{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 15680,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006377551020408163,
      "grad_norm": 0.5606629150294529,
      "learning_rate": 9.999998394288208e-06,
      "loss": 0.32555317878723145,
      "num_input_tokens_seen": 307056,
      "step": 5,
      "train_runtime": 70.644,
      "train_tokens_per_second": 4346.524
    },
    {
      "epoch": 0.0012755102040816326,
      "grad_norm": 0.7593070557109454,
      "learning_rate": 9.999991871085821e-06,
      "loss": 0.3172970533370972,
      "num_input_tokens_seen": 619368,
      "step": 10,
      "train_runtime": 129.5013,
      "train_tokens_per_second": 4782.716
    },
    {
      "epoch": 0.001913265306122449,
      "grad_norm": 0.6388307972758556,
      "learning_rate": 9.999980330042391e-06,
      "loss": 0.3005043983459473,
      "num_input_tokens_seen": 919640,
      "step": 15,
      "train_runtime": 197.7304,
      "train_tokens_per_second": 4650.979
    },
    {
      "epoch": 0.002551020408163265,
      "grad_norm": 0.669146696435932,
      "learning_rate": 9.999963771169503e-06,
      "loss": 0.3088076591491699,
      "num_input_tokens_seen": 1230568,
      "step": 20,
      "train_runtime": 255.401,
      "train_tokens_per_second": 4818.181
    },
    {
      "epoch": 0.0031887755102040817,
      "grad_norm": 0.6777401540799771,
      "learning_rate": 9.999942194483775e-06,
      "loss": 0.3104086399078369,
      "num_input_tokens_seen": 1532272,
      "step": 25,
      "train_runtime": 310.5411,
      "train_tokens_per_second": 4934.201
    },
    {
      "epoch": 0.003826530612244898,
      "grad_norm": 0.6676639429470436,
      "learning_rate": 9.999915600006857e-06,
      "loss": 0.29593493938446047,
      "num_input_tokens_seen": 1816368,
      "step": 30,
      "train_runtime": 371.4824,
      "train_tokens_per_second": 4889.513
    },
    {
      "epoch": 0.004464285714285714,
      "grad_norm": 0.6524667967178304,
      "learning_rate": 9.999883987765443e-06,
      "loss": 0.31024646759033203,
      "num_input_tokens_seen": 2113528,
      "step": 35,
      "train_runtime": 429.8203,
      "train_tokens_per_second": 4917.236
    },
    {
      "epoch": 0.00510204081632653,
      "grad_norm": 0.6353759506595735,
      "learning_rate": 9.999847357791256e-06,
      "loss": 0.3009695529937744,
      "num_input_tokens_seen": 2422408,
      "step": 40,
      "train_runtime": 505.854,
      "train_tokens_per_second": 4788.749
    },
    {
      "epoch": 0.005739795918367347,
      "grad_norm": 0.6662553077885118,
      "learning_rate": 9.999805710121055e-06,
      "loss": 0.30118274688720703,
      "num_input_tokens_seen": 2737912,
      "step": 45,
      "train_runtime": 576.8314,
      "train_tokens_per_second": 4746.468
    },
    {
      "epoch": 0.006377551020408163,
      "grad_norm": 0.6649355594885129,
      "learning_rate": 9.99975904479664e-06,
      "loss": 0.335707688331604,
      "num_input_tokens_seen": 3032176,
      "step": 50,
      "train_runtime": 644.2997,
      "train_tokens_per_second": 4706.158
    },
    {
      "epoch": 0.00701530612244898,
      "grad_norm": 0.6128090342381591,
      "learning_rate": 9.999707361864841e-06,
      "loss": 0.2981316566467285,
      "num_input_tokens_seen": 3326688,
      "step": 55,
      "train_runtime": 706.3582,
      "train_tokens_per_second": 4709.633
    },
    {
      "epoch": 0.007653061224489796,
      "grad_norm": 0.6146153348665865,
      "learning_rate": 9.999650661377527e-06,
      "loss": 0.3245338201522827,
      "num_input_tokens_seen": 3643440,
      "step": 60,
      "train_runtime": 774.5001,
      "train_tokens_per_second": 4704.247
    },
    {
      "epoch": 0.008290816326530613,
      "grad_norm": 0.6785023098079027,
      "learning_rate": 9.999588943391597e-06,
      "loss": 0.3105494022369385,
      "num_input_tokens_seen": 3931984,
      "step": 65,
      "train_runtime": 848.9996,
      "train_tokens_per_second": 4631.314
    },
    {
      "epoch": 0.008928571428571428,
      "grad_norm": 0.6655731547295166,
      "learning_rate": 9.999522207968992e-06,
      "loss": 0.31799840927124023,
      "num_input_tokens_seen": 4232600,
      "step": 70,
      "train_runtime": 911.9567,
      "train_tokens_per_second": 4641.229
    },
    {
      "epoch": 0.009566326530612245,
      "grad_norm": 0.8247868120346744,
      "learning_rate": 9.999450455176686e-06,
      "loss": 0.31391355991363523,
      "num_input_tokens_seen": 4541848,
      "step": 75,
      "train_runtime": 955.2831,
      "train_tokens_per_second": 4754.452
    },
    {
      "epoch": 0.01020408163265306,
      "grad_norm": 0.6199820358842323,
      "learning_rate": 9.99937368508669e-06,
      "loss": 0.3074874639511108,
      "num_input_tokens_seen": 4841560,
      "step": 80,
      "train_runtime": 1018.2827,
      "train_tokens_per_second": 4754.632
    },
    {
      "epoch": 0.010841836734693877,
      "grad_norm": 0.7319306172868103,
      "learning_rate": 9.999291897776043e-06,
      "loss": 0.3182725429534912,
      "num_input_tokens_seen": 5127432,
      "step": 85,
      "train_runtime": 1078.4405,
      "train_tokens_per_second": 4754.487
    },
    {
      "epoch": 0.011479591836734694,
      "grad_norm": 0.6685715685218444,
      "learning_rate": 9.999205093326825e-06,
      "loss": 0.3352250337600708,
      "num_input_tokens_seen": 5436216,
      "step": 90,
      "train_runtime": 1145.626,
      "train_tokens_per_second": 4745.193
    },
    {
      "epoch": 0.01211734693877551,
      "grad_norm": 0.6622114853124953,
      "learning_rate": 9.999113271826154e-06,
      "loss": 0.3200762987136841,
      "num_input_tokens_seen": 5736632,
      "step": 95,
      "train_runtime": 1215.2064,
      "train_tokens_per_second": 4720.706
    },
    {
      "epoch": 0.012755102040816327,
      "grad_norm": 0.6864874287783332,
      "learning_rate": 9.999016433366178e-06,
      "loss": 0.3209428310394287,
      "num_input_tokens_seen": 6036576,
      "step": 100,
      "train_runtime": 1285.2068,
      "train_tokens_per_second": 4696.969
    },
    {
      "epoch": 0.013392857142857142,
      "grad_norm": 0.6354038469659828,
      "learning_rate": 9.998914578044079e-06,
      "loss": 0.3147577285766602,
      "num_input_tokens_seen": 6324568,
      "step": 105,
      "train_runtime": 1349.6796,
      "train_tokens_per_second": 4685.977
    },
    {
      "epoch": 0.01403061224489796,
      "grad_norm": 0.6791525229535978,
      "learning_rate": 9.998807705962079e-06,
      "loss": 0.3091780662536621,
      "num_input_tokens_seen": 6618624,
      "step": 110,
      "train_runtime": 1410.8023,
      "train_tokens_per_second": 4691.39
    },
    {
      "epoch": 0.014668367346938776,
      "grad_norm": 0.5904897872744799,
      "learning_rate": 9.99869581722743e-06,
      "loss": 0.3179654598236084,
      "num_input_tokens_seen": 6928552,
      "step": 115,
      "train_runtime": 1471.3438,
      "train_tokens_per_second": 4708.996
    },
    {
      "epoch": 0.015306122448979591,
      "grad_norm": 0.5866546516408551,
      "learning_rate": 9.99857891195242e-06,
      "loss": 0.30416264533996584,
      "num_input_tokens_seen": 7235768,
      "step": 120,
      "train_runtime": 1541.301,
      "train_tokens_per_second": 4694.585
    },
    {
      "epoch": 0.01594387755102041,
      "grad_norm": 0.5847461573271263,
      "learning_rate": 9.998456990254371e-06,
      "loss": 0.3292226314544678,
      "num_input_tokens_seen": 7548984,
      "step": 125,
      "train_runtime": 1585.4463,
      "train_tokens_per_second": 4761.425
    },
    {
      "epoch": 0.016581632653061226,
      "grad_norm": 0.663765539854714,
      "learning_rate": 9.998330052255642e-06,
      "loss": 0.30205209255218507,
      "num_input_tokens_seen": 7854000,
      "step": 130,
      "train_runtime": 1656.5651,
      "train_tokens_per_second": 4741.136
    },
    {
      "epoch": 0.01721938775510204,
      "grad_norm": 0.5938051231759824,
      "learning_rate": 9.998198098083623e-06,
      "loss": 0.31253373622894287,
      "num_input_tokens_seen": 8150232,
      "step": 135,
      "train_runtime": 1719.6335,
      "train_tokens_per_second": 4739.517
    },
    {
      "epoch": 0.017857142857142856,
      "grad_norm": 0.6766189571803923,
      "learning_rate": 9.998061127870738e-06,
      "loss": 0.3201970100402832,
      "num_input_tokens_seen": 8453184,
      "step": 140,
      "train_runtime": 1773.6419,
      "train_tokens_per_second": 4766.004
    },
    {
      "epoch": 0.018494897959183673,
      "grad_norm": 0.6541348589014065,
      "learning_rate": 9.997919141754448e-06,
      "loss": 0.31698040962219237,
      "num_input_tokens_seen": 8742848,
      "step": 145,
      "train_runtime": 1837.6218,
      "train_tokens_per_second": 4757.697
    },
    {
      "epoch": 0.01913265306122449,
      "grad_norm": 0.6791837085985184,
      "learning_rate": 9.997772139877245e-06,
      "loss": 0.3183109998703003,
      "num_input_tokens_seen": 9051224,
      "step": 150,
      "train_runtime": 1906.2014,
      "train_tokens_per_second": 4748.304
    },
    {
      "epoch": 0.019770408163265307,
      "grad_norm": 0.8304155968229497,
      "learning_rate": 9.997620122386658e-06,
      "loss": 0.3402367115020752,
      "num_input_tokens_seen": 9348064,
      "step": 155,
      "train_runtime": 1974.4495,
      "train_tokens_per_second": 4734.517
    },
    {
      "epoch": 0.02040816326530612,
      "grad_norm": 0.6632211550468946,
      "learning_rate": 9.997463089435242e-06,
      "loss": 0.3259411811828613,
      "num_input_tokens_seen": 9639544,
      "step": 160,
      "train_runtime": 2043.9311,
      "train_tokens_per_second": 4716.179
    },
    {
      "epoch": 0.021045918367346938,
      "grad_norm": 0.6576049807668407,
      "learning_rate": 9.997301041180597e-06,
      "loss": 0.32861177921295165,
      "num_input_tokens_seen": 9945192,
      "step": 165,
      "train_runtime": 2103.3501,
      "train_tokens_per_second": 4728.263
    },
    {
      "epoch": 0.021683673469387755,
      "grad_norm": 0.6304566369226388,
      "learning_rate": 9.997133977785342e-06,
      "loss": 0.31604456901550293,
      "num_input_tokens_seen": 10255232,
      "step": 170,
      "train_runtime": 2165.9496,
      "train_tokens_per_second": 4734.751
    },
    {
      "epoch": 0.022321428571428572,
      "grad_norm": 0.6805488864873742,
      "learning_rate": 9.996961899417145e-06,
      "loss": 0.31748697757720945,
      "num_input_tokens_seen": 10566904,
      "step": 175,
      "train_runtime": 2229.7478,
      "train_tokens_per_second": 4739.058
    },
    {
      "epoch": 0.02295918367346939,
      "grad_norm": 0.6082949994569637,
      "learning_rate": 9.996784806248694e-06,
      "loss": 0.29314508438110354,
      "num_input_tokens_seen": 10877624,
      "step": 180,
      "train_runtime": 2300.7334,
      "train_tokens_per_second": 4727.894
    },
    {
      "epoch": 0.023596938775510203,
      "grad_norm": 0.6497868120512466,
      "learning_rate": 9.996602698457716e-06,
      "loss": 0.3461932182312012,
      "num_input_tokens_seen": 11176016,
      "step": 185,
      "train_runtime": 2358.3838,
      "train_tokens_per_second": 4738.845
    },
    {
      "epoch": 0.02423469387755102,
      "grad_norm": 0.6093118092640974,
      "learning_rate": 9.996415576226967e-06,
      "loss": 0.3246460914611816,
      "num_input_tokens_seen": 11489872,
      "step": 190,
      "train_runtime": 2410.919,
      "train_tokens_per_second": 4765.764
    },
    {
      "epoch": 0.024872448979591837,
      "grad_norm": 0.5901408518219843,
      "learning_rate": 9.996223439744238e-06,
      "loss": 0.34906563758850095,
      "num_input_tokens_seen": 11796712,
      "step": 195,
      "train_runtime": 2471.8639,
      "train_tokens_per_second": 4772.395
    },
    {
      "epoch": 0.025510204081632654,
      "grad_norm": 0.7443418375897536,
      "learning_rate": 9.996026289202352e-06,
      "loss": 0.3157334566116333,
      "num_input_tokens_seen": 12094920,
      "step": 200,
      "train_runtime": 2534.5638,
      "train_tokens_per_second": 4771.993
    },
    {
      "epoch": 0.02614795918367347,
      "grad_norm": 0.6328150179349048,
      "learning_rate": 9.995824124799164e-06,
      "loss": 0.3202739477157593,
      "num_input_tokens_seen": 12402328,
      "step": 205,
      "train_runtime": 2599.7223,
      "train_tokens_per_second": 4770.636
    },
    {
      "epoch": 0.026785714285714284,
      "grad_norm": 0.5679139496449538,
      "learning_rate": 9.995616946737558e-06,
      "loss": 0.30471899509429934,
      "num_input_tokens_seen": 12693232,
      "step": 210,
      "train_runtime": 2660.0942,
      "train_tokens_per_second": 4771.723
    },
    {
      "epoch": 0.0274234693877551,
      "grad_norm": 0.6479245859651804,
      "learning_rate": 9.995404755225454e-06,
      "loss": 0.3205737590789795,
      "num_input_tokens_seen": 12988256,
      "step": 215,
      "train_runtime": 2727.0257,
      "train_tokens_per_second": 4762.792
    },
    {
      "epoch": 0.02806122448979592,
      "grad_norm": 0.6394405912298028,
      "learning_rate": 9.995187550475799e-06,
      "loss": 0.2996486186981201,
      "num_input_tokens_seen": 13274424,
      "step": 220,
      "train_runtime": 2790.309,
      "train_tokens_per_second": 4757.331
    },
    {
      "epoch": 0.028698979591836735,
      "grad_norm": 0.607272545676857,
      "learning_rate": 9.994965332706574e-06,
      "loss": 0.3249512195587158,
      "num_input_tokens_seen": 13570744,
      "step": 225,
      "train_runtime": 2857.2788,
      "train_tokens_per_second": 4749.534
    },
    {
      "epoch": 0.029336734693877552,
      "grad_norm": 0.7294675452895084,
      "learning_rate": 9.994738102140791e-06,
      "loss": 0.33909082412719727,
      "num_input_tokens_seen": 13872696,
      "step": 230,
      "train_runtime": 2919.6481,
      "train_tokens_per_second": 4751.496
    },
    {
      "epoch": 0.029974489795918366,
      "grad_norm": 0.6378127712293848,
      "learning_rate": 9.994505859006489e-06,
      "loss": 0.34965171813964846,
      "num_input_tokens_seen": 14164936,
      "step": 235,
      "train_runtime": 2991.8019,
      "train_tokens_per_second": 4734.583
    },
    {
      "epoch": 0.030612244897959183,
      "grad_norm": 0.591562332095497,
      "learning_rate": 9.994268603536743e-06,
      "loss": 0.330527138710022,
      "num_input_tokens_seen": 14468000,
      "step": 240,
      "train_runtime": 3052.1748,
      "train_tokens_per_second": 4740.226
    },
    {
      "epoch": 0.03125,
      "grad_norm": 0.6503140425037022,
      "learning_rate": 9.994026335969654e-06,
      "loss": 0.34091250896453856,
      "num_input_tokens_seen": 14769440,
      "step": 245,
      "train_runtime": 3121.2753,
      "train_tokens_per_second": 4731.861
    },
    {
      "epoch": 0.03188775510204082,
      "grad_norm": 0.7206538638886743,
      "learning_rate": 9.993779056548357e-06,
      "loss": 0.32621986865997316,
      "num_input_tokens_seen": 15070728,
      "step": 250,
      "train_runtime": 3179.8481,
      "train_tokens_per_second": 4739.449
    },
    {
      "epoch": 0.032525510204081634,
      "grad_norm": 0.6417790379171046,
      "learning_rate": 9.99352676552101e-06,
      "loss": 0.32343032360076907,
      "num_input_tokens_seen": 15381016,
      "step": 255,
      "train_runtime": 3244.6399,
      "train_tokens_per_second": 4740.439
    },
    {
      "epoch": 0.03316326530612245,
      "grad_norm": 0.7257546948478659,
      "learning_rate": 9.993269463140805e-06,
      "loss": 0.33203413486480715,
      "num_input_tokens_seen": 15678832,
      "step": 260,
      "train_runtime": 3313.1898,
      "train_tokens_per_second": 4732.247
    },
    {
      "epoch": 0.03380102040816327,
      "grad_norm": 0.5972157409099551,
      "learning_rate": 9.993007149665968e-06,
      "loss": 0.30761942863464353,
      "num_input_tokens_seen": 16002672,
      "step": 265,
      "train_runtime": 3368.7361,
      "train_tokens_per_second": 4750.349
    },
    {
      "epoch": 0.03443877551020408,
      "grad_norm": 0.6274491114706233,
      "learning_rate": 9.992739825359743e-06,
      "loss": 0.3240455150604248,
      "num_input_tokens_seen": 16299328,
      "step": 270,
      "train_runtime": 3429.672,
      "train_tokens_per_second": 4752.445
    },
    {
      "epoch": 0.035076530612244895,
      "grad_norm": 0.633623366571598,
      "learning_rate": 9.992467490490411e-06,
      "loss": 0.3531185150146484,
      "num_input_tokens_seen": 16600456,
      "step": 275,
      "train_runtime": 3497.6353,
      "train_tokens_per_second": 4746.194
    },
    {
      "epoch": 0.03571428571428571,
      "grad_norm": 0.6160667032053221,
      "learning_rate": 9.99219014533128e-06,
      "loss": 0.3064985752105713,
      "num_input_tokens_seen": 16908000,
      "step": 280,
      "train_runtime": 3560.7808,
      "train_tokens_per_second": 4748.397
    },
    {
      "epoch": 0.03635204081632653,
      "grad_norm": 0.6738097526273257,
      "learning_rate": 9.991907790160686e-06,
      "loss": 0.32353625297546384,
      "num_input_tokens_seen": 17206424,
      "step": 285,
      "train_runtime": 3631.8727,
      "train_tokens_per_second": 4737.618
    },
    {
      "epoch": 0.036989795918367346,
      "grad_norm": 0.6817527588786203,
      "learning_rate": 9.991620425261987e-06,
      "loss": 0.3298201084136963,
      "num_input_tokens_seen": 17512464,
      "step": 290,
      "train_runtime": 3693.5072,
      "train_tokens_per_second": 4741.419
    },
    {
      "epoch": 0.03762755102040816,
      "grad_norm": 0.6571499745360195,
      "learning_rate": 9.99132805092358e-06,
      "loss": 0.3204179048538208,
      "num_input_tokens_seen": 17819256,
      "step": 295,
      "train_runtime": 3751.968,
      "train_tokens_per_second": 4749.309
    },
    {
      "epoch": 0.03826530612244898,
      "grad_norm": 0.6382139716578328,
      "learning_rate": 9.99103066743888e-06,
      "loss": 0.3137251853942871,
      "num_input_tokens_seen": 18118704,
      "step": 300,
      "train_runtime": 3816.7296,
      "train_tokens_per_second": 4747.18
    },
    {
      "epoch": 0.0389030612244898,
      "grad_norm": 0.6468563537536628,
      "learning_rate": 9.990728275106332e-06,
      "loss": 0.33147296905517576,
      "num_input_tokens_seen": 18416920,
      "step": 305,
      "train_runtime": 3888.0192,
      "train_tokens_per_second": 4736.839
    },
    {
      "epoch": 0.039540816326530615,
      "grad_norm": 0.565317175678439,
      "learning_rate": 9.990420874229408e-06,
      "loss": 0.3082550525665283,
      "num_input_tokens_seen": 18712736,
      "step": 310,
      "train_runtime": 3950.42,
      "train_tokens_per_second": 4736.898
    },
    {
      "epoch": 0.04017857142857143,
      "grad_norm": 0.711772451702932,
      "learning_rate": 9.990108465116606e-06,
      "loss": 0.33627796173095703,
      "num_input_tokens_seen": 18999736,
      "step": 315,
      "train_runtime": 4022.8437,
      "train_tokens_per_second": 4722.962
    },
    {
      "epoch": 0.04081632653061224,
      "grad_norm": 0.6024687267299634,
      "learning_rate": 9.989791048081452e-06,
      "loss": 0.34989590644836427,
      "num_input_tokens_seen": 19301608,
      "step": 320,
      "train_runtime": 4091.466,
      "train_tokens_per_second": 4717.529
    },
    {
      "epoch": 0.04145408163265306,
      "grad_norm": 0.6438135368684226,
      "learning_rate": 9.989468623442494e-06,
      "loss": 0.3031839609146118,
      "num_input_tokens_seen": 19594792,
      "step": 325,
      "train_runtime": 4151.5342,
      "train_tokens_per_second": 4719.892
    },
    {
      "epoch": 0.042091836734693876,
      "grad_norm": 0.6809717294079394,
      "learning_rate": 9.989141191523309e-06,
      "loss": 0.3175718545913696,
      "num_input_tokens_seen": 19865664,
      "step": 330,
      "train_runtime": 4224.6826,
      "train_tokens_per_second": 4702.286
    },
    {
      "epoch": 0.04272959183673469,
      "grad_norm": 0.6787774677132214,
      "learning_rate": 9.988808752652498e-06,
      "loss": 0.3163781642913818,
      "num_input_tokens_seen": 20161264,
      "step": 335,
      "train_runtime": 4283.0728,
      "train_tokens_per_second": 4707.196
    },
    {
      "epoch": 0.04336734693877551,
      "grad_norm": 0.606984869535405,
      "learning_rate": 9.988471307163684e-06,
      "loss": 0.3327335834503174,
      "num_input_tokens_seen": 20460944,
      "step": 340,
      "train_runtime": 4347.2456,
      "train_tokens_per_second": 4706.645
    },
    {
      "epoch": 0.04400510204081633,
      "grad_norm": 0.7121877671474408,
      "learning_rate": 9.988128855395522e-06,
      "loss": 0.31911287307739256,
      "num_input_tokens_seen": 20759440,
      "step": 345,
      "train_runtime": 4417.4681,
      "train_tokens_per_second": 4699.398
    },
    {
      "epoch": 0.044642857142857144,
      "grad_norm": 0.7092694846190201,
      "learning_rate": 9.987781397691683e-06,
      "loss": 0.3497852087020874,
      "num_input_tokens_seen": 21051728,
      "step": 350,
      "train_runtime": 4489.3049,
      "train_tokens_per_second": 4689.307
    },
    {
      "epoch": 0.04528061224489796,
      "grad_norm": 0.6829232197373883,
      "learning_rate": 9.987428934400864e-06,
      "loss": 0.33017926216125487,
      "num_input_tokens_seen": 21350144,
      "step": 355,
      "train_runtime": 4561.7355,
      "train_tokens_per_second": 4680.268
    },
    {
      "epoch": 0.04591836734693878,
      "grad_norm": 0.6792644968287634,
      "learning_rate": 9.98707146587679e-06,
      "loss": 0.31848602294921874,
      "num_input_tokens_seen": 21642016,
      "step": 360,
      "train_runtime": 4620.6082,
      "train_tokens_per_second": 4683.802
    },
    {
      "epoch": 0.046556122448979595,
      "grad_norm": 0.5901350660402305,
      "learning_rate": 9.986708992478202e-06,
      "loss": 0.30682835578918455,
      "num_input_tokens_seen": 21955552,
      "step": 365,
      "train_runtime": 4685.473,
      "train_tokens_per_second": 4685.877
    },
    {
      "epoch": 0.047193877551020405,
      "grad_norm": 0.5923571477531686,
      "learning_rate": 9.98634151456887e-06,
      "loss": 0.29576663970947265,
      "num_input_tokens_seen": 22264416,
      "step": 370,
      "train_runtime": 4735.8841,
      "train_tokens_per_second": 4701.216
    },
    {
      "epoch": 0.04783163265306122,
      "grad_norm": 0.6394976500523845,
      "learning_rate": 9.985969032517582e-06,
      "loss": 0.3408482551574707,
      "num_input_tokens_seen": 22582104,
      "step": 375,
      "train_runtime": 4796.2194,
      "train_tokens_per_second": 4708.313
    },
    {
      "epoch": 0.04846938775510204,
      "grad_norm": 0.6948625367608592,
      "learning_rate": 9.985591546698151e-06,
      "loss": 0.32459070682525637,
      "num_input_tokens_seen": 22874592,
      "step": 380,
      "train_runtime": 4863.8457,
      "train_tokens_per_second": 4702.985
    },
    {
      "epoch": 0.049107142857142856,
      "grad_norm": 0.6621128282688693,
      "learning_rate": 9.98520905748941e-06,
      "loss": 0.3638442039489746,
      "num_input_tokens_seen": 23168960,
      "step": 385,
      "train_runtime": 4924.6561,
      "train_tokens_per_second": 4704.686
    },
    {
      "epoch": 0.04974489795918367,
      "grad_norm": 0.5449516416009511,
      "learning_rate": 9.984821565275214e-06,
      "loss": 0.31552510261535643,
      "num_input_tokens_seen": 23476952,
      "step": 390,
      "train_runtime": 4985.8864,
      "train_tokens_per_second": 4708.682
    },
    {
      "epoch": 0.05038265306122449,
      "grad_norm": 0.6218114965530601,
      "learning_rate": 9.984429070444437e-06,
      "loss": 0.3246952533721924,
      "num_input_tokens_seen": 23783704,
      "step": 395,
      "train_runtime": 5042.3193,
      "train_tokens_per_second": 4716.818
    },
    {
      "epoch": 0.05102040816326531,
      "grad_norm": 0.6938120251411585,
      "learning_rate": 9.984031573390977e-06,
      "loss": 0.31642162799835205,
      "num_input_tokens_seen": 24084240,
      "step": 400,
      "train_runtime": 5106.6348,
      "train_tokens_per_second": 4716.264
    },
    {
      "epoch": 0.051658163265306124,
      "grad_norm": 0.6408719146211228,
      "learning_rate": 9.983629074513749e-06,
      "loss": 0.3200521469116211,
      "num_input_tokens_seen": 24392072,
      "step": 405,
      "train_runtime": 5162.8643,
      "train_tokens_per_second": 4724.523
    },
    {
      "epoch": 0.05229591836734694,
      "grad_norm": 0.5789189971057029,
      "learning_rate": 9.983221574216688e-06,
      "loss": 0.3312517166137695,
      "num_input_tokens_seen": 24695208,
      "step": 410,
      "train_runtime": 5230.068,
      "train_tokens_per_second": 4721.776
    },
    {
      "epoch": 0.05293367346938776,
      "grad_norm": 0.6249544848251316,
      "learning_rate": 9.98280907290875e-06,
      "loss": 0.3149777889251709,
      "num_input_tokens_seen": 25001944,
      "step": 415,
      "train_runtime": 5288.4449,
      "train_tokens_per_second": 4727.655
    },
    {
      "epoch": 0.05357142857142857,
      "grad_norm": 0.5877787304045785,
      "learning_rate": 9.98239157100391e-06,
      "loss": 0.313005256652832,
      "num_input_tokens_seen": 25314664,
      "step": 420,
      "train_runtime": 5355.1453,
      "train_tokens_per_second": 4727.167
    },
    {
      "epoch": 0.054209183673469385,
      "grad_norm": 0.6097079957803321,
      "learning_rate": 9.981969068921157e-06,
      "loss": 0.3354280710220337,
      "num_input_tokens_seen": 25624904,
      "step": 425,
      "train_runtime": 5429.7986,
      "train_tokens_per_second": 4719.31
    },
    {
      "epoch": 0.0548469387755102,
      "grad_norm": 0.6231312381575178,
      "learning_rate": 9.981541567084506e-06,
      "loss": 0.30369086265563966,
      "num_input_tokens_seen": 25912224,
      "step": 430,
      "train_runtime": 5494.9138,
      "train_tokens_per_second": 4715.674
    },
    {
      "epoch": 0.05548469387755102,
      "grad_norm": 0.6124856979200631,
      "learning_rate": 9.98110906592298e-06,
      "loss": 0.3130204439163208,
      "num_input_tokens_seen": 26221560,
      "step": 435,
      "train_runtime": 5561.1032,
      "train_tokens_per_second": 4715.172
    },
    {
      "epoch": 0.05612244897959184,
      "grad_norm": 0.7234427591142097,
      "learning_rate": 9.980671565870628e-06,
      "loss": 0.3251868486404419,
      "num_input_tokens_seen": 26509480,
      "step": 440,
      "train_runtime": 5618.0134,
      "train_tokens_per_second": 4718.657
    },
    {
      "epoch": 0.056760204081632654,
      "grad_norm": 0.7264223860858371,
      "learning_rate": 9.98022906736651e-06,
      "loss": 0.338460111618042,
      "num_input_tokens_seen": 26797608,
      "step": 445,
      "train_runtime": 5686.6321,
      "train_tokens_per_second": 4712.386
    },
    {
      "epoch": 0.05739795918367347,
      "grad_norm": 0.587867187729823,
      "learning_rate": 9.979781570854704e-06,
      "loss": 0.3184667110443115,
      "num_input_tokens_seen": 27103592,
      "step": 450,
      "train_runtime": 5750.9363,
      "train_tokens_per_second": 4712.901
    },
    {
      "epoch": 0.05803571428571429,
      "grad_norm": 0.6426574915923813,
      "learning_rate": 9.979329076784305e-06,
      "loss": 0.300250506401062,
      "num_input_tokens_seen": 27398720,
      "step": 455,
      "train_runtime": 5819.9648,
      "train_tokens_per_second": 4707.712
    },
    {
      "epoch": 0.058673469387755105,
      "grad_norm": 0.6225789987596188,
      "learning_rate": 9.978871585609422e-06,
      "loss": 0.32442636489868165,
      "num_input_tokens_seen": 27717824,
      "step": 460,
      "train_runtime": 5873.7344,
      "train_tokens_per_second": 4718.944
    },
    {
      "epoch": 0.05931122448979592,
      "grad_norm": 0.5973787264008296,
      "learning_rate": 9.978409097789178e-06,
      "loss": 0.3366720199584961,
      "num_input_tokens_seen": 28021088,
      "step": 465,
      "train_runtime": 5943.536,
      "train_tokens_per_second": 4714.548
    },
    {
      "epoch": 0.05994897959183673,
      "grad_norm": 0.6069025514534458,
      "learning_rate": 9.977941613787714e-06,
      "loss": 0.2983412504196167,
      "num_input_tokens_seen": 28312744,
      "step": 470,
      "train_runtime": 5999.5463,
      "train_tokens_per_second": 4719.148
    },
    {
      "epoch": 0.06058673469387755,
      "grad_norm": 0.6414932786095137,
      "learning_rate": 9.977469134074181e-06,
      "loss": 0.3117494583129883,
      "num_input_tokens_seen": 28608608,
      "step": 475,
      "train_runtime": 6073.1591,
      "train_tokens_per_second": 4710.663
    },
    {
      "epoch": 0.061224489795918366,
      "grad_norm": 0.5766432270379303,
      "learning_rate": 9.976991659122748e-06,
      "loss": 0.332213020324707,
      "num_input_tokens_seen": 28919968,
      "step": 480,
      "train_runtime": 6142.8908,
      "train_tokens_per_second": 4707.876
    },
    {
      "epoch": 0.06186224489795918,
      "grad_norm": 0.687213434108391,
      "learning_rate": 9.976509189412591e-06,
      "loss": 0.32208664417266847,
      "num_input_tokens_seen": 29221048,
      "step": 485,
      "train_runtime": 6200.1589,
      "train_tokens_per_second": 4712.951
    },
    {
      "epoch": 0.0625,
      "grad_norm": 0.6164832754267603,
      "learning_rate": 9.976021725427905e-06,
      "loss": 0.30185608863830565,
      "num_input_tokens_seen": 29514096,
      "step": 490,
      "train_runtime": 6268.6853,
      "train_tokens_per_second": 4708.18
    },
    {
      "epoch": 0.06313775510204081,
      "grad_norm": 0.5608106101306191,
      "learning_rate": 9.975529267657892e-06,
      "loss": 0.3353261470794678,
      "num_input_tokens_seen": 29819160,
      "step": 495,
      "train_runtime": 6330.802,
      "train_tokens_per_second": 4710.171
    },
    {
      "epoch": 0.06377551020408163,
      "grad_norm": 0.6414173387060472,
      "learning_rate": 9.97503181659677e-06,
      "loss": 0.3376767635345459,
      "num_input_tokens_seen": 30123016,
      "step": 500,
      "train_runtime": 6395.8326,
      "train_tokens_per_second": 4709.788
    },
    {
      "epoch": 0.06441326530612244,
      "grad_norm": 1.0983782636367827,
      "learning_rate": 9.974529372743762e-06,
      "loss": 0.35728614330291747,
      "num_input_tokens_seen": 30428672,
      "step": 505,
      "train_runtime": 6458.7711,
      "train_tokens_per_second": 4711.217
    },
    {
      "epoch": 0.06505102040816327,
      "grad_norm": 0.6220557654114379,
      "learning_rate": 9.97402193660311e-06,
      "loss": 0.34481186866760255,
      "num_input_tokens_seen": 30726152,
      "step": 510,
      "train_runtime": 6524.046,
      "train_tokens_per_second": 4709.677
    },
    {
      "epoch": 0.06568877551020408,
      "grad_norm": 0.5294936170454477,
      "learning_rate": 9.973509508684061e-06,
      "loss": 0.30537350177764894,
      "num_input_tokens_seen": 31040816,
      "step": 515,
      "train_runtime": 6591.8962,
      "train_tokens_per_second": 4708.936
    },
    {
      "epoch": 0.0663265306122449,
      "grad_norm": 0.6660172134985414,
      "learning_rate": 9.972992089500868e-06,
      "loss": 0.3000118494033813,
      "num_input_tokens_seen": 31340616,
      "step": 520,
      "train_runtime": 6655.4303,
      "train_tokens_per_second": 4709.029
    },
    {
      "epoch": 0.06696428571428571,
      "grad_norm": 0.5608559388484903,
      "learning_rate": 9.972469679572802e-06,
      "loss": 0.3160065174102783,
      "num_input_tokens_seen": 31632232,
      "step": 525,
      "train_runtime": 6722.5039,
      "train_tokens_per_second": 4705.424
    },
    {
      "epoch": 0.06760204081632654,
      "grad_norm": 0.6358022650010828,
      "learning_rate": 9.971942279424136e-06,
      "loss": 0.330961275100708,
      "num_input_tokens_seen": 31932512,
      "step": 530,
      "train_runtime": 6793.9904,
      "train_tokens_per_second": 4700.111
    },
    {
      "epoch": 0.06823979591836735,
      "grad_norm": 0.6717442313471188,
      "learning_rate": 9.971409889584151e-06,
      "loss": 0.3318564176559448,
      "num_input_tokens_seen": 32235152,
      "step": 535,
      "train_runtime": 6863.2145,
      "train_tokens_per_second": 4696.801
    },
    {
      "epoch": 0.06887755102040816,
      "grad_norm": 0.6592231573120175,
      "learning_rate": 9.970872510587142e-06,
      "loss": 0.3217033386230469,
      "num_input_tokens_seen": 32533016,
      "step": 540,
      "train_runtime": 6929.4769,
      "train_tokens_per_second": 4694.873
    },
    {
      "epoch": 0.06951530612244898,
      "grad_norm": 0.6719574151234373,
      "learning_rate": 9.970330142972403e-06,
      "loss": 0.30914323329925536,
      "num_input_tokens_seen": 32833312,
      "step": 545,
      "train_runtime": 7000.378,
      "train_tokens_per_second": 4690.22
    },
    {
      "epoch": 0.07015306122448979,
      "grad_norm": 0.5935025976726969,
      "learning_rate": 9.969782787284237e-06,
      "loss": 0.32875947952270507,
      "num_input_tokens_seen": 33147480,
      "step": 550,
      "train_runtime": 7065.8158,
      "train_tokens_per_second": 4691.246
    },
    {
      "epoch": 0.07079081632653061,
      "grad_norm": 0.5404040639433234,
      "learning_rate": 9.969230444071956e-06,
      "loss": 0.2959745407104492,
      "num_input_tokens_seen": 33452880,
      "step": 555,
      "train_runtime": 7124.2157,
      "train_tokens_per_second": 4695.658
    },
    {
      "epoch": 0.07142857142857142,
      "grad_norm": 0.6941661209620434,
      "learning_rate": 9.968673113889874e-06,
      "loss": 0.28325576782226564,
      "num_input_tokens_seen": 33732872,
      "step": 560,
      "train_runtime": 7186.4275,
      "train_tokens_per_second": 4693.97
    },
    {
      "epoch": 0.07206632653061225,
      "grad_norm": 0.6130601638560698,
      "learning_rate": 9.968110797297312e-06,
      "loss": 0.33871145248413087,
      "num_input_tokens_seen": 34031504,
      "step": 565,
      "train_runtime": 7245.269,
      "train_tokens_per_second": 4697.066
    },
    {
      "epoch": 0.07270408163265306,
      "grad_norm": 0.7471357200132628,
      "learning_rate": 9.967543494858591e-06,
      "loss": 0.3374213695526123,
      "num_input_tokens_seen": 34322088,
      "step": 570,
      "train_runtime": 7311.3288,
      "train_tokens_per_second": 4694.371
    },
    {
      "epoch": 0.07334183673469388,
      "grad_norm": 0.5507845697057443,
      "learning_rate": 9.966971207143041e-06,
      "loss": 0.29483556747436523,
      "num_input_tokens_seen": 34628464,
      "step": 575,
      "train_runtime": 7366.2422,
      "train_tokens_per_second": 4700.967
    },
    {
      "epoch": 0.07397959183673469,
      "grad_norm": 0.5780226519591101,
      "learning_rate": 9.966393934724992e-06,
      "loss": 0.304833984375,
      "num_input_tokens_seen": 34935360,
      "step": 580,
      "train_runtime": 7434.2548,
      "train_tokens_per_second": 4699.242
    },
    {
      "epoch": 0.07461734693877552,
      "grad_norm": 0.680301531971061,
      "learning_rate": 9.965811678183777e-06,
      "loss": 0.3297819375991821,
      "num_input_tokens_seen": 35232952,
      "step": 585,
      "train_runtime": 7495.366,
      "train_tokens_per_second": 4700.631
    },
    {
      "epoch": 0.07525510204081633,
      "grad_norm": 0.5924148284649571,
      "learning_rate": 9.965224438103733e-06,
      "loss": 0.32800214290618895,
      "num_input_tokens_seen": 35544336,
      "step": 590,
      "train_runtime": 7562.7599,
      "train_tokens_per_second": 4699.916
    },
    {
      "epoch": 0.07589285714285714,
      "grad_norm": 0.6551797249996265,
      "learning_rate": 9.964632215074193e-06,
      "loss": 0.3100182771682739,
      "num_input_tokens_seen": 35846112,
      "step": 595,
      "train_runtime": 7629.9065,
      "train_tokens_per_second": 4698.106
    },
    {
      "epoch": 0.07653061224489796,
      "grad_norm": 0.6393776518734431,
      "learning_rate": 9.964035009689498e-06,
      "loss": 0.31659588813781736,
      "num_input_tokens_seen": 36143816,
      "step": 600,
      "train_runtime": 7695.312,
      "train_tokens_per_second": 4696.862
    },
    {
      "epoch": 0.07716836734693877,
      "grad_norm": 0.60578047473953,
      "learning_rate": 9.963432822548982e-06,
      "loss": 0.2801633358001709,
      "num_input_tokens_seen": 36442288,
      "step": 605,
      "train_runtime": 7764.4173,
      "train_tokens_per_second": 4693.499
    },
    {
      "epoch": 0.0778061224489796,
      "grad_norm": 0.6609439547465508,
      "learning_rate": 9.962825654256984e-06,
      "loss": 0.3205680847167969,
      "num_input_tokens_seen": 36740720,
      "step": 610,
      "train_runtime": 7837.3296,
      "train_tokens_per_second": 4687.913
    },
    {
      "epoch": 0.0784438775510204,
      "grad_norm": 0.5596926565317494,
      "learning_rate": 9.962213505422838e-06,
      "loss": 0.32973132133483884,
      "num_input_tokens_seen": 37050280,
      "step": 615,
      "train_runtime": 7899.5307,
      "train_tokens_per_second": 4690.187
    },
    {
      "epoch": 0.07908163265306123,
      "grad_norm": 0.6288867716165758,
      "learning_rate": 9.96159637666088e-06,
      "loss": 0.3185285568237305,
      "num_input_tokens_seen": 37362816,
      "step": 620,
      "train_runtime": 7972.0212,
      "train_tokens_per_second": 4686.743
    },
    {
      "epoch": 0.07971938775510204,
      "grad_norm": 0.6053102405034232,
      "learning_rate": 9.96097426859044e-06,
      "loss": 0.30266656875610354,
      "num_input_tokens_seen": 37657664,
      "step": 625,
      "train_runtime": 8036.5517,
      "train_tokens_per_second": 4685.799
    },
    {
      "epoch": 0.08035714285714286,
      "grad_norm": 0.5929508426778245,
      "learning_rate": 9.960347181835848e-06,
      "loss": 0.3362568378448486,
      "num_input_tokens_seen": 37953256,
      "step": 630,
      "train_runtime": 8101.3319,
      "train_tokens_per_second": 4684.817
    },
    {
      "epoch": 0.08099489795918367,
      "grad_norm": 0.5990356640392523,
      "learning_rate": 9.95971511702643e-06,
      "loss": 0.3335370302200317,
      "num_input_tokens_seen": 38261928,
      "step": 635,
      "train_runtime": 8165.5101,
      "train_tokens_per_second": 4685.798
    },
    {
      "epoch": 0.08163265306122448,
      "grad_norm": 0.6128622752949864,
      "learning_rate": 9.959078074796505e-06,
      "loss": 0.30237016677856443,
      "num_input_tokens_seen": 38567456,
      "step": 640,
      "train_runtime": 8231.7885,
      "train_tokens_per_second": 4685.185
    },
    {
      "epoch": 0.08227040816326531,
      "grad_norm": 0.6152463940681897,
      "learning_rate": 9.958436055785391e-06,
      "loss": 0.3088388442993164,
      "num_input_tokens_seen": 38864792,
      "step": 645,
      "train_runtime": 8287.9684,
      "train_tokens_per_second": 4689.303
    },
    {
      "epoch": 0.08290816326530612,
      "grad_norm": 0.6553132749686336,
      "learning_rate": 9.9577890606374e-06,
      "loss": 0.32641143798828126,
      "num_input_tokens_seen": 39140920,
      "step": 650,
      "train_runtime": 8353.9801,
      "train_tokens_per_second": 4685.302
    },
    {
      "epoch": 0.08354591836734694,
      "grad_norm": 0.6119341484296121,
      "learning_rate": 9.957137090001834e-06,
      "loss": 0.34890944957733155,
      "num_input_tokens_seen": 39450344,
      "step": 655,
      "train_runtime": 8418.7079,
      "train_tokens_per_second": 4686.033
    },
    {
      "epoch": 0.08418367346938775,
      "grad_norm": 0.6288933704786654,
      "learning_rate": 9.956480144532992e-06,
      "loss": 0.32751471996307374,
      "num_input_tokens_seen": 39741144,
      "step": 660,
      "train_runtime": 8480.8337,
      "train_tokens_per_second": 4685.995
    },
    {
      "epoch": 0.08482142857142858,
      "grad_norm": 0.5550925641636524,
      "learning_rate": 9.955818224890166e-06,
      "loss": 0.301294732093811,
      "num_input_tokens_seen": 40053448,
      "step": 665,
      "train_runtime": 8536.8798,
      "train_tokens_per_second": 4691.813
    },
    {
      "epoch": 0.08545918367346939,
      "grad_norm": 0.5896776348699012,
      "learning_rate": 9.955151331737638e-06,
      "loss": 0.32529559135437014,
      "num_input_tokens_seen": 40364272,
      "step": 670,
      "train_runtime": 8606.7891,
      "train_tokens_per_second": 4689.818
    },
    {
      "epoch": 0.08609693877551021,
      "grad_norm": 0.6244477227734737,
      "learning_rate": 9.954479465744681e-06,
      "loss": 0.30895309448242186,
      "num_input_tokens_seen": 40652352,
      "step": 675,
      "train_runtime": 8680.4236,
      "train_tokens_per_second": 4683.222
    },
    {
      "epoch": 0.08673469387755102,
      "grad_norm": 0.6537923271745987,
      "learning_rate": 9.95380262758556e-06,
      "loss": 0.34395403861999513,
      "num_input_tokens_seen": 40955904,
      "step": 680,
      "train_runtime": 8734.5293,
      "train_tokens_per_second": 4688.965
    },
    {
      "epoch": 0.08737244897959184,
      "grad_norm": 0.5680360919622338,
      "learning_rate": 9.953120817939532e-06,
      "loss": 0.3261738300323486,
      "num_input_tokens_seen": 41274320,
      "step": 685,
      "train_runtime": 8803.1769,
      "train_tokens_per_second": 4688.571
    },
    {
      "epoch": 0.08801020408163265,
      "grad_norm": 0.6507982924694738,
      "learning_rate": 9.952434037490835e-06,
      "loss": 0.34364681243896483,
      "num_input_tokens_seen": 41565784,
      "step": 690,
      "train_runtime": 8873.5797,
      "train_tokens_per_second": 4684.218
    },
    {
      "epoch": 0.08864795918367346,
      "grad_norm": 0.5721567224791083,
      "learning_rate": 9.951742286928706e-06,
      "loss": 0.31003880500793457,
      "num_input_tokens_seen": 41878216,
      "step": 695,
      "train_runtime": 8941.4147,
      "train_tokens_per_second": 4683.623
    },
    {
      "epoch": 0.08928571428571429,
      "grad_norm": 0.62241533513591,
      "learning_rate": 9.951045566947361e-06,
      "loss": 0.31658937931060793,
      "num_input_tokens_seen": 42162784,
      "step": 700,
      "train_runtime": 9004.6117,
      "train_tokens_per_second": 4682.354
    },
    {
      "epoch": 0.0899234693877551,
      "grad_norm": 0.6436845568443312,
      "learning_rate": 9.950343878246011e-06,
      "loss": 0.33315672874450686,
      "num_input_tokens_seen": 42462184,
      "step": 705,
      "train_runtime": 9073.6235,
      "train_tokens_per_second": 4679.738
    },
    {
      "epoch": 0.09056122448979592,
      "grad_norm": 0.5942064773040654,
      "learning_rate": 9.949637221528847e-06,
      "loss": 0.309171199798584,
      "num_input_tokens_seen": 42764072,
      "step": 710,
      "train_runtime": 9144.3897,
      "train_tokens_per_second": 4676.536
    },
    {
      "epoch": 0.09119897959183673,
      "grad_norm": 0.6031870848834505,
      "learning_rate": 9.94892559750505e-06,
      "loss": 0.30094854831695556,
      "num_input_tokens_seen": 43068888,
      "step": 715,
      "train_runtime": 9208.8944,
      "train_tokens_per_second": 4676.879
    },
    {
      "epoch": 0.09183673469387756,
      "grad_norm": 0.5937684412015127,
      "learning_rate": 9.948209006888782e-06,
      "loss": 0.3068702220916748,
      "num_input_tokens_seen": 43363368,
      "step": 720,
      "train_runtime": 9266.0798,
      "train_tokens_per_second": 4679.797
    },
    {
      "epoch": 0.09247448979591837,
      "grad_norm": 0.626594993813998,
      "learning_rate": 9.947487450399196e-06,
      "loss": 0.30313475131988527,
      "num_input_tokens_seen": 43664912,
      "step": 725,
      "train_runtime": 9327.0004,
      "train_tokens_per_second": 4681.56
    },
    {
      "epoch": 0.09311224489795919,
      "grad_norm": 0.6088546257040371,
      "learning_rate": 9.946760928760419e-06,
      "loss": 0.32722713947296145,
      "num_input_tokens_seen": 43950432,
      "step": 730,
      "train_runtime": 9380.4369,
      "train_tokens_per_second": 4685.329
    },
    {
      "epoch": 0.09375,
      "grad_norm": 0.719430601296753,
      "learning_rate": 9.946029442701572e-06,
      "loss": 0.3238682270050049,
      "num_input_tokens_seen": 44237160,
      "step": 735,
      "train_runtime": 9442.7344,
      "train_tokens_per_second": 4684.783
    },
    {
      "epoch": 0.09438775510204081,
      "grad_norm": 0.6703492333647523,
      "learning_rate": 9.945292992956747e-06,
      "loss": 0.34824090003967284,
      "num_input_tokens_seen": 44524808,
      "step": 740,
      "train_runtime": 9516.0973,
      "train_tokens_per_second": 4678.894
    },
    {
      "epoch": 0.09502551020408163,
      "grad_norm": 0.6648986101307878,
      "learning_rate": 9.944551580265025e-06,
      "loss": 0.2878001928329468,
      "num_input_tokens_seen": 44819888,
      "step": 745,
      "train_runtime": 9584.2294,
      "train_tokens_per_second": 4676.421
    },
    {
      "epoch": 0.09566326530612244,
      "grad_norm": 0.6265596220483821,
      "learning_rate": 9.943805205370467e-06,
      "loss": 0.29811065196990966,
      "num_input_tokens_seen": 45111088,
      "step": 750,
      "train_runtime": 9646.616,
      "train_tokens_per_second": 4676.364
    },
    {
      "epoch": 0.09630102040816327,
      "grad_norm": 0.6369850036746705,
      "learning_rate": 9.94305386902211e-06,
      "loss": 0.3029534339904785,
      "num_input_tokens_seen": 45396712,
      "step": 755,
      "train_runtime": 9706.4685,
      "train_tokens_per_second": 4676.955
    },
    {
      "epoch": 0.09693877551020408,
      "grad_norm": 0.7009399437487792,
      "learning_rate": 9.942297571973973e-06,
      "loss": 0.31508803367614746,
      "num_input_tokens_seen": 45704072,
      "step": 760,
      "train_runtime": 9766.4704,
      "train_tokens_per_second": 4679.692
    },
    {
      "epoch": 0.0975765306122449,
      "grad_norm": 0.6732654773954694,
      "learning_rate": 9.941536314985054e-06,
      "loss": 0.3145891666412354,
      "num_input_tokens_seen": 46011488,
      "step": 765,
      "train_runtime": 9844.4757,
      "train_tokens_per_second": 4673.838
    },
    {
      "epoch": 0.09821428571428571,
      "grad_norm": 0.6714319811854066,
      "learning_rate": 9.940770098819329e-06,
      "loss": 0.2932924747467041,
      "num_input_tokens_seen": 46308584,
      "step": 770,
      "train_runtime": 9906.462,
      "train_tokens_per_second": 4674.584
    },
    {
      "epoch": 0.09885204081632654,
      "grad_norm": 0.5948171857193486,
      "learning_rate": 9.939998924245742e-06,
      "loss": 0.3305630683898926,
      "num_input_tokens_seen": 46611024,
      "step": 775,
      "train_runtime": 9969.1515,
      "train_tokens_per_second": 4675.526
    },
    {
      "epoch": 0.09948979591836735,
      "grad_norm": 0.6881406977355818,
      "learning_rate": 9.93922279203823e-06,
      "loss": 0.31389031410217283,
      "num_input_tokens_seen": 46910264,
      "step": 780,
      "train_runtime": 10022.2978,
      "train_tokens_per_second": 4680.59
    },
    {
      "epoch": 0.10012755102040816,
      "grad_norm": 0.7115771502207534,
      "learning_rate": 9.938441702975689e-06,
      "loss": 0.30338244438171386,
      "num_input_tokens_seen": 47206432,
      "step": 785,
      "train_runtime": 10089.0775,
      "train_tokens_per_second": 4678.964
    },
    {
      "epoch": 0.10076530612244898,
      "grad_norm": 0.6640895019101406,
      "learning_rate": 9.937655657842001e-06,
      "loss": 0.31490967273712156,
      "num_input_tokens_seen": 47499424,
      "step": 790,
      "train_runtime": 10146.8944,
      "train_tokens_per_second": 4681.179
    },
    {
      "epoch": 0.10140306122448979,
      "grad_norm": 0.6623807327573633,
      "learning_rate": 9.936864657426014e-06,
      "loss": 0.31421527862548826,
      "num_input_tokens_seen": 47796664,
      "step": 795,
      "train_runtime": 10219.13,
      "train_tokens_per_second": 4677.175
    },
    {
      "epoch": 0.10204081632653061,
      "grad_norm": 0.661257226744512,
      "learning_rate": 9.936068702521553e-06,
      "loss": 0.341552734375,
      "num_input_tokens_seen": 48091408,
      "step": 800,
      "train_runtime": 10277.118,
      "train_tokens_per_second": 4679.464
    },
    {
      "epoch": 0.10267857142857142,
      "grad_norm": 0.7209962136089281,
      "learning_rate": 9.935267793927416e-06,
      "loss": 0.32032017707824706,
      "num_input_tokens_seen": 48390784,
      "step": 805,
      "train_runtime": 10354.1699,
      "train_tokens_per_second": 4673.555
    },
    {
      "epoch": 0.10331632653061225,
      "grad_norm": 0.5488066440785075,
      "learning_rate": 9.93446193244737e-06,
      "loss": 0.3139784336090088,
      "num_input_tokens_seen": 48696176,
      "step": 810,
      "train_runtime": 10406.7942,
      "train_tokens_per_second": 4679.268
    },
    {
      "epoch": 0.10395408163265306,
      "grad_norm": 0.544898948943717,
      "learning_rate": 9.933651118890151e-06,
      "loss": 0.2795098781585693,
      "num_input_tokens_seen": 49003600,
      "step": 815,
      "train_runtime": 10468.8302,
      "train_tokens_per_second": 4680.905
    },
    {
      "epoch": 0.10459183673469388,
      "grad_norm": 0.6187756471207275,
      "learning_rate": 9.93283535406947e-06,
      "loss": 0.32117714881896975,
      "num_input_tokens_seen": 49303792,
      "step": 820,
      "train_runtime": 10539.9136,
      "train_tokens_per_second": 4677.817
    },
    {
      "epoch": 0.10522959183673469,
      "grad_norm": 0.5417719818828942,
      "learning_rate": 9.932014638804001e-06,
      "loss": 0.3011282444000244,
      "num_input_tokens_seen": 49597720,
      "step": 825,
      "train_runtime": 10611.7683,
      "train_tokens_per_second": 4673.841
    },
    {
      "epoch": 0.10586734693877552,
      "grad_norm": 0.5866159331507957,
      "learning_rate": 9.931188973917393e-06,
      "loss": 0.3179909706115723,
      "num_input_tokens_seen": 49904016,
      "step": 830,
      "train_runtime": 10678.652,
      "train_tokens_per_second": 4673.251
    },
    {
      "epoch": 0.10650510204081633,
      "grad_norm": 0.6378722878023269,
      "learning_rate": 9.930358360238255e-06,
      "loss": 0.3088613271713257,
      "num_input_tokens_seen": 50207048,
      "step": 835,
      "train_runtime": 10742.1626,
      "train_tokens_per_second": 4673.831
    },
    {
      "epoch": 0.10714285714285714,
      "grad_norm": 0.5956580159604341,
      "learning_rate": 9.929522798600168e-06,
      "loss": 0.3251255989074707,
      "num_input_tokens_seen": 50507064,
      "step": 840,
      "train_runtime": 10808.9908,
      "train_tokens_per_second": 4672.69
    },
    {
      "epoch": 0.10778061224489796,
      "grad_norm": 0.5592604343521482,
      "learning_rate": 9.928682289841674e-06,
      "loss": 0.29808411598205564,
      "num_input_tokens_seen": 50799800,
      "step": 845,
      "train_runtime": 10867.1029,
      "train_tokens_per_second": 4674.641
    },
    {
      "epoch": 0.10841836734693877,
      "grad_norm": 0.6821118302048432,
      "learning_rate": 9.927836834806284e-06,
      "loss": 0.31413898468017576,
      "num_input_tokens_seen": 51109008,
      "step": 850,
      "train_runtime": 10928.4495,
      "train_tokens_per_second": 4676.693
    },
    {
      "epoch": 0.1090561224489796,
      "grad_norm": 0.6112797837438632,
      "learning_rate": 9.926986434342471e-06,
      "loss": 0.33620872497558596,
      "num_input_tokens_seen": 51419856,
      "step": 855,
      "train_runtime": 10977.6527,
      "train_tokens_per_second": 4684.048
    },
    {
      "epoch": 0.1096938775510204,
      "grad_norm": 0.6965902197834406,
      "learning_rate": 9.926131089303672e-06,
      "loss": 0.32804648876190184,
      "num_input_tokens_seen": 51718856,
      "step": 860,
      "train_runtime": 11045.6007,
      "train_tokens_per_second": 4682.304
    },
    {
      "epoch": 0.11033163265306123,
      "grad_norm": 0.5922989265164307,
      "learning_rate": 9.925270800548285e-06,
      "loss": 0.2920385360717773,
      "num_input_tokens_seen": 52001312,
      "step": 865,
      "train_runtime": 11110.6562,
      "train_tokens_per_second": 4680.31
    },
    {
      "epoch": 0.11096938775510204,
      "grad_norm": 0.5552690111657114,
      "learning_rate": 9.924405568939667e-06,
      "loss": 0.2939499616622925,
      "num_input_tokens_seen": 52289960,
      "step": 870,
      "train_runtime": 11167.8448,
      "train_tokens_per_second": 4682.189
    },
    {
      "epoch": 0.11160714285714286,
      "grad_norm": 0.7357628357379227,
      "learning_rate": 9.923535395346145e-06,
      "loss": 0.31589181423187257,
      "num_input_tokens_seen": 52579872,
      "step": 875,
      "train_runtime": 11232.4079,
      "train_tokens_per_second": 4681.086
    },
    {
      "epoch": 0.11224489795918367,
      "grad_norm": 0.7322803384441785,
      "learning_rate": 9.92266028064099e-06,
      "loss": 0.31965022087097167,
      "num_input_tokens_seen": 52881552,
      "step": 880,
      "train_runtime": 11301.1359,
      "train_tokens_per_second": 4679.313
    },
    {
      "epoch": 0.11288265306122448,
      "grad_norm": 0.68574343125282,
      "learning_rate": 9.921780225702448e-06,
      "loss": 0.32580366134643557,
      "num_input_tokens_seen": 53180168,
      "step": 885,
      "train_runtime": 11360.1319,
      "train_tokens_per_second": 4681.298
    },
    {
      "epoch": 0.11352040816326531,
      "grad_norm": 0.626630658179409,
      "learning_rate": 9.920895231413714e-06,
      "loss": 0.31912999153137206,
      "num_input_tokens_seen": 53479400,
      "step": 890,
      "train_runtime": 11427.9244,
      "train_tokens_per_second": 4679.712
    },
    {
      "epoch": 0.11415816326530612,
      "grad_norm": 0.6308652018383727,
      "learning_rate": 9.92000529866294e-06,
      "loss": 0.2981255054473877,
      "num_input_tokens_seen": 53757944,
      "step": 895,
      "train_runtime": 11488.6682,
      "train_tokens_per_second": 4679.215
    },
    {
      "epoch": 0.11479591836734694,
      "grad_norm": 0.5792513572099894,
      "learning_rate": 9.919110428343235e-06,
      "loss": 0.336773419380188,
      "num_input_tokens_seen": 54073360,
      "step": 900,
      "train_runtime": 11549.696,
      "train_tokens_per_second": 4681.799
    },
    {
      "epoch": 0.11543367346938775,
      "grad_norm": 0.5578804162195902,
      "learning_rate": 9.918210621352667e-06,
      "loss": 0.29676618576049807,
      "num_input_tokens_seen": 54382992,
      "step": 905,
      "train_runtime": 11615.1619,
      "train_tokens_per_second": 4682.069
    },
    {
      "epoch": 0.11607142857142858,
      "grad_norm": 0.5825139503867761,
      "learning_rate": 9.917305878594251e-06,
      "loss": 0.3137337684631348,
      "num_input_tokens_seen": 54688480,
      "step": 910,
      "train_runtime": 11690.9802,
      "train_tokens_per_second": 4677.835
    },
    {
      "epoch": 0.11670918367346939,
      "grad_norm": 0.6405537860326604,
      "learning_rate": 9.916396200975966e-06,
      "loss": 0.3228217840194702,
      "num_input_tokens_seen": 54978288,
      "step": 915,
      "train_runtime": 11758.4874,
      "train_tokens_per_second": 4675.626
    },
    {
      "epoch": 0.11734693877551021,
      "grad_norm": 0.6152578874032614,
      "learning_rate": 9.91548158941073e-06,
      "loss": 0.3395294189453125,
      "num_input_tokens_seen": 55269312,
      "step": 920,
      "train_runtime": 11826.8238,
      "train_tokens_per_second": 4673.217
    },
    {
      "epoch": 0.11798469387755102,
      "grad_norm": 0.5712778569213676,
      "learning_rate": 9.914562044816424e-06,
      "loss": 0.3318194389343262,
      "num_input_tokens_seen": 55567456,
      "step": 925,
      "train_runtime": 11892.7035,
      "train_tokens_per_second": 4672.399
    },
    {
      "epoch": 0.11862244897959184,
      "grad_norm": 0.5973264820162346,
      "learning_rate": 9.913637568115873e-06,
      "loss": 0.3069302082061768,
      "num_input_tokens_seen": 55862328,
      "step": 930,
      "train_runtime": 11955.3605,
      "train_tokens_per_second": 4672.576
    },
    {
      "epoch": 0.11926020408163265,
      "grad_norm": 0.5917702718278268,
      "learning_rate": 9.912708160236854e-06,
      "loss": 0.2973860502243042,
      "num_input_tokens_seen": 56176552,
      "step": 935,
      "train_runtime": 12021.9816,
      "train_tokens_per_second": 4672.82
    },
    {
      "epoch": 0.11989795918367346,
      "grad_norm": 0.5914721778462996,
      "learning_rate": 9.911773822112094e-06,
      "loss": 0.31302266120910643,
      "num_input_tokens_seen": 56476512,
      "step": 940,
      "train_runtime": 12080.7884,
      "train_tokens_per_second": 4674.903
    },
    {
      "epoch": 0.12053571428571429,
      "grad_norm": 0.5599367349904536,
      "learning_rate": 9.910834554679266e-06,
      "loss": 0.298633337020874,
      "num_input_tokens_seen": 56778048,
      "step": 945,
      "train_runtime": 12133.4272,
      "train_tokens_per_second": 4679.473
    },
    {
      "epoch": 0.1211734693877551,
      "grad_norm": 0.6253758896492649,
      "learning_rate": 9.909890358880991e-06,
      "loss": 0.34549541473388673,
      "num_input_tokens_seen": 57088192,
      "step": 950,
      "train_runtime": 12184.659,
      "train_tokens_per_second": 4685.252
    },
    {
      "epoch": 0.12181122448979592,
      "grad_norm": 0.6700044937088339,
      "learning_rate": 9.908941235664835e-06,
      "loss": 0.3271981716156006,
      "num_input_tokens_seen": 57380160,
      "step": 955,
      "train_runtime": 12244.7229,
      "train_tokens_per_second": 4686.113
    },
    {
      "epoch": 0.12244897959183673,
      "grad_norm": 0.6511179079267093,
      "learning_rate": 9.907987185983307e-06,
      "loss": 0.30856757164001464,
      "num_input_tokens_seen": 57671608,
      "step": 960,
      "train_runtime": 12312.6788,
      "train_tokens_per_second": 4683.92
    },
    {
      "epoch": 0.12308673469387756,
      "grad_norm": 0.6347264543032678,
      "learning_rate": 9.907028210793868e-06,
      "loss": 0.3388669490814209,
      "num_input_tokens_seen": 57988552,
      "step": 965,
      "train_runtime": 12375.9362,
      "train_tokens_per_second": 4685.589
    },
    {
      "epoch": 0.12372448979591837,
      "grad_norm": 0.6987486238581951,
      "learning_rate": 9.906064311058911e-06,
      "loss": 0.35554654598236085,
      "num_input_tokens_seen": 58296672,
      "step": 970,
      "train_runtime": 12436.9652,
      "train_tokens_per_second": 4687.371
    },
    {
      "epoch": 0.12436224489795919,
      "grad_norm": 0.6107891338604124,
      "learning_rate": 9.90509548774578e-06,
      "loss": 0.2888030052185059,
      "num_input_tokens_seen": 58608296,
      "step": 975,
      "train_runtime": 12498.8867,
      "train_tokens_per_second": 4689.081
    },
    {
      "epoch": 0.125,
      "grad_norm": 0.6392507862116199,
      "learning_rate": 9.904121741826757e-06,
      "loss": 0.3465890884399414,
      "num_input_tokens_seen": 58903200,
      "step": 980,
      "train_runtime": 12567.6865,
      "train_tokens_per_second": 4686.877
    },
    {
      "epoch": 0.1256377551020408,
      "grad_norm": 0.6406840589093801,
      "learning_rate": 9.903143074279061e-06,
      "loss": 0.3300570249557495,
      "num_input_tokens_seen": 59205872,
      "step": 985,
      "train_runtime": 12636.9273,
      "train_tokens_per_second": 4685.148
    },
    {
      "epoch": 0.12627551020408162,
      "grad_norm": 0.5817406031322423,
      "learning_rate": 9.902159486084857e-06,
      "loss": 0.3436952829360962,
      "num_input_tokens_seen": 59501888,
      "step": 990,
      "train_runtime": 12708.2782,
      "train_tokens_per_second": 4682.136
    },
    {
      "epoch": 0.12691326530612246,
      "grad_norm": 0.5604581561085965,
      "learning_rate": 9.90117097823124e-06,
      "loss": 0.3070975780487061,
      "num_input_tokens_seen": 59793520,
      "step": 995,
      "train_runtime": 12767.709,
      "train_tokens_per_second": 4683.183
    },
    {
      "epoch": 0.12755102040816327,
      "grad_norm": 0.5731625321807637,
      "learning_rate": 9.900177551710252e-06,
      "loss": 0.30677595138549807,
      "num_input_tokens_seen": 60096304,
      "step": 1000,
      "train_runtime": 12833.3771,
      "train_tokens_per_second": 4682.813
    },
    {
      "epoch": 0.12818877551020408,
      "grad_norm": 0.7803286085707585,
      "learning_rate": 9.899179207518863e-06,
      "loss": 0.3457111597061157,
      "num_input_tokens_seen": 60401392,
      "step": 1005,
      "train_runtime": 12896.7323,
      "train_tokens_per_second": 4683.465
    },
    {
      "epoch": 0.1288265306122449,
      "grad_norm": 0.7006445386826811,
      "learning_rate": 9.89817594665898e-06,
      "loss": 0.3316737174987793,
      "num_input_tokens_seen": 60692632,
      "step": 1010,
      "train_runtime": 12970.2757,
      "train_tokens_per_second": 4679.363
    },
    {
      "epoch": 0.12946428571428573,
      "grad_norm": 0.5850857329875174,
      "learning_rate": 9.897167770137449e-06,
      "loss": 0.3376582145690918,
      "num_input_tokens_seen": 60990224,
      "step": 1015,
      "train_runtime": 13022.514,
      "train_tokens_per_second": 4683.445
    },
    {
      "epoch": 0.13010204081632654,
      "grad_norm": 0.5877881119871815,
      "learning_rate": 9.89615467896604e-06,
      "loss": 0.31652557849884033,
      "num_input_tokens_seen": 61282832,
      "step": 1020,
      "train_runtime": 13086.7786,
      "train_tokens_per_second": 4682.805
    },
    {
      "epoch": 0.13073979591836735,
      "grad_norm": 0.6350167715825561,
      "learning_rate": 9.895136674161466e-06,
      "loss": 0.36972153186798096,
      "num_input_tokens_seen": 61579128,
      "step": 1025,
      "train_runtime": 13146.4427,
      "train_tokens_per_second": 4684.091
    },
    {
      "epoch": 0.13137755102040816,
      "grad_norm": 0.6478268207128107,
      "learning_rate": 9.894113756745362e-06,
      "loss": 0.34624972343444826,
      "num_input_tokens_seen": 61888048,
      "step": 1030,
      "train_runtime": 13207.1108,
      "train_tokens_per_second": 4685.964
    },
    {
      "epoch": 0.13201530612244897,
      "grad_norm": 0.643687752492837,
      "learning_rate": 9.893085927744302e-06,
      "loss": 0.31321775913238525,
      "num_input_tokens_seen": 62182464,
      "step": 1035,
      "train_runtime": 13281.3897,
      "train_tokens_per_second": 4681.925
    },
    {
      "epoch": 0.1326530612244898,
      "grad_norm": 0.7749921845531136,
      "learning_rate": 9.892053188189778e-06,
      "loss": 0.3312078475952148,
      "num_input_tokens_seen": 62481672,
      "step": 1040,
      "train_runtime": 13341.1358,
      "train_tokens_per_second": 4683.385
    },
    {
      "epoch": 0.13329081632653061,
      "grad_norm": 0.6176461633030148,
      "learning_rate": 9.89101553911822e-06,
      "loss": 0.33130059242248533,
      "num_input_tokens_seen": 62770304,
      "step": 1045,
      "train_runtime": 13398.0925,
      "train_tokens_per_second": 4685.018
    },
    {
      "epoch": 0.13392857142857142,
      "grad_norm": 0.5640493570474314,
      "learning_rate": 9.889972981570984e-06,
      "loss": 0.34102232456207277,
      "num_input_tokens_seen": 63079128,
      "step": 1050,
      "train_runtime": 13453.0172,
      "train_tokens_per_second": 4688.846
    },
    {
      "epoch": 0.13456632653061223,
      "grad_norm": 0.6098806851952427,
      "learning_rate": 9.888925516594344e-06,
      "loss": 0.31255135536193845,
      "num_input_tokens_seen": 63374288,
      "step": 1055,
      "train_runtime": 13520.5802,
      "train_tokens_per_second": 4687.246
    },
    {
      "epoch": 0.13520408163265307,
      "grad_norm": 0.6016838976343992,
      "learning_rate": 9.887873145239505e-06,
      "loss": 0.32413506507873535,
      "num_input_tokens_seen": 63685544,
      "step": 1060,
      "train_runtime": 13583.1412,
      "train_tokens_per_second": 4688.573
    },
    {
      "epoch": 0.13584183673469388,
      "grad_norm": 0.6974335096565114,
      "learning_rate": 9.886815868562596e-06,
      "loss": 0.3247077226638794,
      "num_input_tokens_seen": 63985992,
      "step": 1065,
      "train_runtime": 13641.8081,
      "train_tokens_per_second": 4690.433
    },
    {
      "epoch": 0.1364795918367347,
      "grad_norm": 0.5503294330866604,
      "learning_rate": 9.885753687624671e-06,
      "loss": 0.32499494552612307,
      "num_input_tokens_seen": 64290784,
      "step": 1070,
      "train_runtime": 13708.5892,
      "train_tokens_per_second": 4689.818
    },
    {
      "epoch": 0.1371173469387755,
      "grad_norm": 0.5497811927600912,
      "learning_rate": 9.884686603491697e-06,
      "loss": 0.30507540702819824,
      "num_input_tokens_seen": 64580560,
      "step": 1075,
      "train_runtime": 13774.1678,
      "train_tokens_per_second": 4688.527
    },
    {
      "epoch": 0.1377551020408163,
      "grad_norm": 0.7370277106298851,
      "learning_rate": 9.883614617234574e-06,
      "loss": 0.31615376472473145,
      "num_input_tokens_seen": 64887752,
      "step": 1080,
      "train_runtime": 13843.7194,
      "train_tokens_per_second": 4687.162
    },
    {
      "epoch": 0.13839285714285715,
      "grad_norm": 0.5747869854992422,
      "learning_rate": 9.882537729929111e-06,
      "loss": 0.31842334270477296,
      "num_input_tokens_seen": 65177312,
      "step": 1085,
      "train_runtime": 13900.8839,
      "train_tokens_per_second": 4688.717
    },
    {
      "epoch": 0.13903061224489796,
      "grad_norm": 0.5994241942474536,
      "learning_rate": 9.88145594265604e-06,
      "loss": 0.33463475704193113,
      "num_input_tokens_seen": 65477552,
      "step": 1090,
      "train_runtime": 13957.0563,
      "train_tokens_per_second": 4691.358
    },
    {
      "epoch": 0.13966836734693877,
      "grad_norm": 0.5350594766885293,
      "learning_rate": 9.880369256501009e-06,
      "loss": 0.32270474433898927,
      "num_input_tokens_seen": 65773704,
      "step": 1095,
      "train_runtime": 14020.0614,
      "train_tokens_per_second": 4691.399
    },
    {
      "epoch": 0.14030612244897958,
      "grad_norm": 0.5856356402792805,
      "learning_rate": 9.879277672554586e-06,
      "loss": 0.3453704357147217,
      "num_input_tokens_seen": 66078288,
      "step": 1100,
      "train_runtime": 14092.98,
      "train_tokens_per_second": 4688.738
    },
    {
      "epoch": 0.14094387755102042,
      "grad_norm": 0.7066698779118917,
      "learning_rate": 9.878181191912251e-06,
      "loss": 0.342348051071167,
      "num_input_tokens_seen": 66389960,
      "step": 1105,
      "train_runtime": 14148.192,
      "train_tokens_per_second": 4692.47
    },
    {
      "epoch": 0.14158163265306123,
      "grad_norm": 0.6732728396315365,
      "learning_rate": 9.877079815674397e-06,
      "loss": 0.336626935005188,
      "num_input_tokens_seen": 66687024,
      "step": 1110,
      "train_runtime": 14215.1781,
      "train_tokens_per_second": 4691.255
    },
    {
      "epoch": 0.14221938775510204,
      "grad_norm": 0.6253707519183908,
      "learning_rate": 9.875973544946335e-06,
      "loss": 0.3181014060974121,
      "num_input_tokens_seen": 66992032,
      "step": 1115,
      "train_runtime": 14286.1982,
      "train_tokens_per_second": 4689.283
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.6049025984710364,
      "learning_rate": 9.874862380838284e-06,
      "loss": 0.3243386745452881,
      "num_input_tokens_seen": 67301344,
      "step": 1120,
      "train_runtime": 14357.9471,
      "train_tokens_per_second": 4687.393
    },
    {
      "epoch": 0.1434948979591837,
      "grad_norm": 0.581962219439854,
      "learning_rate": 9.873746324465372e-06,
      "loss": 0.32609100341796876,
      "num_input_tokens_seen": 67605624,
      "step": 1125,
      "train_runtime": 14425.1262,
      "train_tokens_per_second": 4686.657
    },
    {
      "epoch": 0.1441326530612245,
      "grad_norm": 0.6891948701654147,
      "learning_rate": 9.872625376947642e-06,
      "loss": 0.36863346099853517,
      "num_input_tokens_seen": 67914000,
      "step": 1130,
      "train_runtime": 14489.5411,
      "train_tokens_per_second": 4687.105
    },
    {
      "epoch": 0.1447704081632653,
      "grad_norm": 0.6585553967582634,
      "learning_rate": 9.871499539410045e-06,
      "loss": 0.336389422416687,
      "num_input_tokens_seen": 68208728,
      "step": 1135,
      "train_runtime": 14551.1305,
      "train_tokens_per_second": 4687.521
    },
    {
      "epoch": 0.14540816326530612,
      "grad_norm": 0.6240599879124094,
      "learning_rate": 9.870368812982433e-06,
      "loss": 0.3170750379562378,
      "num_input_tokens_seen": 68503840,
      "step": 1140,
      "train_runtime": 14623.6933,
      "train_tokens_per_second": 4684.442
    },
    {
      "epoch": 0.14604591836734693,
      "grad_norm": 0.6363383880401341,
      "learning_rate": 9.869233198799572e-06,
      "loss": 0.3022523164749146,
      "num_input_tokens_seen": 68794536,
      "step": 1145,
      "train_runtime": 14683.4451,
      "train_tokens_per_second": 4685.177
    },
    {
      "epoch": 0.14668367346938777,
      "grad_norm": 0.5788965255735574,
      "learning_rate": 9.86809269800113e-06,
      "loss": 0.3102893829345703,
      "num_input_tokens_seen": 69097584,
      "step": 1150,
      "train_runtime": 14746.5166,
      "train_tokens_per_second": 4685.689
    },
    {
      "epoch": 0.14732142857142858,
      "grad_norm": 0.540792417537845,
      "learning_rate": 9.86694731173168e-06,
      "loss": 0.3258485794067383,
      "num_input_tokens_seen": 69400448,
      "step": 1155,
      "train_runtime": 14823.6477,
      "train_tokens_per_second": 4681.739
    },
    {
      "epoch": 0.14795918367346939,
      "grad_norm": 0.6371755914466374,
      "learning_rate": 9.865797041140695e-06,
      "loss": 0.3410828113555908,
      "num_input_tokens_seen": 69694432,
      "step": 1160,
      "train_runtime": 14898.2847,
      "train_tokens_per_second": 4678.017
    },
    {
      "epoch": 0.1485969387755102,
      "grad_norm": 0.6907737129507281,
      "learning_rate": 9.864641887382552e-06,
      "loss": 0.3313009262084961,
      "num_input_tokens_seen": 70000464,
      "step": 1165,
      "train_runtime": 14973.2259,
      "train_tokens_per_second": 4675.042
    },
    {
      "epoch": 0.14923469387755103,
      "grad_norm": 0.645599339462072,
      "learning_rate": 9.863481851616528e-06,
      "loss": 0.33059353828430177,
      "num_input_tokens_seen": 70291160,
      "step": 1170,
      "train_runtime": 15041.846,
      "train_tokens_per_second": 4673.041
    },
    {
      "epoch": 0.14987244897959184,
      "grad_norm": 0.6778562215219686,
      "learning_rate": 9.862316935006803e-06,
      "loss": 0.34509251117706297,
      "num_input_tokens_seen": 70595328,
      "step": 1175,
      "train_runtime": 15107.0039,
      "train_tokens_per_second": 4673.02
    },
    {
      "epoch": 0.15051020408163265,
      "grad_norm": 0.5947368297067033,
      "learning_rate": 9.86114713872245e-06,
      "loss": 0.3179985523223877,
      "num_input_tokens_seen": 70891792,
      "step": 1180,
      "train_runtime": 15179.6228,
      "train_tokens_per_second": 4670.195
    },
    {
      "epoch": 0.15114795918367346,
      "grad_norm": 0.639355512325466,
      "learning_rate": 9.85997246393744e-06,
      "loss": 0.2997492551803589,
      "num_input_tokens_seen": 71187232,
      "step": 1185,
      "train_runtime": 15243.6794,
      "train_tokens_per_second": 4669.951
    },
    {
      "epoch": 0.15178571428571427,
      "grad_norm": 0.5952671858659276,
      "learning_rate": 9.858792911830643e-06,
      "loss": 0.3210294246673584,
      "num_input_tokens_seen": 71495592,
      "step": 1190,
      "train_runtime": 15310.89,
      "train_tokens_per_second": 4669.591
    },
    {
      "epoch": 0.1524234693877551,
      "grad_norm": 0.5932781344053283,
      "learning_rate": 9.857608483585823e-06,
      "loss": 0.32401361465454104,
      "num_input_tokens_seen": 71815632,
      "step": 1195,
      "train_runtime": 15374.9978,
      "train_tokens_per_second": 4670.936
    },
    {
      "epoch": 0.15306122448979592,
      "grad_norm": 0.513803869774437,
      "learning_rate": 9.856419180391636e-06,
      "loss": 0.2997368097305298,
      "num_input_tokens_seen": 72127616,
      "step": 1200,
      "train_runtime": 15437.5576,
      "train_tokens_per_second": 4672.217
    },
    {
      "epoch": 0.15369897959183673,
      "grad_norm": 0.6406041180556482,
      "learning_rate": 9.855225003441629e-06,
      "loss": 0.3473092555999756,
      "num_input_tokens_seen": 72438664,
      "step": 1205,
      "train_runtime": 15502.8818,
      "train_tokens_per_second": 4672.593
    },
    {
      "epoch": 0.15433673469387754,
      "grad_norm": 0.6204091343286625,
      "learning_rate": 9.85402595393424e-06,
      "loss": 0.3153236865997314,
      "num_input_tokens_seen": 72724400,
      "step": 1210,
      "train_runtime": 15570.96,
      "train_tokens_per_second": 4670.515
    },
    {
      "epoch": 0.15497448979591838,
      "grad_norm": 0.6226101352218131,
      "learning_rate": 9.852822033072805e-06,
      "loss": 0.33648636341094973,
      "num_input_tokens_seen": 73022520,
      "step": 1215,
      "train_runtime": 15638.4136,
      "train_tokens_per_second": 4669.433
    },
    {
      "epoch": 0.1556122448979592,
      "grad_norm": 0.616337565200306,
      "learning_rate": 9.85161324206554e-06,
      "loss": 0.3274383544921875,
      "num_input_tokens_seen": 73330568,
      "step": 1220,
      "train_runtime": 15700.4452,
      "train_tokens_per_second": 4670.604
    },
    {
      "epoch": 0.15625,
      "grad_norm": 0.5865853821331624,
      "learning_rate": 9.850399582125548e-06,
      "loss": 0.3142099857330322,
      "num_input_tokens_seen": 73637840,
      "step": 1225,
      "train_runtime": 15757.748,
      "train_tokens_per_second": 4673.12
    },
    {
      "epoch": 0.1568877551020408,
      "grad_norm": 0.6033716859627452,
      "learning_rate": 9.849181054470827e-06,
      "loss": 0.30474565029144285,
      "num_input_tokens_seen": 73923712,
      "step": 1230,
      "train_runtime": 15822.0851,
      "train_tokens_per_second": 4672.185
    },
    {
      "epoch": 0.15752551020408162,
      "grad_norm": 0.5910514385779135,
      "learning_rate": 9.84795766032425e-06,
      "loss": 0.314410138130188,
      "num_input_tokens_seen": 74208384,
      "step": 1235,
      "train_runtime": 15882.1475,
      "train_tokens_per_second": 4672.44
    },
    {
      "epoch": 0.15816326530612246,
      "grad_norm": 0.749622008061032,
      "learning_rate": 9.84672940091358e-06,
      "loss": 0.3144389629364014,
      "num_input_tokens_seen": 74496304,
      "step": 1240,
      "train_runtime": 15950.3528,
      "train_tokens_per_second": 4670.511
    },
    {
      "epoch": 0.15880102040816327,
      "grad_norm": 0.6930546014040261,
      "learning_rate": 9.845496277471463e-06,
      "loss": 0.3085564136505127,
      "num_input_tokens_seen": 74781656,
      "step": 1245,
      "train_runtime": 16018.7634,
      "train_tokens_per_second": 4668.379
    },
    {
      "epoch": 0.15943877551020408,
      "grad_norm": 0.5848372027862598,
      "learning_rate": 9.84425829123542e-06,
      "loss": 0.3176731109619141,
      "num_input_tokens_seen": 75095368,
      "step": 1250,
      "train_runtime": 16084.6142,
      "train_tokens_per_second": 4668.77
    },
    {
      "epoch": 0.1600765306122449,
      "grad_norm": 1.5703473449110465,
      "learning_rate": 9.84301544344786e-06,
      "loss": 0.33735177516937254,
      "num_input_tokens_seen": 75394808,
      "step": 1255,
      "train_runtime": 16144.4235,
      "train_tokens_per_second": 4670.022
    },
    {
      "epoch": 0.16071428571428573,
      "grad_norm": 0.5655628998642258,
      "learning_rate": 9.841767735356069e-06,
      "loss": 0.29922213554382326,
      "num_input_tokens_seen": 75696816,
      "step": 1260,
      "train_runtime": 16213.9234,
      "train_tokens_per_second": 4668.63
    },
    {
      "epoch": 0.16135204081632654,
      "grad_norm": 0.5468474890427991,
      "learning_rate": 9.840515168212208e-06,
      "loss": 0.3142865180969238,
      "num_input_tokens_seen": 75997032,
      "step": 1265,
      "train_runtime": 16269.2171,
      "train_tokens_per_second": 4671.216
    },
    {
      "epoch": 0.16198979591836735,
      "grad_norm": 0.6977623447232346,
      "learning_rate": 9.839257743273313e-06,
      "loss": 0.3169466495513916,
      "num_input_tokens_seen": 76306232,
      "step": 1270,
      "train_runtime": 16336.5091,
      "train_tokens_per_second": 4670.902
    },
    {
      "epoch": 0.16262755102040816,
      "grad_norm": 0.6307116574118067,
      "learning_rate": 9.8379954618013e-06,
      "loss": 0.3264782905578613,
      "num_input_tokens_seen": 76616680,
      "step": 1275,
      "train_runtime": 16403.3467,
      "train_tokens_per_second": 4670.796
    },
    {
      "epoch": 0.16326530612244897,
      "grad_norm": 0.6481160726385983,
      "learning_rate": 9.836728325062956e-06,
      "loss": 0.32020001411437987,
      "num_input_tokens_seen": 76911192,
      "step": 1280,
      "train_runtime": 16466.5796,
      "train_tokens_per_second": 4670.745
    },
    {
      "epoch": 0.1639030612244898,
      "grad_norm": 0.6260219252872821,
      "learning_rate": 9.83545633432994e-06,
      "loss": 0.3197026252746582,
      "num_input_tokens_seen": 77213144,
      "step": 1285,
      "train_runtime": 16538.6986,
      "train_tokens_per_second": 4668.635
    },
    {
      "epoch": 0.16454081632653061,
      "grad_norm": 0.6061939398138922,
      "learning_rate": 9.834179490878789e-06,
      "loss": 0.32415170669555665,
      "num_input_tokens_seen": 77511552,
      "step": 1290,
      "train_runtime": 16608.3086,
      "train_tokens_per_second": 4667.035
    },
    {
      "epoch": 0.16517857142857142,
      "grad_norm": 0.624518323832078,
      "learning_rate": 9.832897795990896e-06,
      "loss": 0.3314162254333496,
      "num_input_tokens_seen": 77819352,
      "step": 1295,
      "train_runtime": 16670.9625,
      "train_tokens_per_second": 4667.958
    },
    {
      "epoch": 0.16581632653061223,
      "grad_norm": 0.6035466373238898,
      "learning_rate": 9.83161125095254e-06,
      "loss": 0.3390120267868042,
      "num_input_tokens_seen": 78124624,
      "step": 1300,
      "train_runtime": 16722.1329,
      "train_tokens_per_second": 4671.929
    },
    {
      "epoch": 0.16645408163265307,
      "grad_norm": 0.604635713790092,
      "learning_rate": 9.830319857054853e-06,
      "loss": 0.32261736392974855,
      "num_input_tokens_seen": 78425016,
      "step": 1305,
      "train_runtime": 16782.9521,
      "train_tokens_per_second": 4672.898
    },
    {
      "epoch": 0.16709183673469388,
      "grad_norm": 0.6308378582995104,
      "learning_rate": 9.82902361559384e-06,
      "loss": 0.3400526523590088,
      "num_input_tokens_seen": 78725448,
      "step": 1310,
      "train_runtime": 16845.0134,
      "train_tokens_per_second": 4673.517
    },
    {
      "epoch": 0.1677295918367347,
      "grad_norm": 0.6775668307882584,
      "learning_rate": 9.82772252787037e-06,
      "loss": 0.32093000411987305,
      "num_input_tokens_seen": 79019432,
      "step": 1315,
      "train_runtime": 16898.2536,
      "train_tokens_per_second": 4676.189
    },
    {
      "epoch": 0.1683673469387755,
      "grad_norm": 0.7150284876508191,
      "learning_rate": 9.826416595190177e-06,
      "loss": 0.32538139820098877,
      "num_input_tokens_seen": 79302880,
      "step": 1320,
      "train_runtime": 16960.6423,
      "train_tokens_per_second": 4675.7
    },
    {
      "epoch": 0.1690051020408163,
      "grad_norm": 0.6080349805720094,
      "learning_rate": 9.825105818863854e-06,
      "loss": 0.2659433603286743,
      "num_input_tokens_seen": 79609552,
      "step": 1325,
      "train_runtime": 17024.5507,
      "train_tokens_per_second": 4676.162
    },
    {
      "epoch": 0.16964285714285715,
      "grad_norm": 0.6644582121488926,
      "learning_rate": 9.823790200206857e-06,
      "loss": 0.35074310302734374,
      "num_input_tokens_seen": 79911872,
      "step": 1330,
      "train_runtime": 17091.2088,
      "train_tokens_per_second": 4675.613
    },
    {
      "epoch": 0.17028061224489796,
      "grad_norm": 0.5988993982839885,
      "learning_rate": 9.822469740539503e-06,
      "loss": 0.3313178539276123,
      "num_input_tokens_seen": 80214880,
      "step": 1335,
      "train_runtime": 17152.7302,
      "train_tokens_per_second": 4676.508
    },
    {
      "epoch": 0.17091836734693877,
      "grad_norm": 0.6734717878509392,
      "learning_rate": 9.821144441186964e-06,
      "loss": 0.29335453510284426,
      "num_input_tokens_seen": 80505568,
      "step": 1340,
      "train_runtime": 17220.5566,
      "train_tokens_per_second": 4674.969
    },
    {
      "epoch": 0.17155612244897958,
      "grad_norm": 0.7450898696825456,
      "learning_rate": 9.819814303479268e-06,
      "loss": 0.3468411922454834,
      "num_input_tokens_seen": 80798056,
      "step": 1345,
      "train_runtime": 17275.7965,
      "train_tokens_per_second": 4676.951
    },
    {
      "epoch": 0.17219387755102042,
      "grad_norm": 0.6193250292674742,
      "learning_rate": 9.818479328751303e-06,
      "loss": 0.3229133129119873,
      "num_input_tokens_seen": 81077272,
      "step": 1350,
      "train_runtime": 17338.1589,
      "train_tokens_per_second": 4676.233
    },
    {
      "epoch": 0.17283163265306123,
      "grad_norm": 0.5734338838942346,
      "learning_rate": 9.817139518342811e-06,
      "loss": 0.34751739501953127,
      "num_input_tokens_seen": 81381760,
      "step": 1355,
      "train_runtime": 17404.5748,
      "train_tokens_per_second": 4675.883
    },
    {
      "epoch": 0.17346938775510204,
      "grad_norm": 0.598503096145298,
      "learning_rate": 9.815794873598384e-06,
      "loss": 0.32517199516296386,
      "num_input_tokens_seen": 81689296,
      "step": 1360,
      "train_runtime": 17468.4889,
      "train_tokens_per_second": 4676.38
    },
    {
      "epoch": 0.17410714285714285,
      "grad_norm": 0.5786479595145131,
      "learning_rate": 9.814445395867467e-06,
      "loss": 0.32674236297607423,
      "num_input_tokens_seen": 81993936,
      "step": 1365,
      "train_runtime": 17535.8677,
      "train_tokens_per_second": 4675.784
    },
    {
      "epoch": 0.1747448979591837,
      "grad_norm": 0.6426414968956544,
      "learning_rate": 9.813091086504356e-06,
      "loss": 0.33481631278991697,
      "num_input_tokens_seen": 82291264,
      "step": 1370,
      "train_runtime": 17598.9769,
      "train_tokens_per_second": 4675.912
    },
    {
      "epoch": 0.1753826530612245,
      "grad_norm": 0.6772863683681803,
      "learning_rate": 9.811731946868192e-06,
      "loss": 0.3312889575958252,
      "num_input_tokens_seen": 82576648,
      "step": 1375,
      "train_runtime": 17674.1013,
      "train_tokens_per_second": 4672.184
    },
    {
      "epoch": 0.1760204081632653,
      "grad_norm": 0.6183891713769821,
      "learning_rate": 9.810367978322971e-06,
      "loss": 0.32770624160766604,
      "num_input_tokens_seen": 82855264,
      "step": 1380,
      "train_runtime": 17739.7033,
      "train_tokens_per_second": 4670.612
    },
    {
      "epoch": 0.17665816326530612,
      "grad_norm": 0.6743162881667827,
      "learning_rate": 9.808999182237527e-06,
      "loss": 0.32761325836181643,
      "num_input_tokens_seen": 83163056,
      "step": 1385,
      "train_runtime": 17809.4298,
      "train_tokens_per_second": 4669.608
    },
    {
      "epoch": 0.17729591836734693,
      "grad_norm": 0.6417615450472127,
      "learning_rate": 9.807625559985546e-06,
      "loss": 0.31441588401794435,
      "num_input_tokens_seen": 83459208,
      "step": 1390,
      "train_runtime": 17870.7623,
      "train_tokens_per_second": 4670.154
    },
    {
      "epoch": 0.17793367346938777,
      "grad_norm": 0.5655903433812112,
      "learning_rate": 9.806247112945548e-06,
      "loss": 0.30176577568054197,
      "num_input_tokens_seen": 83768232,
      "step": 1395,
      "train_runtime": 17933.5064,
      "train_tokens_per_second": 4671.046
    },
    {
      "epoch": 0.17857142857142858,
      "grad_norm": 0.6186427531608422,
      "learning_rate": 9.804863842500906e-06,
      "loss": 0.3389891147613525,
      "num_input_tokens_seen": 84066288,
      "step": 1400,
      "train_runtime": 17998.6148,
      "train_tokens_per_second": 4670.709
    },
    {
      "epoch": 0.17920918367346939,
      "grad_norm": 0.577582621616322,
      "learning_rate": 9.803475750039828e-06,
      "loss": 0.3311573028564453,
      "num_input_tokens_seen": 84367984,
      "step": 1405,
      "train_runtime": 18069.4343,
      "train_tokens_per_second": 4669.099
    },
    {
      "epoch": 0.1798469387755102,
      "grad_norm": 0.5793464491006457,
      "learning_rate": 9.80208283695536e-06,
      "loss": 0.3258392095565796,
      "num_input_tokens_seen": 84669608,
      "step": 1410,
      "train_runtime": 18144.4137,
      "train_tokens_per_second": 4666.428
    },
    {
      "epoch": 0.18048469387755103,
      "grad_norm": 0.576273241426476,
      "learning_rate": 9.80068510464539e-06,
      "loss": 0.3283647298812866,
      "num_input_tokens_seen": 84968192,
      "step": 1415,
      "train_runtime": 18218.7607,
      "train_tokens_per_second": 4663.775
    },
    {
      "epoch": 0.18112244897959184,
      "grad_norm": 0.6027961814579025,
      "learning_rate": 9.799282554512637e-06,
      "loss": 0.31317665576934817,
      "num_input_tokens_seen": 85267272,
      "step": 1420,
      "train_runtime": 18275.5674,
      "train_tokens_per_second": 4665.643
    },
    {
      "epoch": 0.18176020408163265,
      "grad_norm": 0.6472109682658503,
      "learning_rate": 9.797875187964661e-06,
      "loss": 0.36163792610168455,
      "num_input_tokens_seen": 85573528,
      "step": 1425,
      "train_runtime": 18350.5815,
      "train_tokens_per_second": 4663.26
    },
    {
      "epoch": 0.18239795918367346,
      "grad_norm": 0.6570387718669111,
      "learning_rate": 9.79646300641385e-06,
      "loss": 0.33827977180480956,
      "num_input_tokens_seen": 85888168,
      "step": 1430,
      "train_runtime": 18407.6843,
      "train_tokens_per_second": 4665.887
    },
    {
      "epoch": 0.18303571428571427,
      "grad_norm": 0.5915865126409967,
      "learning_rate": 9.795046011277427e-06,
      "loss": 0.2988413333892822,
      "num_input_tokens_seen": 86197056,
      "step": 1435,
      "train_runtime": 18484.4433,
      "train_tokens_per_second": 4663.222
    },
    {
      "epoch": 0.1836734693877551,
      "grad_norm": 0.6319044202710384,
      "learning_rate": 9.793624203977446e-06,
      "loss": 0.3205398082733154,
      "num_input_tokens_seen": 86495336,
      "step": 1440,
      "train_runtime": 18549.8044,
      "train_tokens_per_second": 4662.871
    },
    {
      "epoch": 0.18431122448979592,
      "grad_norm": 0.602782954776995,
      "learning_rate": 9.792197585940792e-06,
      "loss": 0.3298956871032715,
      "num_input_tokens_seen": 86813224,
      "step": 1445,
      "train_runtime": 18619.406,
      "train_tokens_per_second": 4662.513
    },
    {
      "epoch": 0.18494897959183673,
      "grad_norm": 0.5754024523751752,
      "learning_rate": 9.790766158599172e-06,
      "loss": 0.30818729400634765,
      "num_input_tokens_seen": 87113448,
      "step": 1450,
      "train_runtime": 18686.6467,
      "train_tokens_per_second": 4661.802
    },
    {
      "epoch": 0.18558673469387754,
      "grad_norm": 0.6881805944021485,
      "learning_rate": 9.789329923389128e-06,
      "loss": 0.32339422702789306,
      "num_input_tokens_seen": 87413584,
      "step": 1455,
      "train_runtime": 18754.1814,
      "train_tokens_per_second": 4661.018
    },
    {
      "epoch": 0.18622448979591838,
      "grad_norm": 0.6193905311957907,
      "learning_rate": 9.787888881752018e-06,
      "loss": 0.3140754699707031,
      "num_input_tokens_seen": 87714304,
      "step": 1460,
      "train_runtime": 18813.2731,
      "train_tokens_per_second": 4662.363
    },
    {
      "epoch": 0.1868622448979592,
      "grad_norm": 0.5697777601138168,
      "learning_rate": 9.78644303513403e-06,
      "loss": 0.32805964946746824,
      "num_input_tokens_seen": 88010424,
      "step": 1465,
      "train_runtime": 18884.5056,
      "train_tokens_per_second": 4660.457
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.5761455879704065,
      "learning_rate": 9.784992384986173e-06,
      "loss": 0.3325707674026489,
      "num_input_tokens_seen": 88312208,
      "step": 1470,
      "train_runtime": 18945.8281,
      "train_tokens_per_second": 4661.301
    },
    {
      "epoch": 0.1881377551020408,
      "grad_norm": 0.6312574029931073,
      "learning_rate": 9.783536932764273e-06,
      "loss": 0.31943345069885254,
      "num_input_tokens_seen": 88600744,
      "step": 1475,
      "train_runtime": 19010.9206,
      "train_tokens_per_second": 4660.518
    },
    {
      "epoch": 0.18877551020408162,
      "grad_norm": 0.5557668031229552,
      "learning_rate": 9.782076679928981e-06,
      "loss": 0.30451416969299316,
      "num_input_tokens_seen": 88897368,
      "step": 1480,
      "train_runtime": 19074.2654,
      "train_tokens_per_second": 4660.592
    },
    {
      "epoch": 0.18941326530612246,
      "grad_norm": 0.6064797044911235,
      "learning_rate": 9.78061162794576e-06,
      "loss": 0.3167304277420044,
      "num_input_tokens_seen": 89206296,
      "step": 1485,
      "train_runtime": 19135.7143,
      "train_tokens_per_second": 4661.77
    },
    {
      "epoch": 0.19005102040816327,
      "grad_norm": 0.7163669268796902,
      "learning_rate": 9.779141778284893e-06,
      "loss": 0.32610111236572265,
      "num_input_tokens_seen": 89519288,
      "step": 1490,
      "train_runtime": 19186.547,
      "train_tokens_per_second": 4665.732
    },
    {
      "epoch": 0.19068877551020408,
      "grad_norm": 0.6028664586912892,
      "learning_rate": 9.77766713242148e-06,
      "loss": 0.3151542663574219,
      "num_input_tokens_seen": 89810448,
      "step": 1495,
      "train_runtime": 19253.8342,
      "train_tokens_per_second": 4664.549
    },
    {
      "epoch": 0.1913265306122449,
      "grad_norm": 0.5932933121302334,
      "learning_rate": 9.776187691835424e-06,
      "loss": 0.3264111042022705,
      "num_input_tokens_seen": 90118960,
      "step": 1500,
      "train_runtime": 19328.9784,
      "train_tokens_per_second": 4662.376
    },
    {
      "epoch": 0.19196428571428573,
      "grad_norm": 0.5718879186944182,
      "learning_rate": 9.774703458011453e-06,
      "loss": 0.31701092720031737,
      "num_input_tokens_seen": 90423136,
      "step": 1505,
      "train_runtime": 19384.6841,
      "train_tokens_per_second": 4664.669
    },
    {
      "epoch": 0.19260204081632654,
      "grad_norm": 0.5732930244436449,
      "learning_rate": 9.773214432439098e-06,
      "loss": 0.32857470512390136,
      "num_input_tokens_seen": 90716256,
      "step": 1510,
      "train_runtime": 19449.9174,
      "train_tokens_per_second": 4664.095
    },
    {
      "epoch": 0.19323979591836735,
      "grad_norm": 0.6264692762177914,
      "learning_rate": 9.771720616612697e-06,
      "loss": 0.3403751373291016,
      "num_input_tokens_seen": 91000816,
      "step": 1515,
      "train_runtime": 19513.2349,
      "train_tokens_per_second": 4663.543
    },
    {
      "epoch": 0.19387755102040816,
      "grad_norm": 0.5879142904469583,
      "learning_rate": 9.770222012031404e-06,
      "loss": 0.3135849952697754,
      "num_input_tokens_seen": 91292152,
      "step": 1520,
      "train_runtime": 19588.8663,
      "train_tokens_per_second": 4660.41
    },
    {
      "epoch": 0.19451530612244897,
      "grad_norm": 0.6186807155432585,
      "learning_rate": 9.768718620199168e-06,
      "loss": 0.3122847080230713,
      "num_input_tokens_seen": 91584544,
      "step": 1525,
      "train_runtime": 19655.1895,
      "train_tokens_per_second": 4659.56
    },
    {
      "epoch": 0.1951530612244898,
      "grad_norm": 0.6143695774772617,
      "learning_rate": 9.76721044262475e-06,
      "loss": 0.302173376083374,
      "num_input_tokens_seen": 91892136,
      "step": 1530,
      "train_runtime": 19715.1758,
      "train_tokens_per_second": 4660.985
    },
    {
      "epoch": 0.19579081632653061,
      "grad_norm": 0.6423532206679058,
      "learning_rate": 9.765697480821714e-06,
      "loss": 0.32527501583099366,
      "num_input_tokens_seen": 92199840,
      "step": 1535,
      "train_runtime": 19771.8732,
      "train_tokens_per_second": 4663.182
    },
    {
      "epoch": 0.19642857142857142,
      "grad_norm": 0.8069967571373494,
      "learning_rate": 9.764179736308419e-06,
      "loss": 0.3400618553161621,
      "num_input_tokens_seen": 92513480,
      "step": 1540,
      "train_runtime": 19826.0737,
      "train_tokens_per_second": 4666.253
    },
    {
      "epoch": 0.19706632653061223,
      "grad_norm": 0.9233258986747941,
      "learning_rate": 9.762657210608028e-06,
      "loss": 0.3346916675567627,
      "num_input_tokens_seen": 92813864,
      "step": 1545,
      "train_runtime": 19884.5282,
      "train_tokens_per_second": 4667.642
    },
    {
      "epoch": 0.19770408163265307,
      "grad_norm": 0.8489790906182042,
      "learning_rate": 9.761129905248505e-06,
      "loss": 0.3408361911773682,
      "num_input_tokens_seen": 93112808,
      "step": 1550,
      "train_runtime": 19955.849,
      "train_tokens_per_second": 4665.941
    },
    {
      "epoch": 0.19834183673469388,
      "grad_norm": 0.6666863063443117,
      "learning_rate": 9.759597821762603e-06,
      "loss": 0.32660608291625975,
      "num_input_tokens_seen": 93403528,
      "step": 1555,
      "train_runtime": 20025.1716,
      "train_tokens_per_second": 4664.306
    },
    {
      "epoch": 0.1989795918367347,
      "grad_norm": 0.5539952771846446,
      "learning_rate": 9.758060961687879e-06,
      "loss": 0.347269606590271,
      "num_input_tokens_seen": 93706008,
      "step": 1560,
      "train_runtime": 20085.773,
      "train_tokens_per_second": 4665.293
    },
    {
      "epoch": 0.1996173469387755,
      "grad_norm": 0.5466107019623242,
      "learning_rate": 9.756519326566677e-06,
      "loss": 0.3399479627609253,
      "num_input_tokens_seen": 94012760,
      "step": 1565,
      "train_runtime": 20155.7307,
      "train_tokens_per_second": 4664.319
    },
    {
      "epoch": 0.2002551020408163,
      "grad_norm": 0.585516484028218,
      "learning_rate": 9.754972917946138e-06,
      "loss": 0.3284050941467285,
      "num_input_tokens_seen": 94322480,
      "step": 1570,
      "train_runtime": 20220.8537,
      "train_tokens_per_second": 4664.614
    },
    {
      "epoch": 0.20089285714285715,
      "grad_norm": 0.6350719112756239,
      "learning_rate": 9.753421737378188e-06,
      "loss": 0.3520487070083618,
      "num_input_tokens_seen": 94623816,
      "step": 1575,
      "train_runtime": 20283.6263,
      "train_tokens_per_second": 4665.034
    },
    {
      "epoch": 0.20153061224489796,
      "grad_norm": 0.5088503114605749,
      "learning_rate": 9.751865786419546e-06,
      "loss": 0.3247365951538086,
      "num_input_tokens_seen": 94925560,
      "step": 1580,
      "train_runtime": 20340.2822,
      "train_tokens_per_second": 4666.875
    },
    {
      "epoch": 0.20216836734693877,
      "grad_norm": 0.5359700169183711,
      "learning_rate": 9.750305066631717e-06,
      "loss": 0.3115284204483032,
      "num_input_tokens_seen": 95240304,
      "step": 1585,
      "train_runtime": 20397.2195,
      "train_tokens_per_second": 4669.279
    },
    {
      "epoch": 0.20280612244897958,
      "grad_norm": 0.7442337531112403,
      "learning_rate": 9.748739579580995e-06,
      "loss": 0.32700955867767334,
      "num_input_tokens_seen": 95522168,
      "step": 1590,
      "train_runtime": 20455.8691,
      "train_tokens_per_second": 4669.67
    },
    {
      "epoch": 0.20344387755102042,
      "grad_norm": 0.5765541750227784,
      "learning_rate": 9.747169326838451e-06,
      "loss": 0.2974551200866699,
      "num_input_tokens_seen": 95817688,
      "step": 1595,
      "train_runtime": 20518.0377,
      "train_tokens_per_second": 4669.925
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 0.5344399026968158,
      "learning_rate": 9.74559430997995e-06,
      "loss": 0.286482572555542,
      "num_input_tokens_seen": 96124656,
      "step": 1600,
      "train_runtime": 20579.1449,
      "train_tokens_per_second": 4670.974
    },
    {
      "epoch": 0.20471938775510204,
      "grad_norm": 0.5542855767607231,
      "learning_rate": 9.744014530586124e-06,
      "loss": 0.3121712446212769,
      "num_input_tokens_seen": 96432192,
      "step": 1605,
      "train_runtime": 20636.432,
      "train_tokens_per_second": 4672.91
    },
    {
      "epoch": 0.20535714285714285,
      "grad_norm": 0.6417270797527082,
      "learning_rate": 9.742429990242394e-06,
      "loss": 0.3185253143310547,
      "num_input_tokens_seen": 96733072,
      "step": 1610,
      "train_runtime": 20690.8871,
      "train_tokens_per_second": 4675.153
    },
    {
      "epoch": 0.2059948979591837,
      "grad_norm": 0.8196337875958973,
      "learning_rate": 9.740840690538961e-06,
      "loss": 0.34393696784973143,
      "num_input_tokens_seen": 97033976,
      "step": 1615,
      "train_runtime": 20770.3778,
      "train_tokens_per_second": 4671.748
    },
    {
      "epoch": 0.2066326530612245,
      "grad_norm": 0.7553206240898032,
      "learning_rate": 9.739246633070796e-06,
      "loss": 0.3039816617965698,
      "num_input_tokens_seen": 97334640,
      "step": 1620,
      "train_runtime": 20826.8186,
      "train_tokens_per_second": 4673.524
    },
    {
      "epoch": 0.2072704081632653,
      "grad_norm": 0.5848461607387229,
      "learning_rate": 9.737647819437645e-06,
      "loss": 0.32238154411315917,
      "num_input_tokens_seen": 97635840,
      "step": 1625,
      "train_runtime": 20891.1271,
      "train_tokens_per_second": 4673.555
    },
    {
      "epoch": 0.20790816326530612,
      "grad_norm": 0.6590019837579646,
      "learning_rate": 9.73604425124403e-06,
      "loss": 0.3333409309387207,
      "num_input_tokens_seen": 97926552,
      "step": 1630,
      "train_runtime": 20959.8775,
      "train_tokens_per_second": 4672.096
    },
    {
      "epoch": 0.20854591836734693,
      "grad_norm": 0.6800291269132569,
      "learning_rate": 9.734435930099246e-06,
      "loss": 0.32056217193603515,
      "num_input_tokens_seen": 98218656,
      "step": 1635,
      "train_runtime": 21020.074,
      "train_tokens_per_second": 4672.612
    },
    {
      "epoch": 0.20918367346938777,
      "grad_norm": 0.5415844117036988,
      "learning_rate": 9.732822857617352e-06,
      "loss": 0.3002612113952637,
      "num_input_tokens_seen": 98504848,
      "step": 1640,
      "train_runtime": 21077.7169,
      "train_tokens_per_second": 4673.412
    },
    {
      "epoch": 0.20982142857142858,
      "grad_norm": 0.6193481368187909,
      "learning_rate": 9.731205035417183e-06,
      "loss": 0.3319518804550171,
      "num_input_tokens_seen": 98806944,
      "step": 1645,
      "train_runtime": 21140.7188,
      "train_tokens_per_second": 4673.774
    },
    {
      "epoch": 0.21045918367346939,
      "grad_norm": 0.6077166715888609,
      "learning_rate": 9.729582465122334e-06,
      "loss": 0.33286495208740235,
      "num_input_tokens_seen": 99115168,
      "step": 1650,
      "train_runtime": 21195.2064,
      "train_tokens_per_second": 4676.301
    },
    {
      "epoch": 0.2110969387755102,
      "grad_norm": 0.5909430297339548,
      "learning_rate": 9.727955148361166e-06,
      "loss": 0.32822155952453613,
      "num_input_tokens_seen": 99396568,
      "step": 1655,
      "train_runtime": 21261.9658,
      "train_tokens_per_second": 4674.853
    },
    {
      "epoch": 0.21173469387755103,
      "grad_norm": 0.6011483095378239,
      "learning_rate": 9.72632308676681e-06,
      "loss": 0.33887970447540283,
      "num_input_tokens_seen": 99714032,
      "step": 1660,
      "train_runtime": 21331.2956,
      "train_tokens_per_second": 4674.542
    },
    {
      "epoch": 0.21237244897959184,
      "grad_norm": 0.559470750563947,
      "learning_rate": 9.724686281977146e-06,
      "loss": 0.32210845947265626,
      "num_input_tokens_seen": 100000736,
      "step": 1665,
      "train_runtime": 21401.8927,
      "train_tokens_per_second": 4672.518
    },
    {
      "epoch": 0.21301020408163265,
      "grad_norm": 0.6567689329295388,
      "learning_rate": 9.72304473563483e-06,
      "loss": 0.33064305782318115,
      "num_input_tokens_seen": 100289480,
      "step": 1670,
      "train_runtime": 21463.9962,
      "train_tokens_per_second": 4672.451
    },
    {
      "epoch": 0.21364795918367346,
      "grad_norm": 0.6018245890126758,
      "learning_rate": 9.721398449387265e-06,
      "loss": 0.30869359970092775,
      "num_input_tokens_seen": 100577872,
      "step": 1675,
      "train_runtime": 21529.0437,
      "train_tokens_per_second": 4671.73
    },
    {
      "epoch": 0.21428571428571427,
      "grad_norm": 0.6395107833809925,
      "learning_rate": 9.719747424886613e-06,
      "loss": 0.3150346279144287,
      "num_input_tokens_seen": 100863168,
      "step": 1680,
      "train_runtime": 21599.2901,
      "train_tokens_per_second": 4669.745
    },
    {
      "epoch": 0.2149234693877551,
      "grad_norm": 0.5271941265038819,
      "learning_rate": 9.718091663789795e-06,
      "loss": 0.31960508823394773,
      "num_input_tokens_seen": 101162408,
      "step": 1685,
      "train_runtime": 21667.5749,
      "train_tokens_per_second": 4668.839
    },
    {
      "epoch": 0.21556122448979592,
      "grad_norm": 0.5481431804332924,
      "learning_rate": 9.716431167758482e-06,
      "loss": 0.3123780250549316,
      "num_input_tokens_seen": 101460984,
      "step": 1690,
      "train_runtime": 21728.5693,
      "train_tokens_per_second": 4669.474
    },
    {
      "epoch": 0.21619897959183673,
      "grad_norm": 0.6086299934433471,
      "learning_rate": 9.714765938459097e-06,
      "loss": 0.3333261013031006,
      "num_input_tokens_seen": 101761920,
      "step": 1695,
      "train_runtime": 21783.126,
      "train_tokens_per_second": 4671.594
    },
    {
      "epoch": 0.21683673469387754,
      "grad_norm": 0.6365447361149097,
      "learning_rate": 9.713095977562813e-06,
      "loss": 0.3120710849761963,
      "num_input_tokens_seen": 102053720,
      "step": 1700,
      "train_runtime": 21855.0072,
      "train_tokens_per_second": 4669.581
    },
    {
      "epoch": 0.21747448979591838,
      "grad_norm": 0.6855948828164128,
      "learning_rate": 9.711421286745554e-06,
      "loss": 0.3282258749008179,
      "num_input_tokens_seen": 102353896,
      "step": 1705,
      "train_runtime": 21918.2908,
      "train_tokens_per_second": 4669.794
    },
    {
      "epoch": 0.2181122448979592,
      "grad_norm": 0.6620982489776986,
      "learning_rate": 9.70974186768799e-06,
      "loss": 0.3506155252456665,
      "num_input_tokens_seen": 102660488,
      "step": 1710,
      "train_runtime": 21979.9884,
      "train_tokens_per_second": 4670.634
    },
    {
      "epoch": 0.21875,
      "grad_norm": 0.5350707355091577,
      "learning_rate": 9.708057722075532e-06,
      "loss": 0.2788844108581543,
      "num_input_tokens_seen": 102962576,
      "step": 1715,
      "train_runtime": 22039.8013,
      "train_tokens_per_second": 4671.665
    },
    {
      "epoch": 0.2193877551020408,
      "grad_norm": 0.5954127998391153,
      "learning_rate": 9.706368851598343e-06,
      "loss": 0.35607798099517823,
      "num_input_tokens_seen": 103270088,
      "step": 1720,
      "train_runtime": 22100.4497,
      "train_tokens_per_second": 4672.76
    },
    {
      "epoch": 0.22002551020408162,
      "grad_norm": 0.6113164632124081,
      "learning_rate": 9.704675257951316e-06,
      "loss": 0.3244358539581299,
      "num_input_tokens_seen": 103567896,
      "step": 1725,
      "train_runtime": 22171.8036,
      "train_tokens_per_second": 4671.153
    },
    {
      "epoch": 0.22066326530612246,
      "grad_norm": 0.6635617763751173,
      "learning_rate": 9.702976942834096e-06,
      "loss": 0.36413140296936036,
      "num_input_tokens_seen": 103858784,
      "step": 1730,
      "train_runtime": 22242.7813,
      "train_tokens_per_second": 4669.325
    },
    {
      "epoch": 0.22130102040816327,
      "grad_norm": 0.5945460106687103,
      "learning_rate": 9.701273907951058e-06,
      "loss": 0.32247157096862794,
      "num_input_tokens_seen": 104166592,
      "step": 1735,
      "train_runtime": 22311.0144,
      "train_tokens_per_second": 4668.842
    },
    {
      "epoch": 0.22193877551020408,
      "grad_norm": 0.5986015319544813,
      "learning_rate": 9.699566155011317e-06,
      "loss": 0.3243138790130615,
      "num_input_tokens_seen": 104481144,
      "step": 1740,
      "train_runtime": 22377.095,
      "train_tokens_per_second": 4669.111
    },
    {
      "epoch": 0.2225765306122449,
      "grad_norm": 0.6072340691688982,
      "learning_rate": 9.697853685728721e-06,
      "loss": 0.28355393409729,
      "num_input_tokens_seen": 104789096,
      "step": 1745,
      "train_runtime": 22448.0265,
      "train_tokens_per_second": 4668.076
    },
    {
      "epoch": 0.22321428571428573,
      "grad_norm": 0.7710206984061643,
      "learning_rate": 9.696136501821858e-06,
      "loss": 0.3441521406173706,
      "num_input_tokens_seen": 105099160,
      "step": 1750,
      "train_runtime": 22506.2221,
      "train_tokens_per_second": 4669.782
    },
    {
      "epoch": 0.22385204081632654,
      "grad_norm": 0.5534601761751993,
      "learning_rate": 9.694414605014035e-06,
      "loss": 0.2991414785385132,
      "num_input_tokens_seen": 105399368,
      "step": 1755,
      "train_runtime": 22572.1161,
      "train_tokens_per_second": 4669.45
    },
    {
      "epoch": 0.22448979591836735,
      "grad_norm": 0.6135564145614848,
      "learning_rate": 9.6926879970333e-06,
      "loss": 0.34076249599456787,
      "num_input_tokens_seen": 105713400,
      "step": 1760,
      "train_runtime": 22621.1206,
      "train_tokens_per_second": 4673.217
    },
    {
      "epoch": 0.22512755102040816,
      "grad_norm": 0.6390624645538395,
      "learning_rate": 9.690956679612422e-06,
      "loss": 0.3157844543457031,
      "num_input_tokens_seen": 106017008,
      "step": 1765,
      "train_runtime": 22683.2953,
      "train_tokens_per_second": 4673.792
    },
    {
      "epoch": 0.22576530612244897,
      "grad_norm": 0.6247895903764572,
      "learning_rate": 9.6892206544889e-06,
      "loss": 0.33294594287872314,
      "num_input_tokens_seen": 106333352,
      "step": 1770,
      "train_runtime": 22732.1296,
      "train_tokens_per_second": 4677.668
    },
    {
      "epoch": 0.2264030612244898,
      "grad_norm": 0.5941472594118004,
      "learning_rate": 9.687479923404959e-06,
      "loss": 0.3039743900299072,
      "num_input_tokens_seen": 106624944,
      "step": 1775,
      "train_runtime": 22799.7411,
      "train_tokens_per_second": 4676.586
    },
    {
      "epoch": 0.22704081632653061,
      "grad_norm": 0.598925695045739,
      "learning_rate": 9.68573448810754e-06,
      "loss": 0.32722058296203616,
      "num_input_tokens_seen": 106924616,
      "step": 1780,
      "train_runtime": 22861.3056,
      "train_tokens_per_second": 4677.1
    },
    {
      "epoch": 0.22767857142857142,
      "grad_norm": 0.565965130082789,
      "learning_rate": 9.683984350348313e-06,
      "loss": 0.32197444438934325,
      "num_input_tokens_seen": 107232608,
      "step": 1785,
      "train_runtime": 22936.9477,
      "train_tokens_per_second": 4675.104
    },
    {
      "epoch": 0.22831632653061223,
      "grad_norm": 0.6180087221242067,
      "learning_rate": 9.68222951188366e-06,
      "loss": 0.3341755628585815,
      "num_input_tokens_seen": 107531872,
      "step": 1790,
      "train_runtime": 23000.7952,
      "train_tokens_per_second": 4675.137
    },
    {
      "epoch": 0.22895408163265307,
      "grad_norm": 0.5223681605772506,
      "learning_rate": 9.680469974474685e-06,
      "loss": 0.3221714973449707,
      "num_input_tokens_seen": 107841640,
      "step": 1795,
      "train_runtime": 23058.9012,
      "train_tokens_per_second": 4676.79
    },
    {
      "epoch": 0.22959183673469388,
      "grad_norm": 0.5957130183551872,
      "learning_rate": 9.67870573988721e-06,
      "loss": 0.3386493682861328,
      "num_input_tokens_seen": 108140352,
      "step": 1800,
      "train_runtime": 23124.1502,
      "train_tokens_per_second": 4676.511
    },
    {
      "epoch": 0.2302295918367347,
      "grad_norm": 0.6059549249466671,
      "learning_rate": 9.676936809891763e-06,
      "loss": 0.3442098617553711,
      "num_input_tokens_seen": 108453968,
      "step": 1805,
      "train_runtime": 23190.2532,
      "train_tokens_per_second": 4676.705
    },
    {
      "epoch": 0.2308673469387755,
      "grad_norm": 0.6213669171971626,
      "learning_rate": 9.675163186263591e-06,
      "loss": 0.31106042861938477,
      "num_input_tokens_seen": 108741728,
      "step": 1810,
      "train_runtime": 23260.6327,
      "train_tokens_per_second": 4674.926
    },
    {
      "epoch": 0.2315051020408163,
      "grad_norm": 0.6288194323670704,
      "learning_rate": 9.673384870782649e-06,
      "loss": 0.329655385017395,
      "num_input_tokens_seen": 109034568,
      "step": 1815,
      "train_runtime": 23319.9422,
      "train_tokens_per_second": 4675.593
    },
    {
      "epoch": 0.23214285714285715,
      "grad_norm": 0.6339100799532578,
      "learning_rate": 9.6716018652336e-06,
      "loss": 0.3513630390167236,
      "num_input_tokens_seen": 109336904,
      "step": 1820,
      "train_runtime": 23391.3656,
      "train_tokens_per_second": 4674.242
    },
    {
      "epoch": 0.23278061224489796,
      "grad_norm": 0.59353528931617,
      "learning_rate": 9.669814171405818e-06,
      "loss": 0.3033082723617554,
      "num_input_tokens_seen": 109642616,
      "step": 1825,
      "train_runtime": 23445.8954,
      "train_tokens_per_second": 4676.41
    },
    {
      "epoch": 0.23341836734693877,
      "grad_norm": 0.6255898925485908,
      "learning_rate": 9.668021791093372e-06,
      "loss": 0.3282336235046387,
      "num_input_tokens_seen": 109949272,
      "step": 1830,
      "train_runtime": 23501.9415,
      "train_tokens_per_second": 4678.306
    },
    {
      "epoch": 0.23405612244897958,
      "grad_norm": 0.7250785340279433,
      "learning_rate": 9.666224726095048e-06,
      "loss": 0.3378988981246948,
      "num_input_tokens_seen": 110248184,
      "step": 1835,
      "train_runtime": 23566.4014,
      "train_tokens_per_second": 4678.193
    },
    {
      "epoch": 0.23469387755102042,
      "grad_norm": 0.6944942590307985,
      "learning_rate": 9.66442297821432e-06,
      "loss": 0.3463714122772217,
      "num_input_tokens_seen": 110544184,
      "step": 1840,
      "train_runtime": 23623.7931,
      "train_tokens_per_second": 4679.358
    },
    {
      "epoch": 0.23533163265306123,
      "grad_norm": 0.6219284149204517,
      "learning_rate": 9.662616549259374e-06,
      "loss": 0.3074399471282959,
      "num_input_tokens_seen": 110843984,
      "step": 1845,
      "train_runtime": 23694.1152,
      "train_tokens_per_second": 4678.123
    },
    {
      "epoch": 0.23596938775510204,
      "grad_norm": 0.558501639276207,
      "learning_rate": 9.660805441043082e-06,
      "loss": 0.30753483772277834,
      "num_input_tokens_seen": 111144448,
      "step": 1850,
      "train_runtime": 23750.718,
      "train_tokens_per_second": 4679.625
    },
    {
      "epoch": 0.23660714285714285,
      "grad_norm": 1.0253823706929464,
      "learning_rate": 9.65898965538302e-06,
      "loss": 0.34722580909729006,
      "num_input_tokens_seen": 111440120,
      "step": 1855,
      "train_runtime": 23816.6419,
      "train_tokens_per_second": 4679.086
    },
    {
      "epoch": 0.2372448979591837,
      "grad_norm": 0.6442962618904584,
      "learning_rate": 9.657169194101457e-06,
      "loss": 0.3536875009536743,
      "num_input_tokens_seen": 111738960,
      "step": 1860,
      "train_runtime": 23892.2735,
      "train_tokens_per_second": 4676.782
    },
    {
      "epoch": 0.2378826530612245,
      "grad_norm": 0.5836927414515857,
      "learning_rate": 9.655344059025351e-06,
      "loss": 0.32279133796691895,
      "num_input_tokens_seen": 112021544,
      "step": 1865,
      "train_runtime": 23955.8069,
      "train_tokens_per_second": 4676.175
    },
    {
      "epoch": 0.2385204081632653,
      "grad_norm": 0.6952663262351617,
      "learning_rate": 9.653514251986354e-06,
      "loss": 0.3028008460998535,
      "num_input_tokens_seen": 112318408,
      "step": 1870,
      "train_runtime": 24014.3599,
      "train_tokens_per_second": 4677.135
    },
    {
      "epoch": 0.23915816326530612,
      "grad_norm": 0.527700252310987,
      "learning_rate": 9.651679774820804e-06,
      "loss": 0.31937668323516843,
      "num_input_tokens_seen": 112617648,
      "step": 1875,
      "train_runtime": 24078.3066,
      "train_tokens_per_second": 4677.142
    },
    {
      "epoch": 0.23979591836734693,
      "grad_norm": 0.562703611670736,
      "learning_rate": 9.649840629369728e-06,
      "loss": 0.3241946935653687,
      "num_input_tokens_seen": 112930624,
      "step": 1880,
      "train_runtime": 24142.8503,
      "train_tokens_per_second": 4677.601
    },
    {
      "epoch": 0.24043367346938777,
      "grad_norm": 0.623722641429278,
      "learning_rate": 9.647996817478835e-06,
      "loss": 0.344280481338501,
      "num_input_tokens_seen": 113217640,
      "step": 1885,
      "train_runtime": 24200.6276,
      "train_tokens_per_second": 4678.294
    },
    {
      "epoch": 0.24107142857142858,
      "grad_norm": 0.5647700791617076,
      "learning_rate": 9.646148340998521e-06,
      "loss": 0.3104014158248901,
      "num_input_tokens_seen": 113529480,
      "step": 1890,
      "train_runtime": 24257.7697,
      "train_tokens_per_second": 4680.129
    },
    {
      "epoch": 0.24170918367346939,
      "grad_norm": 0.5667491923917192,
      "learning_rate": 9.64429520178386e-06,
      "loss": 0.31038522720336914,
      "num_input_tokens_seen": 113819120,
      "step": 1895,
      "train_runtime": 24321.9091,
      "train_tokens_per_second": 4679.695
    },
    {
      "epoch": 0.2423469387755102,
      "grad_norm": 0.6427769837002572,
      "learning_rate": 9.642437401694608e-06,
      "loss": 0.3571403980255127,
      "num_input_tokens_seen": 114132296,
      "step": 1900,
      "train_runtime": 24392.6403,
      "train_tokens_per_second": 4678.964
    },
    {
      "epoch": 0.24298469387755103,
      "grad_norm": 0.6209772677893988,
      "learning_rate": 9.640574942595195e-06,
      "loss": 0.32203919887542726,
      "num_input_tokens_seen": 114447456,
      "step": 1905,
      "train_runtime": 24458.6073,
      "train_tokens_per_second": 4679.23
    },
    {
      "epoch": 0.24362244897959184,
      "grad_norm": 0.5637953152973806,
      "learning_rate": 9.638707826354732e-06,
      "loss": 0.31209521293640136,
      "num_input_tokens_seen": 114753352,
      "step": 1910,
      "train_runtime": 24524.4172,
      "train_tokens_per_second": 4679.147
    },
    {
      "epoch": 0.24426020408163265,
      "grad_norm": 0.5823798976109281,
      "learning_rate": 9.636836054846997e-06,
      "loss": 0.31901235580444337,
      "num_input_tokens_seen": 115050112,
      "step": 1915,
      "train_runtime": 24597.96,
      "train_tokens_per_second": 4677.222
    },
    {
      "epoch": 0.24489795918367346,
      "grad_norm": 0.6719305469028826,
      "learning_rate": 9.634959629950444e-06,
      "loss": 0.29732065200805663,
      "num_input_tokens_seen": 115350832,
      "step": 1920,
      "train_runtime": 24655.5555,
      "train_tokens_per_second": 4678.493
    },
    {
      "epoch": 0.24553571428571427,
      "grad_norm": 0.5579926470705943,
      "learning_rate": 9.6330785535482e-06,
      "loss": 0.3192960262298584,
      "num_input_tokens_seen": 115655024,
      "step": 1925,
      "train_runtime": 24715.6406,
      "train_tokens_per_second": 4679.427
    },
    {
      "epoch": 0.2461734693877551,
      "grad_norm": 0.6004158154783408,
      "learning_rate": 9.631192827528054e-06,
      "loss": 0.3167712688446045,
      "num_input_tokens_seen": 115940248,
      "step": 1930,
      "train_runtime": 24778.0303,
      "train_tokens_per_second": 4679.155
    },
    {
      "epoch": 0.24681122448979592,
      "grad_norm": 0.7097341672547058,
      "learning_rate": 9.629302453782462e-06,
      "loss": 0.34239945411682127,
      "num_input_tokens_seen": 116243072,
      "step": 1935,
      "train_runtime": 24829.7494,
      "train_tokens_per_second": 4681.605
    },
    {
      "epoch": 0.24744897959183673,
      "grad_norm": 0.6007324013896582,
      "learning_rate": 9.62740743420855e-06,
      "loss": 0.3372066020965576,
      "num_input_tokens_seen": 116557320,
      "step": 1940,
      "train_runtime": 24896.0702,
      "train_tokens_per_second": 4681.756
    },
    {
      "epoch": 0.24808673469387754,
      "grad_norm": 0.6010879011596942,
      "learning_rate": 9.625507770708098e-06,
      "loss": 0.33537166118621825,
      "num_input_tokens_seen": 116850296,
      "step": 1945,
      "train_runtime": 24956.4545,
      "train_tokens_per_second": 4682.167
    },
    {
      "epoch": 0.24872448979591838,
      "grad_norm": 0.5642263779586972,
      "learning_rate": 9.623603465187554e-06,
      "loss": 0.3403872013092041,
      "num_input_tokens_seen": 117165232,
      "step": 1950,
      "train_runtime": 25016.4062,
      "train_tokens_per_second": 4683.536
    },
    {
      "epoch": 0.2493622448979592,
      "grad_norm": 0.646102123535778,
      "learning_rate": 9.621694519558021e-06,
      "loss": 0.3353994369506836,
      "num_input_tokens_seen": 117469912,
      "step": 1955,
      "train_runtime": 25089.8824,
      "train_tokens_per_second": 4681.963
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7292598852552631,
      "learning_rate": 9.619780935735258e-06,
      "loss": 0.3624607563018799,
      "num_input_tokens_seen": 117767632,
      "step": 1960,
      "train_runtime": 25158.8153,
      "train_tokens_per_second": 4680.969
    },
    {
      "epoch": 0.25063775510204084,
      "grad_norm": 0.5308857230100679,
      "learning_rate": 9.617862715639685e-06,
      "loss": 0.2849001407623291,
      "num_input_tokens_seen": 118078936,
      "step": 1965,
      "train_runtime": 25227.7331,
      "train_tokens_per_second": 4680.521
    },
    {
      "epoch": 0.2512755102040816,
      "grad_norm": 0.6435926803053078,
      "learning_rate": 9.615939861196364e-06,
      "loss": 0.3471491813659668,
      "num_input_tokens_seen": 118373224,
      "step": 1970,
      "train_runtime": 25301.05,
      "train_tokens_per_second": 4678.589
    },
    {
      "epoch": 0.25191326530612246,
      "grad_norm": 0.6160911738361701,
      "learning_rate": 9.614012374335014e-06,
      "loss": 0.3345517158508301,
      "num_input_tokens_seen": 118665768,
      "step": 1975,
      "train_runtime": 25374.3773,
      "train_tokens_per_second": 4676.598
    },
    {
      "epoch": 0.25255102040816324,
      "grad_norm": 0.6078198965546298,
      "learning_rate": 9.612080256990006e-06,
      "loss": 0.3112973213195801,
      "num_input_tokens_seen": 118963984,
      "step": 1980,
      "train_runtime": 25436.7886,
      "train_tokens_per_second": 4676.848
    },
    {
      "epoch": 0.2531887755102041,
      "grad_norm": 0.5884732352967934,
      "learning_rate": 9.610143511100354e-06,
      "loss": 0.30260891914367677,
      "num_input_tokens_seen": 119243072,
      "step": 1985,
      "train_runtime": 25507.8858,
      "train_tokens_per_second": 4674.753
    },
    {
      "epoch": 0.2538265306122449,
      "grad_norm": 0.5867017741557491,
      "learning_rate": 9.608202138609714e-06,
      "loss": 0.31354618072509766,
      "num_input_tokens_seen": 119547376,
      "step": 1990,
      "train_runtime": 25579.3014,
      "train_tokens_per_second": 4673.598
    },
    {
      "epoch": 0.2544642857142857,
      "grad_norm": 0.5930499940162742,
      "learning_rate": 9.606256141466394e-06,
      "loss": 0.3089960813522339,
      "num_input_tokens_seen": 119848944,
      "step": 1995,
      "train_runtime": 25641.2909,
      "train_tokens_per_second": 4674.06
    },
    {
      "epoch": 0.25510204081632654,
      "grad_norm": 0.6399697120868858,
      "learning_rate": 9.604305521623335e-06,
      "loss": 0.29999723434448244,
      "num_input_tokens_seen": 120155520,
      "step": 2000,
      "train_runtime": 25709.5198,
      "train_tokens_per_second": 4673.581
    },
    {
      "epoch": 0.2557397959183674,
      "grad_norm": 0.5492487573374648,
      "learning_rate": 9.602350281038123e-06,
      "loss": 0.3378730297088623,
      "num_input_tokens_seen": 120461080,
      "step": 2005,
      "train_runtime": 25772.7865,
      "train_tokens_per_second": 4673.964
    },
    {
      "epoch": 0.25637755102040816,
      "grad_norm": 0.704438904242841,
      "learning_rate": 9.600390421672976e-06,
      "loss": 0.34598360061645506,
      "num_input_tokens_seen": 120752152,
      "step": 2010,
      "train_runtime": 25835.8802,
      "train_tokens_per_second": 4673.816
    },
    {
      "epoch": 0.257015306122449,
      "grad_norm": 0.5801593494794286,
      "learning_rate": 9.59842594549475e-06,
      "loss": 0.3566749572753906,
      "num_input_tokens_seen": 121057624,
      "step": 2015,
      "train_runtime": 25906.0535,
      "train_tokens_per_second": 4672.947
    },
    {
      "epoch": 0.2576530612244898,
      "grad_norm": 0.6376080616812323,
      "learning_rate": 9.596456854474933e-06,
      "loss": 0.3009960174560547,
      "num_input_tokens_seen": 121356392,
      "step": 2020,
      "train_runtime": 25976.6194,
      "train_tokens_per_second": 4671.755
    },
    {
      "epoch": 0.2582908163265306,
      "grad_norm": 0.5979766863747837,
      "learning_rate": 9.594483150589647e-06,
      "loss": 0.32624316215515137,
      "num_input_tokens_seen": 121656992,
      "step": 2025,
      "train_runtime": 26028.5388,
      "train_tokens_per_second": 4673.985
    },
    {
      "epoch": 0.25892857142857145,
      "grad_norm": 0.6312514232256536,
      "learning_rate": 9.592504835819642e-06,
      "loss": 0.33791747093200686,
      "num_input_tokens_seen": 121958920,
      "step": 2030,
      "train_runtime": 26087.6457,
      "train_tokens_per_second": 4674.968
    },
    {
      "epoch": 0.25956632653061223,
      "grad_norm": 0.6681405212622115,
      "learning_rate": 9.590521912150294e-06,
      "loss": 0.32710092067718505,
      "num_input_tokens_seen": 122265488,
      "step": 2035,
      "train_runtime": 26139.3938,
      "train_tokens_per_second": 4677.442
    },
    {
      "epoch": 0.2602040816326531,
      "grad_norm": 0.7037855571998192,
      "learning_rate": 9.588534381571606e-06,
      "loss": 0.31521337032318114,
      "num_input_tokens_seen": 122573968,
      "step": 2040,
      "train_runtime": 26194.6883,
      "train_tokens_per_second": 4679.344
    },
    {
      "epoch": 0.26084183673469385,
      "grad_norm": 0.5783322471293008,
      "learning_rate": 9.586542246078203e-06,
      "loss": 0.330059552192688,
      "num_input_tokens_seen": 122862792,
      "step": 2045,
      "train_runtime": 26263.2862,
      "train_tokens_per_second": 4678.12
    },
    {
      "epoch": 0.2614795918367347,
      "grad_norm": 0.6150497473375843,
      "learning_rate": 9.584545507669333e-06,
      "loss": 0.30462510585784913,
      "num_input_tokens_seen": 123159968,
      "step": 2050,
      "train_runtime": 26334.5848,
      "train_tokens_per_second": 4676.739
    },
    {
      "epoch": 0.26211734693877553,
      "grad_norm": 0.59872654736947,
      "learning_rate": 9.582544168348864e-06,
      "loss": 0.31244144439697263,
      "num_input_tokens_seen": 123452512,
      "step": 2055,
      "train_runtime": 26402.5919,
      "train_tokens_per_second": 4675.772
    },
    {
      "epoch": 0.2627551020408163,
      "grad_norm": 0.6211287049459203,
      "learning_rate": 9.580538230125275e-06,
      "loss": 0.344005823135376,
      "num_input_tokens_seen": 123734472,
      "step": 2060,
      "train_runtime": 26468.247,
      "train_tokens_per_second": 4674.827
    },
    {
      "epoch": 0.26339285714285715,
      "grad_norm": 0.6481823971187728,
      "learning_rate": 9.57852769501167e-06,
      "loss": 0.3296398639678955,
      "num_input_tokens_seen": 124037960,
      "step": 2065,
      "train_runtime": 26530.3794,
      "train_tokens_per_second": 4675.318
    },
    {
      "epoch": 0.26403061224489793,
      "grad_norm": 0.5941205675123112,
      "learning_rate": 9.576512565025759e-06,
      "loss": 0.31540956497192385,
      "num_input_tokens_seen": 124342752,
      "step": 2070,
      "train_runtime": 26592.1225,
      "train_tokens_per_second": 4675.924
    },
    {
      "epoch": 0.26466836734693877,
      "grad_norm": 0.6326195230505037,
      "learning_rate": 9.574492842189865e-06,
      "loss": 0.34804620742797854,
      "num_input_tokens_seen": 124662472,
      "step": 2075,
      "train_runtime": 26647.0273,
      "train_tokens_per_second": 4678.288
    },
    {
      "epoch": 0.2653061224489796,
      "grad_norm": 0.5693984902338113,
      "learning_rate": 9.572468528530923e-06,
      "loss": 0.29468479156494143,
      "num_input_tokens_seen": 124958736,
      "step": 2080,
      "train_runtime": 26710.9839,
      "train_tokens_per_second": 4678.178
    },
    {
      "epoch": 0.2659438775510204,
      "grad_norm": 0.6512563017816041,
      "learning_rate": 9.570439626080472e-06,
      "loss": 0.3527271509170532,
      "num_input_tokens_seen": 125241776,
      "step": 2085,
      "train_runtime": 26772.1002,
      "train_tokens_per_second": 4678.071
    },
    {
      "epoch": 0.26658163265306123,
      "grad_norm": 0.6206313058829079,
      "learning_rate": 9.568406136874657e-06,
      "loss": 0.3105809450149536,
      "num_input_tokens_seen": 125535920,
      "step": 2090,
      "train_runtime": 26835.7399,
      "train_tokens_per_second": 4677.938
    },
    {
      "epoch": 0.26721938775510207,
      "grad_norm": 0.5644830315991491,
      "learning_rate": 9.566368062954227e-06,
      "loss": 0.3023581027984619,
      "num_input_tokens_seen": 125827480,
      "step": 2095,
      "train_runtime": 26894.1398,
      "train_tokens_per_second": 4678.621
    },
    {
      "epoch": 0.26785714285714285,
      "grad_norm": 0.6662881978380841,
      "learning_rate": 9.564325406364532e-06,
      "loss": 0.33686540126800535,
      "num_input_tokens_seen": 126148872,
      "step": 2100,
      "train_runtime": 26946.1488,
      "train_tokens_per_second": 4681.518
    },
    {
      "epoch": 0.2684948979591837,
      "grad_norm": 0.616541008841765,
      "learning_rate": 9.562278169155518e-06,
      "loss": 0.3285606622695923,
      "num_input_tokens_seen": 126463968,
      "step": 2105,
      "train_runtime": 27001.0604,
      "train_tokens_per_second": 4683.667
    },
    {
      "epoch": 0.26913265306122447,
      "grad_norm": 0.5501720018606887,
      "learning_rate": 9.560226353381735e-06,
      "loss": 0.3161233186721802,
      "num_input_tokens_seen": 126769104,
      "step": 2110,
      "train_runtime": 27066.9531,
      "train_tokens_per_second": 4683.538
    },
    {
      "epoch": 0.2697704081632653,
      "grad_norm": 0.6098186680772301,
      "learning_rate": 9.55816996110232e-06,
      "loss": 0.33042201995849607,
      "num_input_tokens_seen": 127070176,
      "step": 2115,
      "train_runtime": 27121.159,
      "train_tokens_per_second": 4685.278
    },
    {
      "epoch": 0.27040816326530615,
      "grad_norm": 0.6562972170179081,
      "learning_rate": 9.556108994381007e-06,
      "loss": 0.33913230895996094,
      "num_input_tokens_seen": 127378504,
      "step": 2120,
      "train_runtime": 27184.3758,
      "train_tokens_per_second": 4685.725
    },
    {
      "epoch": 0.2710459183673469,
      "grad_norm": 0.5696608649074028,
      "learning_rate": 9.55404345528612e-06,
      "loss": 0.32880458831787107,
      "num_input_tokens_seen": 127693280,
      "step": 2125,
      "train_runtime": 27238.5903,
      "train_tokens_per_second": 4687.955
    },
    {
      "epoch": 0.27168367346938777,
      "grad_norm": 0.5776721935255191,
      "learning_rate": 9.551973345890572e-06,
      "loss": 0.3231329441070557,
      "num_input_tokens_seen": 128000488,
      "step": 2130,
      "train_runtime": 27293.669,
      "train_tokens_per_second": 4689.75
    },
    {
      "epoch": 0.27232142857142855,
      "grad_norm": 0.6742303614240489,
      "learning_rate": 9.549898668271862e-06,
      "loss": 0.32097625732421875,
      "num_input_tokens_seen": 128303912,
      "step": 2135,
      "train_runtime": 27344.549,
      "train_tokens_per_second": 4692.12
    },
    {
      "epoch": 0.2729591836734694,
      "grad_norm": 0.5630324691347439,
      "learning_rate": 9.547819424512075e-06,
      "loss": 0.333871865272522,
      "num_input_tokens_seen": 128597200,
      "step": 2140,
      "train_runtime": 27404.847,
      "train_tokens_per_second": 4692.498
    },
    {
      "epoch": 0.2735969387755102,
      "grad_norm": 0.6737715351955407,
      "learning_rate": 9.545735616697875e-06,
      "loss": 0.3379852294921875,
      "num_input_tokens_seen": 128907856,
      "step": 2145,
      "train_runtime": 27474.3731,
      "train_tokens_per_second": 4691.931
    },
    {
      "epoch": 0.274234693877551,
      "grad_norm": 0.633589763494367,
      "learning_rate": 9.54364724692051e-06,
      "loss": 0.34932861328125,
      "num_input_tokens_seen": 129221728,
      "step": 2150,
      "train_runtime": 27551.387,
      "train_tokens_per_second": 4690.208
    },
    {
      "epoch": 0.27487244897959184,
      "grad_norm": 0.5976884715434537,
      "learning_rate": 9.541554317275807e-06,
      "loss": 0.30597634315490724,
      "num_input_tokens_seen": 129523728,
      "step": 2155,
      "train_runtime": 27605.7458,
      "train_tokens_per_second": 4691.912
    },
    {
      "epoch": 0.2755102040816326,
      "grad_norm": 0.5936498054938893,
      "learning_rate": 9.539456829864164e-06,
      "loss": 0.2829718112945557,
      "num_input_tokens_seen": 129812504,
      "step": 2160,
      "train_runtime": 27680.8759,
      "train_tokens_per_second": 4689.61
    },
    {
      "epoch": 0.27614795918367346,
      "grad_norm": 0.6135433871825898,
      "learning_rate": 9.537354786790558e-06,
      "loss": 0.335500693321228,
      "num_input_tokens_seen": 130124616,
      "step": 2165,
      "train_runtime": 27750.1509,
      "train_tokens_per_second": 4689.15
    },
    {
      "epoch": 0.2767857142857143,
      "grad_norm": 0.6283514694093978,
      "learning_rate": 9.535248190164537e-06,
      "loss": 0.34310569763183596,
      "num_input_tokens_seen": 130420944,
      "step": 2170,
      "train_runtime": 27816.7927,
      "train_tokens_per_second": 4688.569
    },
    {
      "epoch": 0.2774234693877551,
      "grad_norm": 0.5532100827303795,
      "learning_rate": 9.533137042100214e-06,
      "loss": 0.3102876663208008,
      "num_input_tokens_seen": 130705888,
      "step": 2175,
      "train_runtime": 27874.0843,
      "train_tokens_per_second": 4689.155
    },
    {
      "epoch": 0.2780612244897959,
      "grad_norm": 0.551989879026994,
      "learning_rate": 9.531021344716275e-06,
      "loss": 0.30455660820007324,
      "num_input_tokens_seen": 131007712,
      "step": 2180,
      "train_runtime": 27941.7726,
      "train_tokens_per_second": 4688.597
    },
    {
      "epoch": 0.27869897959183676,
      "grad_norm": 0.604405208631062,
      "learning_rate": 9.528901100135971e-06,
      "loss": 0.3334838390350342,
      "num_input_tokens_seen": 131301160,
      "step": 2185,
      "train_runtime": 28007.0995,
      "train_tokens_per_second": 4688.138
    },
    {
      "epoch": 0.27933673469387754,
      "grad_norm": 0.6372287038655521,
      "learning_rate": 9.526776310487116e-06,
      "loss": 0.30718910694122314,
      "num_input_tokens_seen": 131601576,
      "step": 2190,
      "train_runtime": 28074.0945,
      "train_tokens_per_second": 4687.652
    },
    {
      "epoch": 0.2799744897959184,
      "grad_norm": 0.6095463320329509,
      "learning_rate": 9.524646977902083e-06,
      "loss": 0.3460432291030884,
      "num_input_tokens_seen": 131907152,
      "step": 2195,
      "train_runtime": 28135.9899,
      "train_tokens_per_second": 4688.2
    },
    {
      "epoch": 0.28061224489795916,
      "grad_norm": 0.5801013513430481,
      "learning_rate": 9.522513104517807e-06,
      "loss": 0.3240336894989014,
      "num_input_tokens_seen": 132205376,
      "step": 2200,
      "train_runtime": 28200.4482,
      "train_tokens_per_second": 4688.059
    },
    {
      "epoch": 0.28125,
      "grad_norm": 0.7268316975488225,
      "learning_rate": 9.52037469247578e-06,
      "loss": 0.3053292274475098,
      "num_input_tokens_seen": 132499760,
      "step": 2205,
      "train_runtime": 28257.3095,
      "train_tokens_per_second": 4689.044
    },
    {
      "epoch": 0.28188775510204084,
      "grad_norm": 0.5689089557303342,
      "learning_rate": 9.518231743922049e-06,
      "loss": 0.33064699172973633,
      "num_input_tokens_seen": 132799952,
      "step": 2210,
      "train_runtime": 28331.5474,
      "train_tokens_per_second": 4687.353
    },
    {
      "epoch": 0.2825255102040816,
      "grad_norm": 0.6724396550763708,
      "learning_rate": 9.516084261007207e-06,
      "loss": 0.32482156753540037,
      "num_input_tokens_seen": 133086712,
      "step": 2215,
      "train_runtime": 28405.2893,
      "train_tokens_per_second": 4685.279
    },
    {
      "epoch": 0.28316326530612246,
      "grad_norm": 0.6347967696237345,
      "learning_rate": 9.513932245886409e-06,
      "loss": 0.29505276679992676,
      "num_input_tokens_seen": 133376464,
      "step": 2220,
      "train_runtime": 28467.0786,
      "train_tokens_per_second": 4685.288
    },
    {
      "epoch": 0.28380102040816324,
      "grad_norm": 0.5464938949973542,
      "learning_rate": 9.511775700719347e-06,
      "loss": 0.32130861282348633,
      "num_input_tokens_seen": 133678184,
      "step": 2225,
      "train_runtime": 28536.8175,
      "train_tokens_per_second": 4684.411
    },
    {
      "epoch": 0.2844387755102041,
      "grad_norm": 0.579213261484982,
      "learning_rate": 9.50961462767027e-06,
      "loss": 0.3393184900283813,
      "num_input_tokens_seen": 133974496,
      "step": 2230,
      "train_runtime": 28611.7797,
      "train_tokens_per_second": 4682.494
    },
    {
      "epoch": 0.2850765306122449,
      "grad_norm": 0.6172083632586678,
      "learning_rate": 9.507449028907961e-06,
      "loss": 0.327873158454895,
      "num_input_tokens_seen": 134264256,
      "step": 2235,
      "train_runtime": 28680.0516,
      "train_tokens_per_second": 4681.451
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.6569159177539318,
      "learning_rate": 9.505278906605752e-06,
      "loss": 0.32256019115448,
      "num_input_tokens_seen": 134555712,
      "step": 2240,
      "train_runtime": 28741.5535,
      "train_tokens_per_second": 4681.574
    },
    {
      "epoch": 0.28635204081632654,
      "grad_norm": 0.6489904409400776,
      "learning_rate": 9.503104262941515e-06,
      "loss": 0.3126920938491821,
      "num_input_tokens_seen": 134843824,
      "step": 2245,
      "train_runtime": 28809.5578,
      "train_tokens_per_second": 4680.524
    },
    {
      "epoch": 0.2869897959183674,
      "grad_norm": 0.5889950079552234,
      "learning_rate": 9.500925100097652e-06,
      "loss": 0.3146326541900635,
      "num_input_tokens_seen": 135143624,
      "step": 2250,
      "train_runtime": 28885.544,
      "train_tokens_per_second": 4678.59
    },
    {
      "epoch": 0.28762755102040816,
      "grad_norm": 0.5901022593560107,
      "learning_rate": 9.498741420261109e-06,
      "loss": 0.3119138240814209,
      "num_input_tokens_seen": 135438736,
      "step": 2255,
      "train_runtime": 28945.8891,
      "train_tokens_per_second": 4679.032
    },
    {
      "epoch": 0.288265306122449,
      "grad_norm": 0.6679007996119871,
      "learning_rate": 9.496553225623358e-06,
      "loss": 0.3327902317047119,
      "num_input_tokens_seen": 135721752,
      "step": 2260,
      "train_runtime": 29006.2014,
      "train_tokens_per_second": 4679.06
    },
    {
      "epoch": 0.2889030612244898,
      "grad_norm": 0.6418452396909745,
      "learning_rate": 9.494360518380405e-06,
      "loss": 0.30309200286865234,
      "num_input_tokens_seen": 136006624,
      "step": 2265,
      "train_runtime": 29078.9405,
      "train_tokens_per_second": 4677.152
    },
    {
      "epoch": 0.2895408163265306,
      "grad_norm": 0.6740226014371311,
      "learning_rate": 9.49216330073279e-06,
      "loss": 0.3591219663619995,
      "num_input_tokens_seen": 136304952,
      "step": 2270,
      "train_runtime": 29136.3858,
      "train_tokens_per_second": 4678.17
    },
    {
      "epoch": 0.29017857142857145,
      "grad_norm": 0.6190880149541824,
      "learning_rate": 9.489961574885567e-06,
      "loss": 0.30411219596862793,
      "num_input_tokens_seen": 136585840,
      "step": 2275,
      "train_runtime": 29200.6372,
      "train_tokens_per_second": 4677.495
    },
    {
      "epoch": 0.29081632653061223,
      "grad_norm": 0.5512260948689923,
      "learning_rate": 9.487755343048328e-06,
      "loss": 0.32656924724578856,
      "num_input_tokens_seen": 136885088,
      "step": 2280,
      "train_runtime": 29269.8852,
      "train_tokens_per_second": 4676.653
    },
    {
      "epoch": 0.2914540816326531,
      "grad_norm": 0.6453574262630075,
      "learning_rate": 9.485544607435177e-06,
      "loss": 0.3365833044052124,
      "num_input_tokens_seen": 137186696,
      "step": 2285,
      "train_runtime": 29332.5357,
      "train_tokens_per_second": 4676.946
    },
    {
      "epoch": 0.29209183673469385,
      "grad_norm": 0.5523789076458793,
      "learning_rate": 9.483329370264743e-06,
      "loss": 0.3248692750930786,
      "num_input_tokens_seen": 137492544,
      "step": 2290,
      "train_runtime": 29396.2427,
      "train_tokens_per_second": 4677.215
    },
    {
      "epoch": 0.2927295918367347,
      "grad_norm": 0.596344921349206,
      "learning_rate": 9.48110963376017e-06,
      "loss": 0.3270201444625854,
      "num_input_tokens_seen": 137782080,
      "step": 2295,
      "train_runtime": 29468.9381,
      "train_tokens_per_second": 4675.502
    },
    {
      "epoch": 0.29336734693877553,
      "grad_norm": 0.6736132108784195,
      "learning_rate": 9.47888540014912e-06,
      "loss": 0.3187110900878906,
      "num_input_tokens_seen": 138078120,
      "step": 2300,
      "train_runtime": 29538.7803,
      "train_tokens_per_second": 4674.469
    },
    {
      "epoch": 0.2940051020408163,
      "grad_norm": 0.532669593986322,
      "learning_rate": 9.476656671663766e-06,
      "loss": 0.30268163681030275,
      "num_input_tokens_seen": 138377568,
      "step": 2305,
      "train_runtime": 29604.2275,
      "train_tokens_per_second": 4674.25
    },
    {
      "epoch": 0.29464285714285715,
      "grad_norm": 0.6510441114111867,
      "learning_rate": 9.474423450540794e-06,
      "loss": 0.33988251686096194,
      "num_input_tokens_seen": 138671160,
      "step": 2310,
      "train_runtime": 29670.8385,
      "train_tokens_per_second": 4673.652
    },
    {
      "epoch": 0.29528061224489793,
      "grad_norm": 0.6234702256034333,
      "learning_rate": 9.472185739021396e-06,
      "loss": 0.31470139026641847,
      "num_input_tokens_seen": 138968528,
      "step": 2315,
      "train_runtime": 29738.0543,
      "train_tokens_per_second": 4673.087
    },
    {
      "epoch": 0.29591836734693877,
      "grad_norm": 0.5751340719081197,
      "learning_rate": 9.469943539351272e-06,
      "loss": 0.31812400817871095,
      "num_input_tokens_seen": 139276064,
      "step": 2320,
      "train_runtime": 29805.6272,
      "train_tokens_per_second": 4672.811
    },
    {
      "epoch": 0.2965561224489796,
      "grad_norm": 0.5786021273078671,
      "learning_rate": 9.467696853780625e-06,
      "loss": 0.33053481578826904,
      "num_input_tokens_seen": 139576128,
      "step": 2325,
      "train_runtime": 29860.1304,
      "train_tokens_per_second": 4674.331
    },
    {
      "epoch": 0.2971938775510204,
      "grad_norm": 0.5986777323470593,
      "learning_rate": 9.465445684564165e-06,
      "loss": 0.3223258495330811,
      "num_input_tokens_seen": 139880552,
      "step": 2330,
      "train_runtime": 29926.5424,
      "train_tokens_per_second": 4674.13
    },
    {
      "epoch": 0.29783163265306123,
      "grad_norm": 0.5571699592153114,
      "learning_rate": 9.463190033961092e-06,
      "loss": 0.34364633560180663,
      "num_input_tokens_seen": 140180800,
      "step": 2335,
      "train_runtime": 29985.9569,
      "train_tokens_per_second": 4674.882
    },
    {
      "epoch": 0.29846938775510207,
      "grad_norm": 0.5900640933697718,
      "learning_rate": 9.46092990423511e-06,
      "loss": 0.3437541723251343,
      "num_input_tokens_seen": 140482296,
      "step": 2340,
      "train_runtime": 30050.2785,
      "train_tokens_per_second": 4674.908
    },
    {
      "epoch": 0.29910714285714285,
      "grad_norm": 0.6193560109419097,
      "learning_rate": 9.45866529765442e-06,
      "loss": 0.34493107795715333,
      "num_input_tokens_seen": 140776768,
      "step": 2345,
      "train_runtime": 30119.3252,
      "train_tokens_per_second": 4673.968
    },
    {
      "epoch": 0.2997448979591837,
      "grad_norm": 0.6205275388754196,
      "learning_rate": 9.456396216491713e-06,
      "loss": 0.3377701759338379,
      "num_input_tokens_seen": 141076136,
      "step": 2350,
      "train_runtime": 30177.7701,
      "train_tokens_per_second": 4674.836
    },
    {
      "epoch": 0.30038265306122447,
      "grad_norm": 0.6397239800403964,
      "learning_rate": 9.454122663024167e-06,
      "loss": 0.31418709754943847,
      "num_input_tokens_seen": 141374512,
      "step": 2355,
      "train_runtime": 30240.8034,
      "train_tokens_per_second": 4674.959
    },
    {
      "epoch": 0.3010204081632653,
      "grad_norm": 0.5815300830224596,
      "learning_rate": 9.451844639533453e-06,
      "loss": 0.31237623691558836,
      "num_input_tokens_seen": 141660368,
      "step": 2360,
      "train_runtime": 30309.4086,
      "train_tokens_per_second": 4673.808
    },
    {
      "epoch": 0.30165816326530615,
      "grad_norm": 0.6687419180167138,
      "learning_rate": 9.44956214830573e-06,
      "loss": 0.3478510618209839,
      "num_input_tokens_seen": 141967256,
      "step": 2365,
      "train_runtime": 30368.9784,
      "train_tokens_per_second": 4674.746
    },
    {
      "epoch": 0.3022959183673469,
      "grad_norm": 0.5828907587240244,
      "learning_rate": 9.447275191631631e-06,
      "loss": 0.3015713691711426,
      "num_input_tokens_seen": 142263600,
      "step": 2370,
      "train_runtime": 30433.5815,
      "train_tokens_per_second": 4674.56
    },
    {
      "epoch": 0.30293367346938777,
      "grad_norm": 0.7619381501207457,
      "learning_rate": 9.444983771806281e-06,
      "loss": 0.31589689254760744,
      "num_input_tokens_seen": 142557944,
      "step": 2375,
      "train_runtime": 30484.0685,
      "train_tokens_per_second": 4676.474
    },
    {
      "epoch": 0.30357142857142855,
      "grad_norm": 0.7039368564710913,
      "learning_rate": 9.44268789112928e-06,
      "loss": 0.35042586326599123,
      "num_input_tokens_seen": 142858264,
      "step": 2380,
      "train_runtime": 30547.7524,
      "train_tokens_per_second": 4676.556
    },
    {
      "epoch": 0.3042091836734694,
      "grad_norm": 0.5527063373626678,
      "learning_rate": 9.440387551904705e-06,
      "loss": 0.2907209634780884,
      "num_input_tokens_seen": 143148344,
      "step": 2385,
      "train_runtime": 30609.9933,
      "train_tokens_per_second": 4676.523
    },
    {
      "epoch": 0.3048469387755102,
      "grad_norm": 0.5773185427370752,
      "learning_rate": 9.438082756441103e-06,
      "loss": 0.33542962074279786,
      "num_input_tokens_seen": 143466320,
      "step": 2390,
      "train_runtime": 30657.9794,
      "train_tokens_per_second": 4679.575
    },
    {
      "epoch": 0.305484693877551,
      "grad_norm": 0.6069011727187676,
      "learning_rate": 9.4357735070515e-06,
      "loss": 0.34381320476531985,
      "num_input_tokens_seen": 143768480,
      "step": 2395,
      "train_runtime": 30727.0062,
      "train_tokens_per_second": 4678.896
    },
    {
      "epoch": 0.30612244897959184,
      "grad_norm": 0.6224162118341987,
      "learning_rate": 9.433459806053392e-06,
      "loss": 0.30916447639465333,
      "num_input_tokens_seen": 144080664,
      "step": 2400,
      "train_runtime": 30796.7448,
      "train_tokens_per_second": 4678.438
    },
    {
      "epoch": 0.3067602040816326,
      "grad_norm": 0.6058702743286325,
      "learning_rate": 9.431141655768733e-06,
      "loss": 0.3193027496337891,
      "num_input_tokens_seen": 144385056,
      "step": 2405,
      "train_runtime": 30866.2766,
      "train_tokens_per_second": 4677.761
    },
    {
      "epoch": 0.30739795918367346,
      "grad_norm": 0.5938586145105029,
      "learning_rate": 9.428819058523953e-06,
      "loss": 0.31721904277801516,
      "num_input_tokens_seen": 144691264,
      "step": 2410,
      "train_runtime": 30923.7767,
      "train_tokens_per_second": 4678.965
    },
    {
      "epoch": 0.3080357142857143,
      "grad_norm": 0.5326567040194892,
      "learning_rate": 9.42649201664994e-06,
      "loss": 0.3103544473648071,
      "num_input_tokens_seen": 144999024,
      "step": 2415,
      "train_runtime": 30999.5501,
      "train_tokens_per_second": 4677.456
    },
    {
      "epoch": 0.3086734693877551,
      "grad_norm": 0.6147451958275432,
      "learning_rate": 9.424160532482042e-06,
      "loss": 0.33042163848876954,
      "num_input_tokens_seen": 145300312,
      "step": 2420,
      "train_runtime": 31065.25,
      "train_tokens_per_second": 4677.262
    },
    {
      "epoch": 0.3093112244897959,
      "grad_norm": 0.6699472678811865,
      "learning_rate": 9.421824608360068e-06,
      "loss": 0.31264162063598633,
      "num_input_tokens_seen": 145603704,
      "step": 2425,
      "train_runtime": 31131.0589,
      "train_tokens_per_second": 4677.12
    },
    {
      "epoch": 0.30994897959183676,
      "grad_norm": 0.6012355174720134,
      "learning_rate": 9.419484246628279e-06,
      "loss": 0.3024762153625488,
      "num_input_tokens_seen": 145885752,
      "step": 2430,
      "train_runtime": 31199.3458,
      "train_tokens_per_second": 4675.923
    },
    {
      "epoch": 0.31058673469387754,
      "grad_norm": 0.5707752988400112,
      "learning_rate": 9.417139449635394e-06,
      "loss": 0.3117316722869873,
      "num_input_tokens_seen": 146187696,
      "step": 2435,
      "train_runtime": 31264.451,
      "train_tokens_per_second": 4675.844
    },
    {
      "epoch": 0.3112244897959184,
      "grad_norm": 0.6416544677241595,
      "learning_rate": 9.414790219734578e-06,
      "loss": 0.3597858905792236,
      "num_input_tokens_seen": 146492336,
      "step": 2440,
      "train_runtime": 31317.289,
      "train_tokens_per_second": 4677.683
    },
    {
      "epoch": 0.31186224489795916,
      "grad_norm": 0.6166769162022172,
      "learning_rate": 9.41243655928345e-06,
      "loss": 0.30897738933563235,
      "num_input_tokens_seen": 146810920,
      "step": 2445,
      "train_runtime": 31384.647,
      "train_tokens_per_second": 4677.794
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.6108503616106563,
      "learning_rate": 9.410078470644069e-06,
      "loss": 0.30934574604034426,
      "num_input_tokens_seen": 147095600,
      "step": 2450,
      "train_runtime": 31453.8324,
      "train_tokens_per_second": 4676.556
    },
    {
      "epoch": 0.31313775510204084,
      "grad_norm": 0.4554145588657622,
      "learning_rate": 9.407715956182942e-06,
      "loss": 0.2786367893218994,
      "num_input_tokens_seen": 147406976,
      "step": 2455,
      "train_runtime": 31521.4732,
      "train_tokens_per_second": 4676.399
    },
    {
      "epoch": 0.3137755102040816,
      "grad_norm": 0.5642409078592039,
      "learning_rate": 9.405349018271021e-06,
      "loss": 0.3179809093475342,
      "num_input_tokens_seen": 147685440,
      "step": 2460,
      "train_runtime": 31586.8845,
      "train_tokens_per_second": 4675.53
    },
    {
      "epoch": 0.31441326530612246,
      "grad_norm": 0.5953585134819426,
      "learning_rate": 9.40297765928369e-06,
      "loss": 0.3162703990936279,
      "num_input_tokens_seen": 147986592,
      "step": 2465,
      "train_runtime": 31656.4537,
      "train_tokens_per_second": 4674.768
    },
    {
      "epoch": 0.31505102040816324,
      "grad_norm": 0.559696592077772,
      "learning_rate": 9.400601881600776e-06,
      "loss": 0.30867977142333985,
      "num_input_tokens_seen": 148295552,
      "step": 2470,
      "train_runtime": 31722.1295,
      "train_tokens_per_second": 4674.83
    },
    {
      "epoch": 0.3156887755102041,
      "grad_norm": 0.7691180619862527,
      "learning_rate": 9.398221687606535e-06,
      "loss": 0.3155848741531372,
      "num_input_tokens_seen": 148575152,
      "step": 2475,
      "train_runtime": 31785.5175,
      "train_tokens_per_second": 4674.303
    },
    {
      "epoch": 0.3163265306122449,
      "grad_norm": 0.6003783135059826,
      "learning_rate": 9.395837079689661e-06,
      "loss": 0.3411431789398193,
      "num_input_tokens_seen": 148877328,
      "step": 2480,
      "train_runtime": 31855.9629,
      "train_tokens_per_second": 4673.452
    },
    {
      "epoch": 0.3169642857142857,
      "grad_norm": 0.7310249378431494,
      "learning_rate": 9.393448060243272e-06,
      "loss": 0.30136809349060056,
      "num_input_tokens_seen": 149175304,
      "step": 2485,
      "train_runtime": 31924.3398,
      "train_tokens_per_second": 4672.776
    },
    {
      "epoch": 0.31760204081632654,
      "grad_norm": 0.609029292002024,
      "learning_rate": 9.391054631664918e-06,
      "loss": 0.28966469764709474,
      "num_input_tokens_seen": 149466488,
      "step": 2490,
      "train_runtime": 31992.955,
      "train_tokens_per_second": 4671.856
    },
    {
      "epoch": 0.3182397959183674,
      "grad_norm": 0.5693332037854462,
      "learning_rate": 9.388656796356569e-06,
      "loss": 0.31182107925415037,
      "num_input_tokens_seen": 149768536,
      "step": 2495,
      "train_runtime": 32067.2172,
      "train_tokens_per_second": 4670.456
    },
    {
      "epoch": 0.31887755102040816,
      "grad_norm": 0.5508709642797137,
      "learning_rate": 9.386254556724622e-06,
      "loss": 0.31195390224456787,
      "num_input_tokens_seen": 150078072,
      "step": 2500,
      "train_runtime": 32125.383,
      "train_tokens_per_second": 4671.635
    },
    {
      "epoch": 0.319515306122449,
      "grad_norm": 0.6332043098303607,
      "learning_rate": 9.383847915179892e-06,
      "loss": 0.3054932117462158,
      "num_input_tokens_seen": 150381440,
      "step": 2505,
      "train_runtime": 32187.5312,
      "train_tokens_per_second": 4672.04
    },
    {
      "epoch": 0.3201530612244898,
      "grad_norm": 0.697380145824633,
      "learning_rate": 9.381436874137613e-06,
      "loss": 0.35328240394592286,
      "num_input_tokens_seen": 150684960,
      "step": 2510,
      "train_runtime": 32247.0597,
      "train_tokens_per_second": 4672.828
    },
    {
      "epoch": 0.3207908163265306,
      "grad_norm": 0.8395252940112994,
      "learning_rate": 9.37902143601743e-06,
      "loss": 0.3169242858886719,
      "num_input_tokens_seen": 150986632,
      "step": 2515,
      "train_runtime": 32302.2906,
      "train_tokens_per_second": 4674.177
    },
    {
      "epoch": 0.32142857142857145,
      "grad_norm": 0.641319800087093,
      "learning_rate": 9.376601603243408e-06,
      "loss": 0.337641167640686,
      "num_input_tokens_seen": 151275848,
      "step": 2520,
      "train_runtime": 32355.6892,
      "train_tokens_per_second": 4675.402
    },
    {
      "epoch": 0.32206632653061223,
      "grad_norm": 0.8507791204923795,
      "learning_rate": 9.374177378244014e-06,
      "loss": 0.3014083385467529,
      "num_input_tokens_seen": 151568936,
      "step": 2525,
      "train_runtime": 32424.5142,
      "train_tokens_per_second": 4674.517
    },
    {
      "epoch": 0.3227040816326531,
      "grad_norm": 0.5841578663857016,
      "learning_rate": 9.37174876345213e-06,
      "loss": 0.34109816551208494,
      "num_input_tokens_seen": 151873808,
      "step": 2530,
      "train_runtime": 32481.7854,
      "train_tokens_per_second": 4675.661
    },
    {
      "epoch": 0.32334183673469385,
      "grad_norm": 0.6087239471988137,
      "learning_rate": 9.369315761305039e-06,
      "loss": 0.3376614093780518,
      "num_input_tokens_seen": 152175944,
      "step": 2535,
      "train_runtime": 32549.2355,
      "train_tokens_per_second": 4675.254
    },
    {
      "epoch": 0.3239795918367347,
      "grad_norm": 0.6321419126349881,
      "learning_rate": 9.36687837424443e-06,
      "loss": 0.3259952306747437,
      "num_input_tokens_seen": 152474144,
      "step": 2540,
      "train_runtime": 32605.5044,
      "train_tokens_per_second": 4676.331
    },
    {
      "epoch": 0.32461734693877553,
      "grad_norm": 0.6305802677195149,
      "learning_rate": 9.364436604716389e-06,
      "loss": 0.3154458999633789,
      "num_input_tokens_seen": 152777320,
      "step": 2545,
      "train_runtime": 32669.1717,
      "train_tokens_per_second": 4676.498
    },
    {
      "epoch": 0.3252551020408163,
      "grad_norm": 0.6190955912840279,
      "learning_rate": 9.361990455171405e-06,
      "loss": 0.34942078590393066,
      "num_input_tokens_seen": 153075104,
      "step": 2550,
      "train_runtime": 32732.8173,
      "train_tokens_per_second": 4676.503
    },
    {
      "epoch": 0.32589285714285715,
      "grad_norm": 0.5692325189918546,
      "learning_rate": 9.359539928064358e-06,
      "loss": 0.3225237846374512,
      "num_input_tokens_seen": 153360088,
      "step": 2555,
      "train_runtime": 32798.0862,
      "train_tokens_per_second": 4675.885
    },
    {
      "epoch": 0.32653061224489793,
      "grad_norm": 0.5707301790567272,
      "learning_rate": 9.357085025854525e-06,
      "loss": 0.31881611347198485,
      "num_input_tokens_seen": 153662392,
      "step": 2560,
      "train_runtime": 32864.1581,
      "train_tokens_per_second": 4675.683
    },
    {
      "epoch": 0.32716836734693877,
      "grad_norm": 0.6025023211113483,
      "learning_rate": 9.354625751005568e-06,
      "loss": 0.3189993381500244,
      "num_input_tokens_seen": 153965352,
      "step": 2565,
      "train_runtime": 32922.1984,
      "train_tokens_per_second": 4676.642
    },
    {
      "epoch": 0.3278061224489796,
      "grad_norm": 0.668348688818242,
      "learning_rate": 9.352162105985544e-06,
      "loss": 0.2961996555328369,
      "num_input_tokens_seen": 154268976,
      "step": 2570,
      "train_runtime": 32987.4387,
      "train_tokens_per_second": 4676.598
    },
    {
      "epoch": 0.3284438775510204,
      "grad_norm": 0.6012599795128927,
      "learning_rate": 9.349694093266893e-06,
      "loss": 0.32537577152252195,
      "num_input_tokens_seen": 154572360,
      "step": 2575,
      "train_runtime": 33040.7815,
      "train_tokens_per_second": 4678.23
    },
    {
      "epoch": 0.32908163265306123,
      "grad_norm": 0.6362604300955796,
      "learning_rate": 9.347221715326437e-06,
      "loss": 0.3090221405029297,
      "num_input_tokens_seen": 154874224,
      "step": 2580,
      "train_runtime": 33106.8149,
      "train_tokens_per_second": 4678.016
    },
    {
      "epoch": 0.32971938775510207,
      "grad_norm": 0.6232342613496551,
      "learning_rate": 9.344744974645382e-06,
      "loss": 0.3642828941345215,
      "num_input_tokens_seen": 155176232,
      "step": 2585,
      "train_runtime": 33155.6486,
      "train_tokens_per_second": 4680.235
    },
    {
      "epoch": 0.33035714285714285,
      "grad_norm": 0.5659453051690176,
      "learning_rate": 9.342263873709307e-06,
      "loss": 0.3241748571395874,
      "num_input_tokens_seen": 155478928,
      "step": 2590,
      "train_runtime": 33233.6455,
      "train_tokens_per_second": 4678.359
    },
    {
      "epoch": 0.3309948979591837,
      "grad_norm": 0.6008320540897968,
      "learning_rate": 9.339778415008171e-06,
      "loss": 0.32315621376037595,
      "num_input_tokens_seen": 155777776,
      "step": 2595,
      "train_runtime": 33301.0643,
      "train_tokens_per_second": 4677.862
    },
    {
      "epoch": 0.33163265306122447,
      "grad_norm": 0.5383881192962248,
      "learning_rate": 9.337288601036307e-06,
      "loss": 0.33694684505462646,
      "num_input_tokens_seen": 156076760,
      "step": 2600,
      "train_runtime": 33362.4084,
      "train_tokens_per_second": 4678.222
    },
    {
      "epoch": 0.3322704081632653,
      "grad_norm": 0.9747619675710759,
      "learning_rate": 9.334794434292416e-06,
      "loss": 0.31783132553100585,
      "num_input_tokens_seen": 156376176,
      "step": 2605,
      "train_runtime": 33427.4902,
      "train_tokens_per_second": 4678.071
    },
    {
      "epoch": 0.33290816326530615,
      "grad_norm": 0.5734884113962813,
      "learning_rate": 9.33229591727957e-06,
      "loss": 0.33543288707733154,
      "num_input_tokens_seen": 156692888,
      "step": 2610,
      "train_runtime": 33481.8063,
      "train_tokens_per_second": 4679.941
    },
    {
      "epoch": 0.3335459183673469,
      "grad_norm": 0.6058629357240961,
      "learning_rate": 9.329793052505203e-06,
      "loss": 0.32535901069641116,
      "num_input_tokens_seen": 156994264,
      "step": 2615,
      "train_runtime": 33555.0729,
      "train_tokens_per_second": 4678.704
    },
    {
      "epoch": 0.33418367346938777,
      "grad_norm": 0.6183279915744094,
      "learning_rate": 9.327285842481114e-06,
      "loss": 0.3213878393173218,
      "num_input_tokens_seen": 157279728,
      "step": 2620,
      "train_runtime": 33626.2153,
      "train_tokens_per_second": 4677.295
    },
    {
      "epoch": 0.33482142857142855,
      "grad_norm": 0.5495162121213072,
      "learning_rate": 9.324774289723469e-06,
      "loss": 0.3177948474884033,
      "num_input_tokens_seen": 157572640,
      "step": 2625,
      "train_runtime": 33679.8091,
      "train_tokens_per_second": 4678.549
    },
    {
      "epoch": 0.3354591836734694,
      "grad_norm": 0.5591209518424833,
      "learning_rate": 9.322258396752781e-06,
      "loss": 0.2949415922164917,
      "num_input_tokens_seen": 157869944,
      "step": 2630,
      "train_runtime": 33742.3061,
      "train_tokens_per_second": 4678.695
    },
    {
      "epoch": 0.3360969387755102,
      "grad_norm": 0.5719415058049913,
      "learning_rate": 9.319738166093925e-06,
      "loss": 0.31316614151000977,
      "num_input_tokens_seen": 158157168,
      "step": 2635,
      "train_runtime": 33808.7789,
      "train_tokens_per_second": 4677.991
    },
    {
      "epoch": 0.336734693877551,
      "grad_norm": 0.5826705577574305,
      "learning_rate": 9.31721360027613e-06,
      "loss": 0.32461071014404297,
      "num_input_tokens_seen": 158480072,
      "step": 2640,
      "train_runtime": 33859.1303,
      "train_tokens_per_second": 4680.571
    },
    {
      "epoch": 0.33737244897959184,
      "grad_norm": 0.5889465111150584,
      "learning_rate": 9.314684701832974e-06,
      "loss": 0.3051118850708008,
      "num_input_tokens_seen": 158777024,
      "step": 2645,
      "train_runtime": 33928.0924,
      "train_tokens_per_second": 4679.81
    },
    {
      "epoch": 0.3380102040816326,
      "grad_norm": 0.660741713565453,
      "learning_rate": 9.312151473302381e-06,
      "loss": 0.31853926181793213,
      "num_input_tokens_seen": 159063112,
      "step": 2650,
      "train_runtime": 33992.0852,
      "train_tokens_per_second": 4679.416
    },
    {
      "epoch": 0.33864795918367346,
      "grad_norm": 0.5938152207243252,
      "learning_rate": 9.309613917226627e-06,
      "loss": 0.31599428653717043,
      "num_input_tokens_seen": 159377008,
      "step": 2655,
      "train_runtime": 34058.1005,
      "train_tokens_per_second": 4679.562
    },
    {
      "epoch": 0.3392857142857143,
      "grad_norm": 0.6360821211627453,
      "learning_rate": 9.307072036152324e-06,
      "loss": 0.336210823059082,
      "num_input_tokens_seen": 159676160,
      "step": 2660,
      "train_runtime": 34120.1166,
      "train_tokens_per_second": 4679.825
    },
    {
      "epoch": 0.3399234693877551,
      "grad_norm": 0.5966107426536384,
      "learning_rate": 9.304525832630426e-06,
      "loss": 0.3047550439834595,
      "num_input_tokens_seen": 159984008,
      "step": 2665,
      "train_runtime": 34173.4165,
      "train_tokens_per_second": 4681.534
    },
    {
      "epoch": 0.3405612244897959,
      "grad_norm": 0.6756774734313965,
      "learning_rate": 9.30197530921623e-06,
      "loss": 0.323002552986145,
      "num_input_tokens_seen": 160278584,
      "step": 2670,
      "train_runtime": 34235.8639,
      "train_tokens_per_second": 4681.599
    },
    {
      "epoch": 0.34119897959183676,
      "grad_norm": 0.5767799189063431,
      "learning_rate": 9.29942046846936e-06,
      "loss": 0.30430898666381834,
      "num_input_tokens_seen": 160570328,
      "step": 2675,
      "train_runtime": 34302.707,
      "train_tokens_per_second": 4680.981
    },
    {
      "epoch": 0.34183673469387754,
      "grad_norm": 0.5411949368670815,
      "learning_rate": 9.29686131295378e-06,
      "loss": 0.3525160074234009,
      "num_input_tokens_seen": 160883952,
      "step": 2680,
      "train_runtime": 34364.4493,
      "train_tokens_per_second": 4681.697
    },
    {
      "epoch": 0.3424744897959184,
      "grad_norm": 0.59458431338836,
      "learning_rate": 9.294297845237778e-06,
      "loss": 0.302536940574646,
      "num_input_tokens_seen": 161174664,
      "step": 2685,
      "train_runtime": 34426.1941,
      "train_tokens_per_second": 4681.745
    },
    {
      "epoch": 0.34311224489795916,
      "grad_norm": 0.5883334196101253,
      "learning_rate": 9.291730067893978e-06,
      "loss": 0.34590141773223876,
      "num_input_tokens_seen": 161479088,
      "step": 2690,
      "train_runtime": 34482.2939,
      "train_tokens_per_second": 4682.957
    },
    {
      "epoch": 0.34375,
      "grad_norm": 0.5472357039730193,
      "learning_rate": 9.28915798349932e-06,
      "loss": 0.3102647066116333,
      "num_input_tokens_seen": 161787824,
      "step": 2695,
      "train_runtime": 34549.4021,
      "train_tokens_per_second": 4682.797
    },
    {
      "epoch": 0.34438775510204084,
      "grad_norm": 0.5392811374401831,
      "learning_rate": 9.28658159463507e-06,
      "loss": 0.3034888982772827,
      "num_input_tokens_seen": 162090808,
      "step": 2700,
      "train_runtime": 34618.9419,
      "train_tokens_per_second": 4682.142
    },
    {
      "epoch": 0.3450255102040816,
      "grad_norm": 0.6128208366625503,
      "learning_rate": 9.284000903886818e-06,
      "loss": 0.33556556701660156,
      "num_input_tokens_seen": 162395336,
      "step": 2705,
      "train_runtime": 34675.9325,
      "train_tokens_per_second": 4683.229
    },
    {
      "epoch": 0.34566326530612246,
      "grad_norm": 0.5920738541816492,
      "learning_rate": 9.281415913844465e-06,
      "loss": 0.3328749656677246,
      "num_input_tokens_seen": 162704312,
      "step": 2710,
      "train_runtime": 34740.415,
      "train_tokens_per_second": 4683.43
    },
    {
      "epoch": 0.34630102040816324,
      "grad_norm": 0.5626938659231687,
      "learning_rate": 9.278826627102228e-06,
      "loss": 0.3225257396697998,
      "num_input_tokens_seen": 163009696,
      "step": 2715,
      "train_runtime": 34808.6108,
      "train_tokens_per_second": 4683.028
    },
    {
      "epoch": 0.3469387755102041,
      "grad_norm": 0.6090967600997974,
      "learning_rate": 9.276233046258637e-06,
      "loss": 0.31754341125488283,
      "num_input_tokens_seen": 163307656,
      "step": 2720,
      "train_runtime": 34883.4983,
      "train_tokens_per_second": 4681.516
    },
    {
      "epoch": 0.3475765306122449,
      "grad_norm": 0.5780285311782188,
      "learning_rate": 9.273635173916535e-06,
      "loss": 0.3150130271911621,
      "num_input_tokens_seen": 163608672,
      "step": 2725,
      "train_runtime": 34948.7192,
      "train_tokens_per_second": 4681.393
    },
    {
      "epoch": 0.3482142857142857,
      "grad_norm": 0.5952943749771893,
      "learning_rate": 9.271033012683063e-06,
      "loss": 0.31302311420440676,
      "num_input_tokens_seen": 163922288,
      "step": 2730,
      "train_runtime": 35009.2128,
      "train_tokens_per_second": 4682.261
    },
    {
      "epoch": 0.34885204081632654,
      "grad_norm": 0.5937140156910731,
      "learning_rate": 9.268426565169677e-06,
      "loss": 0.2982464790344238,
      "num_input_tokens_seen": 164223960,
      "step": 2735,
      "train_runtime": 35080.0222,
      "train_tokens_per_second": 4681.41
    },
    {
      "epoch": 0.3494897959183674,
      "grad_norm": 0.5730896448133669,
      "learning_rate": 9.265815833992126e-06,
      "loss": 0.32490172386169436,
      "num_input_tokens_seen": 164527904,
      "step": 2740,
      "train_runtime": 35148.8704,
      "train_tokens_per_second": 4680.887
    },
    {
      "epoch": 0.35012755102040816,
      "grad_norm": 0.7064587156090127,
      "learning_rate": 9.263200821770462e-06,
      "loss": 0.33021042346954343,
      "num_input_tokens_seen": 164830720,
      "step": 2745,
      "train_runtime": 35224.2632,
      "train_tokens_per_second": 4679.465
    },
    {
      "epoch": 0.350765306122449,
      "grad_norm": 0.6771130884511102,
      "learning_rate": 9.260581531129032e-06,
      "loss": 0.3445593357086182,
      "num_input_tokens_seen": 165145048,
      "step": 2750,
      "train_runtime": 35294.5474,
      "train_tokens_per_second": 4679.053
    },
    {
      "epoch": 0.3514030612244898,
      "grad_norm": 0.5884157999002683,
      "learning_rate": 9.257957964696475e-06,
      "loss": 0.3210636615753174,
      "num_input_tokens_seen": 165452016,
      "step": 2755,
      "train_runtime": 35362.1859,
      "train_tokens_per_second": 4678.784
    },
    {
      "epoch": 0.3520408163265306,
      "grad_norm": 0.6076898201423231,
      "learning_rate": 9.255330125105728e-06,
      "loss": 0.32838950157165525,
      "num_input_tokens_seen": 165746352,
      "step": 2760,
      "train_runtime": 35427.8831,
      "train_tokens_per_second": 4678.415
    },
    {
      "epoch": 0.35267857142857145,
      "grad_norm": 0.6218117249933128,
      "learning_rate": 9.252698014994008e-06,
      "loss": 0.32069547176361085,
      "num_input_tokens_seen": 166059632,
      "step": 2765,
      "train_runtime": 35492.7951,
      "train_tokens_per_second": 4678.686
    },
    {
      "epoch": 0.35331632653061223,
      "grad_norm": 0.5975109095759474,
      "learning_rate": 9.250061637002822e-06,
      "loss": 0.3303449392318726,
      "num_input_tokens_seen": 166377536,
      "step": 2770,
      "train_runtime": 35547.221,
      "train_tokens_per_second": 4680.465
    },
    {
      "epoch": 0.3539540816326531,
      "grad_norm": 0.609305190711373,
      "learning_rate": 9.24742099377796e-06,
      "loss": 0.33595004081726076,
      "num_input_tokens_seen": 166680352,
      "step": 2775,
      "train_runtime": 35607.2614,
      "train_tokens_per_second": 4681.078
    },
    {
      "epoch": 0.35459183673469385,
      "grad_norm": 0.5837509994898966,
      "learning_rate": 9.244776087969492e-06,
      "loss": 0.32201323509216306,
      "num_input_tokens_seen": 166988808,
      "step": 2780,
      "train_runtime": 35667.4106,
      "train_tokens_per_second": 4681.832
    },
    {
      "epoch": 0.3552295918367347,
      "grad_norm": 0.5524478839275134,
      "learning_rate": 9.242126922231763e-06,
      "loss": 0.3159002780914307,
      "num_input_tokens_seen": 167284544,
      "step": 2785,
      "train_runtime": 35731.2887,
      "train_tokens_per_second": 4681.738
    },
    {
      "epoch": 0.35586734693877553,
      "grad_norm": 0.6107485769850661,
      "learning_rate": 9.239473499223399e-06,
      "loss": 0.34589133262634275,
      "num_input_tokens_seen": 167600864,
      "step": 2790,
      "train_runtime": 35803.4834,
      "train_tokens_per_second": 4681.133
    },
    {
      "epoch": 0.3565051020408163,
      "grad_norm": 0.6283645408979892,
      "learning_rate": 9.236815821607295e-06,
      "loss": 0.33268554210662843,
      "num_input_tokens_seen": 167902712,
      "step": 2795,
      "train_runtime": 35864.9739,
      "train_tokens_per_second": 4681.523
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.6203263014262271,
      "learning_rate": 9.234153892050616e-06,
      "loss": 0.31472318172454833,
      "num_input_tokens_seen": 168202368,
      "step": 2800,
      "train_runtime": 35937.563,
      "train_tokens_per_second": 4680.406
    },
    {
      "epoch": 0.35778061224489793,
      "grad_norm": 0.5850516113592246,
      "learning_rate": 9.231487713224793e-06,
      "loss": 0.28984746932983396,
      "num_input_tokens_seen": 168472288,
      "step": 2805,
      "train_runtime": 36006.3167,
      "train_tokens_per_second": 4678.965
    },
    {
      "epoch": 0.35841836734693877,
      "grad_norm": 0.5770863363835493,
      "learning_rate": 9.228817287805523e-06,
      "loss": 0.32077693939208984,
      "num_input_tokens_seen": 168776944,
      "step": 2810,
      "train_runtime": 36068.1892,
      "train_tokens_per_second": 4679.385
    },
    {
      "epoch": 0.3590561224489796,
      "grad_norm": 0.6104714968428147,
      "learning_rate": 9.226142618472765e-06,
      "loss": 0.3266309261322021,
      "num_input_tokens_seen": 169074648,
      "step": 2815,
      "train_runtime": 36144.8528,
      "train_tokens_per_second": 4677.696
    },
    {
      "epoch": 0.3596938775510204,
      "grad_norm": 0.5829999977939861,
      "learning_rate": 9.223463707910736e-06,
      "loss": 0.3029497146606445,
      "num_input_tokens_seen": 169363720,
      "step": 2820,
      "train_runtime": 36216.3979,
      "train_tokens_per_second": 4676.437
    },
    {
      "epoch": 0.36033163265306123,
      "grad_norm": 0.5888224016142662,
      "learning_rate": 9.22078055880791e-06,
      "loss": 0.3287858724594116,
      "num_input_tokens_seen": 169667624,
      "step": 2825,
      "train_runtime": 36274.5372,
      "train_tokens_per_second": 4677.32
    },
    {
      "epoch": 0.36096938775510207,
      "grad_norm": 0.6344613567674637,
      "learning_rate": 9.218093173857014e-06,
      "loss": 0.3351175785064697,
      "num_input_tokens_seen": 169943176,
      "step": 2830,
      "train_runtime": 36335.8157,
      "train_tokens_per_second": 4677.016
    },
    {
      "epoch": 0.36160714285714285,
      "grad_norm": 0.5388841134081853,
      "learning_rate": 9.21540155575503e-06,
      "loss": 0.31265280246734617,
      "num_input_tokens_seen": 170242256,
      "step": 2835,
      "train_runtime": 36393.7576,
      "train_tokens_per_second": 4677.787
    },
    {
      "epoch": 0.3622448979591837,
      "grad_norm": 0.6093139387451828,
      "learning_rate": 9.212705707203177e-06,
      "loss": 0.3472312927246094,
      "num_input_tokens_seen": 170547664,
      "step": 2840,
      "train_runtime": 36458.8101,
      "train_tokens_per_second": 4677.818
    },
    {
      "epoch": 0.36288265306122447,
      "grad_norm": 0.652622476960517,
      "learning_rate": 9.210005630906936e-06,
      "loss": 0.35820655822753905,
      "num_input_tokens_seen": 170836896,
      "step": 2845,
      "train_runtime": 36524.0788,
      "train_tokens_per_second": 4677.377
    },
    {
      "epoch": 0.3635204081632653,
      "grad_norm": 0.6800502452751199,
      "learning_rate": 9.207301329576017e-06,
      "loss": 0.3250127792358398,
      "num_input_tokens_seen": 171124112,
      "step": 2850,
      "train_runtime": 36591.694,
      "train_tokens_per_second": 4676.583
    },
    {
      "epoch": 0.36415816326530615,
      "grad_norm": 0.6458571217852985,
      "learning_rate": 9.204592805924378e-06,
      "loss": 0.30792794227600095,
      "num_input_tokens_seen": 171415632,
      "step": 2855,
      "train_runtime": 36654.0505,
      "train_tokens_per_second": 4676.581
    },
    {
      "epoch": 0.3647959183673469,
      "grad_norm": 0.6138572708964783,
      "learning_rate": 9.201880062670207e-06,
      "loss": 0.3266331911087036,
      "num_input_tokens_seen": 171710064,
      "step": 2860,
      "train_runtime": 36715.3574,
      "train_tokens_per_second": 4676.791
    },
    {
      "epoch": 0.36543367346938777,
      "grad_norm": 0.6277275327646562,
      "learning_rate": 9.199163102535937e-06,
      "loss": 0.32140178680419923,
      "num_input_tokens_seen": 171999240,
      "step": 2865,
      "train_runtime": 36780.1837,
      "train_tokens_per_second": 4676.411
    },
    {
      "epoch": 0.36607142857142855,
      "grad_norm": 0.6434628865578913,
      "learning_rate": 9.196441928248222e-06,
      "loss": 0.3261692762374878,
      "num_input_tokens_seen": 172288472,
      "step": 2870,
      "train_runtime": 36848.3777,
      "train_tokens_per_second": 4675.605
    },
    {
      "epoch": 0.3667091836734694,
      "grad_norm": 0.6073808497029558,
      "learning_rate": 9.193716542537956e-06,
      "loss": 0.30363967418670657,
      "num_input_tokens_seen": 172588344,
      "step": 2875,
      "train_runtime": 36917.1158,
      "train_tokens_per_second": 4675.022
    },
    {
      "epoch": 0.3673469387755102,
      "grad_norm": 0.569207919774566,
      "learning_rate": 9.19098694814025e-06,
      "loss": 0.32173728942871094,
      "num_input_tokens_seen": 172875936,
      "step": 2880,
      "train_runtime": 36985.191,
      "train_tokens_per_second": 4674.193
    },
    {
      "epoch": 0.367984693877551,
      "grad_norm": 0.5974814328107237,
      "learning_rate": 9.188253147794443e-06,
      "loss": 0.3502845287322998,
      "num_input_tokens_seen": 173184040,
      "step": 2885,
      "train_runtime": 37048.7315,
      "train_tokens_per_second": 4674.493
    },
    {
      "epoch": 0.36862244897959184,
      "grad_norm": 0.6585625245888168,
      "learning_rate": 9.185515144244097e-06,
      "loss": 0.32525386810302737,
      "num_input_tokens_seen": 173471360,
      "step": 2890,
      "train_runtime": 37114.1096,
      "train_tokens_per_second": 4674.0
    },
    {
      "epoch": 0.3692602040816326,
      "grad_norm": 0.6083140568131817,
      "learning_rate": 9.182772940236986e-06,
      "loss": 0.33935277462005614,
      "num_input_tokens_seen": 173774960,
      "step": 2895,
      "train_runtime": 37177.3464,
      "train_tokens_per_second": 4674.216
    },
    {
      "epoch": 0.36989795918367346,
      "grad_norm": 0.6226672976351433,
      "learning_rate": 9.180026538525106e-06,
      "loss": 0.3294100522994995,
      "num_input_tokens_seen": 174073352,
      "step": 2900,
      "train_runtime": 37238.7128,
      "train_tokens_per_second": 4674.527
    },
    {
      "epoch": 0.3705357142857143,
      "grad_norm": 0.6169838965189202,
      "learning_rate": 9.177275941864663e-06,
      "loss": 0.34671285152435305,
      "num_input_tokens_seen": 174387672,
      "step": 2905,
      "train_runtime": 37302.8993,
      "train_tokens_per_second": 4674.909
    },
    {
      "epoch": 0.3711734693877551,
      "grad_norm": 0.6323971217256894,
      "learning_rate": 9.174521153016072e-06,
      "loss": 0.33532469272613524,
      "num_input_tokens_seen": 174680184,
      "step": 2910,
      "train_runtime": 37369.6818,
      "train_tokens_per_second": 4674.382
    },
    {
      "epoch": 0.3718112244897959,
      "grad_norm": 0.6457474184064677,
      "learning_rate": 9.171762174743954e-06,
      "loss": 0.3320272207260132,
      "num_input_tokens_seen": 174977832,
      "step": 2915,
      "train_runtime": 37431.3959,
      "train_tokens_per_second": 4674.627
    },
    {
      "epoch": 0.37244897959183676,
      "grad_norm": 0.6449675776625434,
      "learning_rate": 9.168999009817141e-06,
      "loss": 0.3314546585083008,
      "num_input_tokens_seen": 175264952,
      "step": 2920,
      "train_runtime": 37493.4442,
      "train_tokens_per_second": 4674.549
    },
    {
      "epoch": 0.37308673469387754,
      "grad_norm": 0.554900892997336,
      "learning_rate": 9.166231661008658e-06,
      "loss": 0.302897310256958,
      "num_input_tokens_seen": 175559528,
      "step": 2925,
      "train_runtime": 37557.9035,
      "train_tokens_per_second": 4674.37
    },
    {
      "epoch": 0.3737244897959184,
      "grad_norm": 0.5205229094984876,
      "learning_rate": 9.163460131095735e-06,
      "loss": 0.32758631706237795,
      "num_input_tokens_seen": 175861776,
      "step": 2930,
      "train_runtime": 37624.4119,
      "train_tokens_per_second": 4674.14
    },
    {
      "epoch": 0.37436224489795916,
      "grad_norm": 0.603082200961075,
      "learning_rate": 9.160684422859794e-06,
      "loss": 0.3190114736557007,
      "num_input_tokens_seen": 176159896,
      "step": 2935,
      "train_runtime": 37698.0916,
      "train_tokens_per_second": 4672.913
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.5625469295871551,
      "learning_rate": 9.157904539086454e-06,
      "loss": 0.30655779838562014,
      "num_input_tokens_seen": 176464624,
      "step": 2940,
      "train_runtime": 37756.5104,
      "train_tokens_per_second": 4673.754
    },
    {
      "epoch": 0.37563775510204084,
      "grad_norm": 0.5597768880272735,
      "learning_rate": 9.15512048256552e-06,
      "loss": 0.2903357982635498,
      "num_input_tokens_seen": 176765064,
      "step": 2945,
      "train_runtime": 37813.0883,
      "train_tokens_per_second": 4674.706
    },
    {
      "epoch": 0.3762755102040816,
      "grad_norm": 0.6270106554429195,
      "learning_rate": 9.15233225609099e-06,
      "loss": 0.35812888145446775,
      "num_input_tokens_seen": 177072888,
      "step": 2950,
      "train_runtime": 37876.0431,
      "train_tokens_per_second": 4675.063
    },
    {
      "epoch": 0.37691326530612246,
      "grad_norm": 0.668670832201587,
      "learning_rate": 9.149539862461044e-06,
      "loss": 0.3203724384307861,
      "num_input_tokens_seen": 177373176,
      "step": 2955,
      "train_runtime": 37939.6809,
      "train_tokens_per_second": 4675.136
    },
    {
      "epoch": 0.37755102040816324,
      "grad_norm": 0.5660199250068496,
      "learning_rate": 9.14674330447804e-06,
      "loss": 0.303403377532959,
      "num_input_tokens_seen": 177674416,
      "step": 2960,
      "train_runtime": 37990.9536,
      "train_tokens_per_second": 4676.756
    },
    {
      "epoch": 0.3781887755102041,
      "grad_norm": 0.7481698219271392,
      "learning_rate": 9.143942584948525e-06,
      "loss": 0.30948538780212403,
      "num_input_tokens_seen": 177967240,
      "step": 2965,
      "train_runtime": 38056.6991,
      "train_tokens_per_second": 4676.371
    },
    {
      "epoch": 0.3788265306122449,
      "grad_norm": 0.5222886232940973,
      "learning_rate": 9.141137706683212e-06,
      "loss": 0.3155945301055908,
      "num_input_tokens_seen": 178253216,
      "step": 2970,
      "train_runtime": 38121.8658,
      "train_tokens_per_second": 4675.879
    },
    {
      "epoch": 0.3794642857142857,
      "grad_norm": 0.49631703953331113,
      "learning_rate": 9.138328672496994e-06,
      "loss": 0.31378564834594724,
      "num_input_tokens_seen": 178544880,
      "step": 2975,
      "train_runtime": 38183.9745,
      "train_tokens_per_second": 4675.911
    },
    {
      "epoch": 0.38010204081632654,
      "grad_norm": 0.6436058231086087,
      "learning_rate": 9.135515485208933e-06,
      "loss": 0.3197082757949829,
      "num_input_tokens_seen": 178850032,
      "step": 2980,
      "train_runtime": 38253.6709,
      "train_tokens_per_second": 4675.369
    },
    {
      "epoch": 0.3807397959183674,
      "grad_norm": 0.5725211682100639,
      "learning_rate": 9.132698147642258e-06,
      "loss": 0.31581218242645265,
      "num_input_tokens_seen": 179168080,
      "step": 2985,
      "train_runtime": 38309.5937,
      "train_tokens_per_second": 4676.846
    },
    {
      "epoch": 0.38137755102040816,
      "grad_norm": 0.6246530414856488,
      "learning_rate": 9.129876662624366e-06,
      "loss": 0.3519841194152832,
      "num_input_tokens_seen": 179479912,
      "step": 2990,
      "train_runtime": 38379.5778,
      "train_tokens_per_second": 4676.443
    },
    {
      "epoch": 0.382015306122449,
      "grad_norm": 0.6094334734716867,
      "learning_rate": 9.127051032986814e-06,
      "loss": 0.2976883888244629,
      "num_input_tokens_seen": 179771336,
      "step": 2995,
      "train_runtime": 38446.0382,
      "train_tokens_per_second": 4675.939
    },
    {
      "epoch": 0.3826530612244898,
      "grad_norm": 0.579574719034915,
      "learning_rate": 9.124221261565316e-06,
      "loss": 0.3473198413848877,
      "num_input_tokens_seen": 180073608,
      "step": 3000,
      "train_runtime": 38511.5091,
      "train_tokens_per_second": 4675.839
    },
    {
      "epoch": 0.3832908163265306,
      "grad_norm": 0.6510082302461844,
      "learning_rate": 9.121387351199748e-06,
      "loss": 0.32548723220825193,
      "num_input_tokens_seen": 180367392,
      "step": 3005,
      "train_runtime": 38574.6434,
      "train_tokens_per_second": 4675.802
    },
    {
      "epoch": 0.38392857142857145,
      "grad_norm": 0.6382767685052189,
      "learning_rate": 9.118549304734136e-06,
      "loss": 0.32240729331970214,
      "num_input_tokens_seen": 180668864,
      "step": 3010,
      "train_runtime": 38634.8884,
      "train_tokens_per_second": 4676.314
    },
    {
      "epoch": 0.38456632653061223,
      "grad_norm": 0.5632490493326954,
      "learning_rate": 9.115707125016658e-06,
      "loss": 0.33153977394104006,
      "num_input_tokens_seen": 180960696,
      "step": 3015,
      "train_runtime": 38688.9969,
      "train_tokens_per_second": 4677.317
    },
    {
      "epoch": 0.3852040816326531,
      "grad_norm": 0.6514142403869994,
      "learning_rate": 9.112860814899637e-06,
      "loss": 0.32599313259124757,
      "num_input_tokens_seen": 181273312,
      "step": 3020,
      "train_runtime": 38750.2165,
      "train_tokens_per_second": 4677.995
    },
    {
      "epoch": 0.38584183673469385,
      "grad_norm": 0.6182466837580296,
      "learning_rate": 9.110010377239552e-06,
      "loss": 0.33211407661437986,
      "num_input_tokens_seen": 181574648,
      "step": 3025,
      "train_runtime": 38815.5721,
      "train_tokens_per_second": 4677.882
    },
    {
      "epoch": 0.3864795918367347,
      "grad_norm": 0.5601908246102765,
      "learning_rate": 9.107155814897007e-06,
      "loss": 0.324910306930542,
      "num_input_tokens_seen": 181866008,
      "step": 3030,
      "train_runtime": 38880.295,
      "train_tokens_per_second": 4677.588
    },
    {
      "epoch": 0.38711734693877553,
      "grad_norm": 0.4971285244063714,
      "learning_rate": 9.10429713073676e-06,
      "loss": 0.3084129810333252,
      "num_input_tokens_seen": 182175176,
      "step": 3035,
      "train_runtime": 38938.018,
      "train_tokens_per_second": 4678.594
    },
    {
      "epoch": 0.3877551020408163,
      "grad_norm": 0.6178381426220323,
      "learning_rate": 9.101434327627696e-06,
      "loss": 0.34023480415344237,
      "num_input_tokens_seen": 182486080,
      "step": 3040,
      "train_runtime": 38996.5405,
      "train_tokens_per_second": 4679.545
    },
    {
      "epoch": 0.38839285714285715,
      "grad_norm": 0.5548632507090535,
      "learning_rate": 9.098567408442844e-06,
      "loss": 0.3091944932937622,
      "num_input_tokens_seen": 182802528,
      "step": 3045,
      "train_runtime": 39052.9148,
      "train_tokens_per_second": 4680.893
    },
    {
      "epoch": 0.38903061224489793,
      "grad_norm": 0.5474032113628822,
      "learning_rate": 9.095696376059353e-06,
      "loss": 0.3133708000183105,
      "num_input_tokens_seen": 183105160,
      "step": 3050,
      "train_runtime": 39109.3133,
      "train_tokens_per_second": 4681.881
    },
    {
      "epoch": 0.38966836734693877,
      "grad_norm": 0.6843299763951822,
      "learning_rate": 9.092821233358507e-06,
      "loss": 0.33481783866882325,
      "num_input_tokens_seen": 183408712,
      "step": 3055,
      "train_runtime": 39170.6318,
      "train_tokens_per_second": 4682.302
    },
    {
      "epoch": 0.3903061224489796,
      "grad_norm": 0.5811761384996502,
      "learning_rate": 9.08994198322571e-06,
      "loss": 0.3331594705581665,
      "num_input_tokens_seen": 183706944,
      "step": 3060,
      "train_runtime": 39238.5567,
      "train_tokens_per_second": 4681.797
    },
    {
      "epoch": 0.3909438775510204,
      "grad_norm": 0.5674901161494683,
      "learning_rate": 9.087058628550492e-06,
      "loss": 0.31420512199401857,
      "num_input_tokens_seen": 184015168,
      "step": 3065,
      "train_runtime": 39303.8812,
      "train_tokens_per_second": 4681.857
    },
    {
      "epoch": 0.39158163265306123,
      "grad_norm": 0.7158630858394088,
      "learning_rate": 9.084171172226502e-06,
      "loss": 0.2939667463302612,
      "num_input_tokens_seen": 184332648,
      "step": 3070,
      "train_runtime": 39354.5198,
      "train_tokens_per_second": 4683.9
    },
    {
      "epoch": 0.39221938775510207,
      "grad_norm": 0.5816620511179589,
      "learning_rate": 9.081279617151502e-06,
      "loss": 0.3151503086090088,
      "num_input_tokens_seen": 184623808,
      "step": 3075,
      "train_runtime": 39403.7366,
      "train_tokens_per_second": 4685.439
    },
    {
      "epoch": 0.39285714285714285,
      "grad_norm": 0.6445589588862652,
      "learning_rate": 9.078383966227369e-06,
      "loss": 0.32155604362487794,
      "num_input_tokens_seen": 184932032,
      "step": 3080,
      "train_runtime": 39463.4086,
      "train_tokens_per_second": 4686.165
    },
    {
      "epoch": 0.3934948979591837,
      "grad_norm": 0.5216398653856932,
      "learning_rate": 9.075484222360094e-06,
      "loss": 0.307496976852417,
      "num_input_tokens_seen": 185235592,
      "step": 3085,
      "train_runtime": 39523.4911,
      "train_tokens_per_second": 4686.721
    },
    {
      "epoch": 0.39413265306122447,
      "grad_norm": 0.7094243044216487,
      "learning_rate": 9.07258038845977e-06,
      "loss": 0.32785470485687257,
      "num_input_tokens_seen": 185545640,
      "step": 3090,
      "train_runtime": 39583.9255,
      "train_tokens_per_second": 4687.399
    },
    {
      "epoch": 0.3947704081632653,
      "grad_norm": 0.6309446058605152,
      "learning_rate": 9.069672467440598e-06,
      "loss": 0.3231490135192871,
      "num_input_tokens_seen": 185845600,
      "step": 3095,
      "train_runtime": 39658.7876,
      "train_tokens_per_second": 4686.114
    },
    {
      "epoch": 0.39540816326530615,
      "grad_norm": 0.585772820107222,
      "learning_rate": 9.066760462220878e-06,
      "loss": 0.337585973739624,
      "num_input_tokens_seen": 186156072,
      "step": 3100,
      "train_runtime": 39717.11,
      "train_tokens_per_second": 4687.05
    },
    {
      "epoch": 0.3960459183673469,
      "grad_norm": 0.6192860021409862,
      "learning_rate": 9.063844375723014e-06,
      "loss": 0.31592369079589844,
      "num_input_tokens_seen": 186463704,
      "step": 3105,
      "train_runtime": 39786.4089,
      "train_tokens_per_second": 4686.618
    },
    {
      "epoch": 0.39668367346938777,
      "grad_norm": 0.5311206414365938,
      "learning_rate": 9.060924210873501e-06,
      "loss": 0.3220933437347412,
      "num_input_tokens_seen": 186785992,
      "step": 3110,
      "train_runtime": 39847.0541,
      "train_tokens_per_second": 4687.573
    },
    {
      "epoch": 0.39732142857142855,
      "grad_norm": 0.5881670251104146,
      "learning_rate": 9.057999970602926e-06,
      "loss": 0.3254377841949463,
      "num_input_tokens_seen": 187099280,
      "step": 3115,
      "train_runtime": 39910.9643,
      "train_tokens_per_second": 4687.917
    },
    {
      "epoch": 0.3979591836734694,
      "grad_norm": 0.5769219565213103,
      "learning_rate": 9.055071657845973e-06,
      "loss": 0.31120288372039795,
      "num_input_tokens_seen": 187390408,
      "step": 3120,
      "train_runtime": 39982.7885,
      "train_tokens_per_second": 4686.777
    },
    {
      "epoch": 0.3985969387755102,
      "grad_norm": 0.693372345161722,
      "learning_rate": 9.052139275541404e-06,
      "loss": 0.3250961065292358,
      "num_input_tokens_seen": 187701720,
      "step": 3125,
      "train_runtime": 40040.1779,
      "train_tokens_per_second": 4687.834
    },
    {
      "epoch": 0.399234693877551,
      "grad_norm": 0.5960444987475249,
      "learning_rate": 9.049202826632072e-06,
      "loss": 0.2999398231506348,
      "num_input_tokens_seen": 187998736,
      "step": 3130,
      "train_runtime": 40096.9613,
      "train_tokens_per_second": 4688.603
    },
    {
      "epoch": 0.39987244897959184,
      "grad_norm": 0.5463541161895422,
      "learning_rate": 9.046262314064907e-06,
      "loss": 0.33198843002319334,
      "num_input_tokens_seen": 188297576,
      "step": 3135,
      "train_runtime": 40161.2612,
      "train_tokens_per_second": 4688.537
    },
    {
      "epoch": 0.4005102040816326,
      "grad_norm": 0.6241038136594013,
      "learning_rate": 9.04331774079092e-06,
      "loss": 0.3063666343688965,
      "num_input_tokens_seen": 188595928,
      "step": 3140,
      "train_runtime": 40207.4881,
      "train_tokens_per_second": 4690.567
    },
    {
      "epoch": 0.40114795918367346,
      "grad_norm": 0.6073385562269218,
      "learning_rate": 9.040369109765196e-06,
      "loss": 0.3146181583404541,
      "num_input_tokens_seen": 188895056,
      "step": 3145,
      "train_runtime": 40270.6058,
      "train_tokens_per_second": 4690.644
    },
    {
      "epoch": 0.4017857142857143,
      "grad_norm": 0.6020250384039302,
      "learning_rate": 9.037416423946891e-06,
      "loss": 0.3437767505645752,
      "num_input_tokens_seen": 189191384,
      "step": 3150,
      "train_runtime": 40334.6693,
      "train_tokens_per_second": 4690.54
    },
    {
      "epoch": 0.4024234693877551,
      "grad_norm": 0.6119629066804524,
      "learning_rate": 9.034459686299232e-06,
      "loss": 0.31041080951690675,
      "num_input_tokens_seen": 189495232,
      "step": 3155,
      "train_runtime": 40395.4823,
      "train_tokens_per_second": 4691.001
    },
    {
      "epoch": 0.4030612244897959,
      "grad_norm": 0.5490785230095873,
      "learning_rate": 9.031498899789513e-06,
      "loss": 0.3348876714706421,
      "num_input_tokens_seen": 189813744,
      "step": 3160,
      "train_runtime": 40457.3509,
      "train_tokens_per_second": 4691.7
    },
    {
      "epoch": 0.40369897959183676,
      "grad_norm": 0.6039839232260207,
      "learning_rate": 9.028534067389087e-06,
      "loss": 0.325526762008667,
      "num_input_tokens_seen": 190113264,
      "step": 3165,
      "train_runtime": 40516.0089,
      "train_tokens_per_second": 4692.3
    },
    {
      "epoch": 0.40433673469387754,
      "grad_norm": 0.6303117453449549,
      "learning_rate": 9.025565192073374e-06,
      "loss": 0.3193602323532104,
      "num_input_tokens_seen": 190413432,
      "step": 3170,
      "train_runtime": 40585.7812,
      "train_tokens_per_second": 4691.629
    },
    {
      "epoch": 0.4049744897959184,
      "grad_norm": 0.502369365072304,
      "learning_rate": 9.022592276821843e-06,
      "loss": 0.314216160774231,
      "num_input_tokens_seen": 190718384,
      "step": 3175,
      "train_runtime": 40647.5957,
      "train_tokens_per_second": 4691.997
    },
    {
      "epoch": 0.40561224489795916,
      "grad_norm": 0.679147229110305,
      "learning_rate": 9.019615324618027e-06,
      "loss": 0.3158599615097046,
      "num_input_tokens_seen": 191023384,
      "step": 3180,
      "train_runtime": 40710.5992,
      "train_tokens_per_second": 4692.227
    },
    {
      "epoch": 0.40625,
      "grad_norm": 2.1548596354114316,
      "learning_rate": 9.016634338449504e-06,
      "loss": 0.3082659006118774,
      "num_input_tokens_seen": 191334512,
      "step": 3185,
      "train_runtime": 40784.1677,
      "train_tokens_per_second": 4691.392
    },
    {
      "epoch": 0.40688775510204084,
      "grad_norm": 0.6126987296501494,
      "learning_rate": 9.0136493213079e-06,
      "loss": 0.35937573909759524,
      "num_input_tokens_seen": 191644088,
      "step": 3190,
      "train_runtime": 40837.2098,
      "train_tokens_per_second": 4692.879
    },
    {
      "epoch": 0.4075255102040816,
      "grad_norm": 0.5911598977521823,
      "learning_rate": 9.01066027618889e-06,
      "loss": 0.32959134578704835,
      "num_input_tokens_seen": 191943728,
      "step": 3195,
      "train_runtime": 40905.0155,
      "train_tokens_per_second": 4692.425
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 0.6158375431469726,
      "learning_rate": 9.007667206092188e-06,
      "loss": 0.3300974130630493,
      "num_input_tokens_seen": 192245368,
      "step": 3200,
      "train_runtime": 40978.8867,
      "train_tokens_per_second": 4691.327
    },
    {
      "epoch": 0.40880102040816324,
      "grad_norm": 0.5524067403238563,
      "learning_rate": 9.00467011402155e-06,
      "loss": 0.3166048526763916,
      "num_input_tokens_seen": 192550496,
      "step": 3205,
      "train_runtime": 41056.7877,
      "train_tokens_per_second": 4689.858
    },
    {
      "epoch": 0.4094387755102041,
      "grad_norm": 0.6526435021969289,
      "learning_rate": 9.001669002984767e-06,
      "loss": 0.31752676963806153,
      "num_input_tokens_seen": 192839112,
      "step": 3210,
      "train_runtime": 41113.8029,
      "train_tokens_per_second": 4690.374
    },
    {
      "epoch": 0.4100765306122449,
      "grad_norm": 0.6171092895650897,
      "learning_rate": 8.998663875993666e-06,
      "loss": 0.3099684238433838,
      "num_input_tokens_seen": 193145744,
      "step": 3215,
      "train_runtime": 41173.0537,
      "train_tokens_per_second": 4691.072
    },
    {
      "epoch": 0.4107142857142857,
      "grad_norm": 0.6351840250528298,
      "learning_rate": 8.995654736064096e-06,
      "loss": 0.3157944679260254,
      "num_input_tokens_seen": 193444720,
      "step": 3220,
      "train_runtime": 41242.4823,
      "train_tokens_per_second": 4690.424
    },
    {
      "epoch": 0.41135204081632654,
      "grad_norm": 0.7096499688646882,
      "learning_rate": 8.992641586215944e-06,
      "loss": 0.31511194705963136,
      "num_input_tokens_seen": 193740152,
      "step": 3225,
      "train_runtime": 41316.6862,
      "train_tokens_per_second": 4689.15
    },
    {
      "epoch": 0.4119897959183674,
      "grad_norm": 0.6489642469997576,
      "learning_rate": 8.989624429473116e-06,
      "loss": 0.3469348192214966,
      "num_input_tokens_seen": 194038616,
      "step": 3230,
      "train_runtime": 41372.4449,
      "train_tokens_per_second": 4690.045
    },
    {
      "epoch": 0.41262755102040816,
      "grad_norm": 0.6253957179146292,
      "learning_rate": 8.986603268863536e-06,
      "loss": 0.30385029315948486,
      "num_input_tokens_seen": 194342968,
      "step": 3235,
      "train_runtime": 41435.5938,
      "train_tokens_per_second": 4690.242
    },
    {
      "epoch": 0.413265306122449,
      "grad_norm": 0.5764196777561061,
      "learning_rate": 8.983578107419153e-06,
      "loss": 0.3338628768920898,
      "num_input_tokens_seen": 194634464,
      "step": 3240,
      "train_runtime": 41499.3601,
      "train_tokens_per_second": 4690.059
    },
    {
      "epoch": 0.4139030612244898,
      "grad_norm": 0.6363802995158825,
      "learning_rate": 8.980548948175926e-06,
      "loss": 0.32214794158935545,
      "num_input_tokens_seen": 194934600,
      "step": 3245,
      "train_runtime": 41570.4501,
      "train_tokens_per_second": 4689.259
    },
    {
      "epoch": 0.4145408163265306,
      "grad_norm": 0.6255302689335023,
      "learning_rate": 8.97751579417383e-06,
      "loss": 0.347865629196167,
      "num_input_tokens_seen": 195236360,
      "step": 3250,
      "train_runtime": 41640.6809,
      "train_tokens_per_second": 4688.597
    },
    {
      "epoch": 0.41517857142857145,
      "grad_norm": 0.5504023160480871,
      "learning_rate": 8.974478648456845e-06,
      "loss": 0.33169217109680177,
      "num_input_tokens_seen": 195547784,
      "step": 3255,
      "train_runtime": 41708.5101,
      "train_tokens_per_second": 4688.438
    },
    {
      "epoch": 0.41581632653061223,
      "grad_norm": 0.599057696663512,
      "learning_rate": 8.971437514072959e-06,
      "loss": 0.3121060848236084,
      "num_input_tokens_seen": 195850144,
      "step": 3260,
      "train_runtime": 41778.9639,
      "train_tokens_per_second": 4687.769
    },
    {
      "epoch": 0.4164540816326531,
      "grad_norm": 0.5589495788658304,
      "learning_rate": 8.968392394074164e-06,
      "loss": 0.3321085929870605,
      "num_input_tokens_seen": 196149992,
      "step": 3265,
      "train_runtime": 41836.1766,
      "train_tokens_per_second": 4688.526
    },
    {
      "epoch": 0.41709183673469385,
      "grad_norm": 0.8438741101842504,
      "learning_rate": 8.965343291516448e-06,
      "loss": 0.3220966339111328,
      "num_input_tokens_seen": 196442800,
      "step": 3270,
      "train_runtime": 41908.493,
      "train_tokens_per_second": 4687.422
    },
    {
      "epoch": 0.4177295918367347,
      "grad_norm": 0.5219612583404035,
      "learning_rate": 8.962290209459801e-06,
      "loss": 0.29323763847351075,
      "num_input_tokens_seen": 196751416,
      "step": 3275,
      "train_runtime": 41961.715,
      "train_tokens_per_second": 4688.832
    },
    {
      "epoch": 0.41836734693877553,
      "grad_norm": 0.5921272335273982,
      "learning_rate": 8.959233150968203e-06,
      "loss": 0.30167346000671386,
      "num_input_tokens_seen": 197039568,
      "step": 3280,
      "train_runtime": 42022.326,
      "train_tokens_per_second": 4688.926
    },
    {
      "epoch": 0.4190051020408163,
      "grad_norm": 0.6436079802230608,
      "learning_rate": 8.956172119109625e-06,
      "loss": 0.32533998489379884,
      "num_input_tokens_seen": 197316768,
      "step": 3285,
      "train_runtime": 42088.9143,
      "train_tokens_per_second": 4688.094
    },
    {
      "epoch": 0.41964285714285715,
      "grad_norm": 0.5625984955814649,
      "learning_rate": 8.953107116956027e-06,
      "loss": 0.33030574321746825,
      "num_input_tokens_seen": 197621504,
      "step": 3290,
      "train_runtime": 42147.4511,
      "train_tokens_per_second": 4688.813
    },
    {
      "epoch": 0.42028061224489793,
      "grad_norm": 0.6472563142078905,
      "learning_rate": 8.950038147583353e-06,
      "loss": 0.32565693855285643,
      "num_input_tokens_seen": 197910040,
      "step": 3295,
      "train_runtime": 42210.7532,
      "train_tokens_per_second": 4688.617
    },
    {
      "epoch": 0.42091836734693877,
      "grad_norm": 0.651944716977726,
      "learning_rate": 8.946965214071528e-06,
      "loss": 0.3277165174484253,
      "num_input_tokens_seen": 198210360,
      "step": 3300,
      "train_runtime": 42276.57,
      "train_tokens_per_second": 4688.421
    },
    {
      "epoch": 0.4215561224489796,
      "grad_norm": 0.5644835498940443,
      "learning_rate": 8.943888319504456e-06,
      "loss": 0.3059228897094727,
      "num_input_tokens_seen": 198520280,
      "step": 3305,
      "train_runtime": 42340.6205,
      "train_tokens_per_second": 4688.648
    },
    {
      "epoch": 0.4221938775510204,
      "grad_norm": 0.6174447870952193,
      "learning_rate": 8.940807466970014e-06,
      "loss": 0.33121767044067385,
      "num_input_tokens_seen": 198822224,
      "step": 3310,
      "train_runtime": 42395.7598,
      "train_tokens_per_second": 4689.672
    },
    {
      "epoch": 0.42283163265306123,
      "grad_norm": 0.6052243828745715,
      "learning_rate": 8.937722659560054e-06,
      "loss": 0.342440128326416,
      "num_input_tokens_seen": 199124776,
      "step": 3315,
      "train_runtime": 42464.2008,
      "train_tokens_per_second": 4689.239
    },
    {
      "epoch": 0.42346938775510207,
      "grad_norm": 0.5528499168638967,
      "learning_rate": 8.934633900370395e-06,
      "loss": 0.3259343385696411,
      "num_input_tokens_seen": 199420408,
      "step": 3320,
      "train_runtime": 42534.3094,
      "train_tokens_per_second": 4688.46
    },
    {
      "epoch": 0.42410714285714285,
      "grad_norm": 0.5671579367989095,
      "learning_rate": 8.931541192500822e-06,
      "loss": 0.29687554836273194,
      "num_input_tokens_seen": 199716672,
      "step": 3325,
      "train_runtime": 42598.2667,
      "train_tokens_per_second": 4688.376
    },
    {
      "epoch": 0.4247448979591837,
      "grad_norm": 0.5894132936373059,
      "learning_rate": 8.928444539055086e-06,
      "loss": 0.33872098922729493,
      "num_input_tokens_seen": 200021208,
      "step": 3330,
      "train_runtime": 42668.2397,
      "train_tokens_per_second": 4687.824
    },
    {
      "epoch": 0.42538265306122447,
      "grad_norm": 0.6090440054505855,
      "learning_rate": 8.92534394314089e-06,
      "loss": 0.33945250511169434,
      "num_input_tokens_seen": 200315992,
      "step": 3335,
      "train_runtime": 42726.5478,
      "train_tokens_per_second": 4688.326
    },
    {
      "epoch": 0.4260204081632653,
      "grad_norm": 0.5735284959805205,
      "learning_rate": 8.922239407869904e-06,
      "loss": 0.3439453601837158,
      "num_input_tokens_seen": 200609824,
      "step": 3340,
      "train_runtime": 42790.9851,
      "train_tokens_per_second": 4688.133
    },
    {
      "epoch": 0.42665816326530615,
      "grad_norm": 0.5581658857242167,
      "learning_rate": 8.919130936357743e-06,
      "loss": 0.3368233680725098,
      "num_input_tokens_seen": 200902288,
      "step": 3345,
      "train_runtime": 42866.4423,
      "train_tokens_per_second": 4686.703
    },
    {
      "epoch": 0.4272959183673469,
      "grad_norm": 0.6014445852209643,
      "learning_rate": 8.916018531723975e-06,
      "loss": 0.3475268602371216,
      "num_input_tokens_seen": 201211416,
      "step": 3350,
      "train_runtime": 42934.4717,
      "train_tokens_per_second": 4686.477
    },
    {
      "epoch": 0.42793367346938777,
      "grad_norm": 0.6456452430025559,
      "learning_rate": 8.91290219709212e-06,
      "loss": 0.31551270484924315,
      "num_input_tokens_seen": 201488776,
      "step": 3355,
      "train_runtime": 42999.6964,
      "train_tokens_per_second": 4685.819
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.6640341757996214,
      "learning_rate": 8.909781935589629e-06,
      "loss": 0.3253334522247314,
      "num_input_tokens_seen": 201782120,
      "step": 3360,
      "train_runtime": 43065.6986,
      "train_tokens_per_second": 4685.449
    },
    {
      "epoch": 0.4292091836734694,
      "grad_norm": 0.6483824588057587,
      "learning_rate": 8.906657750347909e-06,
      "loss": 0.32048940658569336,
      "num_input_tokens_seen": 202068240,
      "step": 3365,
      "train_runtime": 43134.2271,
      "train_tokens_per_second": 4684.638
    },
    {
      "epoch": 0.4298469387755102,
      "grad_norm": 0.5715684018538307,
      "learning_rate": 8.903529644502297e-06,
      "loss": 0.3232470989227295,
      "num_input_tokens_seen": 202369128,
      "step": 3370,
      "train_runtime": 43202.7163,
      "train_tokens_per_second": 4684.176
    },
    {
      "epoch": 0.430484693877551,
      "grad_norm": 0.5447639244848349,
      "learning_rate": 8.900397621192063e-06,
      "loss": 0.33262906074523924,
      "num_input_tokens_seen": 202668864,
      "step": 3375,
      "train_runtime": 43261.3826,
      "train_tokens_per_second": 4684.752
    },
    {
      "epoch": 0.43112244897959184,
      "grad_norm": 0.6136167689578654,
      "learning_rate": 8.897261683560413e-06,
      "loss": 0.3484243869781494,
      "num_input_tokens_seen": 202992192,
      "step": 3380,
      "train_runtime": 43312.0423,
      "train_tokens_per_second": 4686.738
    },
    {
      "epoch": 0.4317602040816326,
      "grad_norm": 0.5605678764252943,
      "learning_rate": 8.89412183475448e-06,
      "loss": 0.31671900749206544,
      "num_input_tokens_seen": 203292680,
      "step": 3385,
      "train_runtime": 43385.9122,
      "train_tokens_per_second": 4685.684
    },
    {
      "epoch": 0.43239795918367346,
      "grad_norm": 0.5449801304902008,
      "learning_rate": 8.890978077925323e-06,
      "loss": 0.30098576545715333,
      "num_input_tokens_seen": 203602792,
      "step": 3390,
      "train_runtime": 43455.4191,
      "train_tokens_per_second": 4685.326
    },
    {
      "epoch": 0.4330357142857143,
      "grad_norm": 0.5718214775448611,
      "learning_rate": 8.887830416227918e-06,
      "loss": 0.3349436283111572,
      "num_input_tokens_seen": 203894240,
      "step": 3395,
      "train_runtime": 43516.0658,
      "train_tokens_per_second": 4685.493
    },
    {
      "epoch": 0.4336734693877551,
      "grad_norm": 0.5491846722687903,
      "learning_rate": 8.884678852821165e-06,
      "loss": 0.3207132339477539,
      "num_input_tokens_seen": 204195296,
      "step": 3400,
      "train_runtime": 43584.9617,
      "train_tokens_per_second": 4684.994
    },
    {
      "epoch": 0.4343112244897959,
      "grad_norm": 0.5131986116157814,
      "learning_rate": 8.881523390867876e-06,
      "loss": 0.28901073932647703,
      "num_input_tokens_seen": 204501952,
      "step": 3405,
      "train_runtime": 43643.4227,
      "train_tokens_per_second": 4685.745
    },
    {
      "epoch": 0.43494897959183676,
      "grad_norm": 0.6369571741647311,
      "learning_rate": 8.878364033534781e-06,
      "loss": 0.316038990020752,
      "num_input_tokens_seen": 204808760,
      "step": 3410,
      "train_runtime": 43701.254,
      "train_tokens_per_second": 4686.565
    },
    {
      "epoch": 0.43558673469387754,
      "grad_norm": 0.6049679474241683,
      "learning_rate": 8.875200783992514e-06,
      "loss": 0.3199471950531006,
      "num_input_tokens_seen": 205114392,
      "step": 3415,
      "train_runtime": 43759.3217,
      "train_tokens_per_second": 4687.33
    },
    {
      "epoch": 0.4362244897959184,
      "grad_norm": 0.5442531763542,
      "learning_rate": 8.872033645415617e-06,
      "loss": 0.3352459669113159,
      "num_input_tokens_seen": 205429936,
      "step": 3420,
      "train_runtime": 43820.5566,
      "train_tokens_per_second": 4687.981
    },
    {
      "epoch": 0.43686224489795916,
      "grad_norm": 0.60890948257868,
      "learning_rate": 8.868862620982534e-06,
      "loss": 0.32459220886230467,
      "num_input_tokens_seen": 205724432,
      "step": 3425,
      "train_runtime": 43889.6241,
      "train_tokens_per_second": 4687.314
    },
    {
      "epoch": 0.4375,
      "grad_norm": 0.5715140225599952,
      "learning_rate": 8.86568771387561e-06,
      "loss": 0.2873693466186523,
      "num_input_tokens_seen": 206029616,
      "step": 3430,
      "train_runtime": 43962.5793,
      "train_tokens_per_second": 4686.477
    },
    {
      "epoch": 0.43813775510204084,
      "grad_norm": 0.6279456681396556,
      "learning_rate": 8.862508927281086e-06,
      "loss": 0.3436110973358154,
      "num_input_tokens_seen": 206319016,
      "step": 3435,
      "train_runtime": 44037.1796,
      "train_tokens_per_second": 4685.11
    },
    {
      "epoch": 0.4387755102040816,
      "grad_norm": 0.6260792472646204,
      "learning_rate": 8.859326264389098e-06,
      "loss": 0.30514523983001707,
      "num_input_tokens_seen": 206606112,
      "step": 3440,
      "train_runtime": 44100.5969,
      "train_tokens_per_second": 4684.882
    },
    {
      "epoch": 0.43941326530612246,
      "grad_norm": 0.68482154832881,
      "learning_rate": 8.856139728393667e-06,
      "loss": 0.32379128932952883,
      "num_input_tokens_seen": 206904232,
      "step": 3445,
      "train_runtime": 44168.5824,
      "train_tokens_per_second": 4684.421
    },
    {
      "epoch": 0.44005102040816324,
      "grad_norm": 0.5930425986425011,
      "learning_rate": 8.852949322492708e-06,
      "loss": 0.28973195552825926,
      "num_input_tokens_seen": 207193664,
      "step": 3450,
      "train_runtime": 44231.8882,
      "train_tokens_per_second": 4684.26
    },
    {
      "epoch": 0.4406887755102041,
      "grad_norm": 0.6825580182781947,
      "learning_rate": 8.849755049888014e-06,
      "loss": 0.29935970306396487,
      "num_input_tokens_seen": 207494832,
      "step": 3455,
      "train_runtime": 44287.5443,
      "train_tokens_per_second": 4685.174
    },
    {
      "epoch": 0.4413265306122449,
      "grad_norm": 0.5133954359714166,
      "learning_rate": 8.84655691378526e-06,
      "loss": 0.29735455513000486,
      "num_input_tokens_seen": 207804120,
      "step": 3460,
      "train_runtime": 44348.9126,
      "train_tokens_per_second": 4685.664
    },
    {
      "epoch": 0.4419642857142857,
      "grad_norm": 0.7101324223192657,
      "learning_rate": 8.843354917394e-06,
      "loss": 0.32294692993164065,
      "num_input_tokens_seen": 208093840,
      "step": 3465,
      "train_runtime": 44422.4075,
      "train_tokens_per_second": 4684.434
    },
    {
      "epoch": 0.44260204081632654,
      "grad_norm": 0.616180186681375,
      "learning_rate": 8.840149063927663e-06,
      "loss": 0.3034783601760864,
      "num_input_tokens_seen": 208411648,
      "step": 3470,
      "train_runtime": 44483.9149,
      "train_tokens_per_second": 4685.101
    },
    {
      "epoch": 0.4432397959183674,
      "grad_norm": 0.6407549783367804,
      "learning_rate": 8.836939356603542e-06,
      "loss": 0.3212604522705078,
      "num_input_tokens_seen": 208712352,
      "step": 3475,
      "train_runtime": 44555.0981,
      "train_tokens_per_second": 4684.365
    },
    {
      "epoch": 0.44387755102040816,
      "grad_norm": 0.6310875509643901,
      "learning_rate": 8.833725798642809e-06,
      "loss": 0.3165870666503906,
      "num_input_tokens_seen": 209023584,
      "step": 3480,
      "train_runtime": 44627.7031,
      "train_tokens_per_second": 4683.718
    },
    {
      "epoch": 0.444515306122449,
      "grad_norm": 0.5486220172333803,
      "learning_rate": 8.830508393270487e-06,
      "loss": 0.3032956123352051,
      "num_input_tokens_seen": 209326680,
      "step": 3485,
      "train_runtime": 44687.319,
      "train_tokens_per_second": 4684.252
    },
    {
      "epoch": 0.4451530612244898,
      "grad_norm": 0.6036304484912378,
      "learning_rate": 8.827287143715472e-06,
      "loss": 0.3312588930130005,
      "num_input_tokens_seen": 209634992,
      "step": 3490,
      "train_runtime": 44743.8105,
      "train_tokens_per_second": 4685.229
    },
    {
      "epoch": 0.4457908163265306,
      "grad_norm": 1.0000561430667463,
      "learning_rate": 8.824062053210512e-06,
      "loss": 0.3453768253326416,
      "num_input_tokens_seen": 209945984,
      "step": 3495,
      "train_runtime": 44799.5629,
      "train_tokens_per_second": 4686.34
    },
    {
      "epoch": 0.44642857142857145,
      "grad_norm": 0.6165732258516156,
      "learning_rate": 8.820833124992207e-06,
      "loss": 0.31481189727783204,
      "num_input_tokens_seen": 210254920,
      "step": 3500,
      "train_runtime": 44865.7239,
      "train_tokens_per_second": 4686.315
    },
    {
      "epoch": 0.44706632653061223,
      "grad_norm": 0.635726610334787,
      "learning_rate": 8.817600362301018e-06,
      "loss": 0.33425703048706057,
      "num_input_tokens_seen": 210556560,
      "step": 3505,
      "train_runtime": 44924.0301,
      "train_tokens_per_second": 4686.947
    },
    {
      "epoch": 0.4477040816326531,
      "grad_norm": 0.566236154272825,
      "learning_rate": 8.814363768381242e-06,
      "loss": 0.33902735710144044,
      "num_input_tokens_seen": 210871448,
      "step": 3510,
      "train_runtime": 44987.2966,
      "train_tokens_per_second": 4687.355
    },
    {
      "epoch": 0.44834183673469385,
      "grad_norm": 0.5145047175941144,
      "learning_rate": 8.81112334648103e-06,
      "loss": 0.31283936500549314,
      "num_input_tokens_seen": 211172816,
      "step": 3515,
      "train_runtime": 45042.6304,
      "train_tokens_per_second": 4688.288
    },
    {
      "epoch": 0.4489795918367347,
      "grad_norm": 0.5673226459992542,
      "learning_rate": 8.807879099852372e-06,
      "loss": 0.30037946701049806,
      "num_input_tokens_seen": 211474016,
      "step": 3520,
      "train_runtime": 45115.5993,
      "train_tokens_per_second": 4687.381
    },
    {
      "epoch": 0.44961734693877553,
      "grad_norm": 0.5881209697924512,
      "learning_rate": 8.804631031751096e-06,
      "loss": 0.29651522636413574,
      "num_input_tokens_seen": 211771800,
      "step": 3525,
      "train_runtime": 45174.4221,
      "train_tokens_per_second": 4687.87
    },
    {
      "epoch": 0.4502551020408163,
      "grad_norm": 0.6182947239481263,
      "learning_rate": 8.801379145436866e-06,
      "loss": 0.2972166299819946,
      "num_input_tokens_seen": 212068808,
      "step": 3530,
      "train_runtime": 45241.6059,
      "train_tokens_per_second": 4687.473
    },
    {
      "epoch": 0.45089285714285715,
      "grad_norm": 0.5918083309528803,
      "learning_rate": 8.798123444173174e-06,
      "loss": 0.3081668853759766,
      "num_input_tokens_seen": 212370816,
      "step": 3535,
      "train_runtime": 45312.2187,
      "train_tokens_per_second": 4686.833
    },
    {
      "epoch": 0.45153061224489793,
      "grad_norm": 0.6423743956345168,
      "learning_rate": 8.794863931227346e-06,
      "loss": 0.30899815559387206,
      "num_input_tokens_seen": 212667944,
      "step": 3540,
      "train_runtime": 45381.4461,
      "train_tokens_per_second": 4686.231
    },
    {
      "epoch": 0.45216836734693877,
      "grad_norm": 0.566686236522869,
      "learning_rate": 8.79160060987053e-06,
      "loss": 0.3186926364898682,
      "num_input_tokens_seen": 212976656,
      "step": 3545,
      "train_runtime": 45441.7075,
      "train_tokens_per_second": 4686.81
    },
    {
      "epoch": 0.4528061224489796,
      "grad_norm": 0.5891548119707445,
      "learning_rate": 8.788333483377699e-06,
      "loss": 0.33061072826385496,
      "num_input_tokens_seen": 213278936,
      "step": 3550,
      "train_runtime": 45501.3328,
      "train_tokens_per_second": 4687.312
    },
    {
      "epoch": 0.4534438775510204,
      "grad_norm": 0.6233804886050337,
      "learning_rate": 8.785062555027637e-06,
      "loss": 0.31149797439575194,
      "num_input_tokens_seen": 213573832,
      "step": 3555,
      "train_runtime": 45558.5219,
      "train_tokens_per_second": 4687.901
    },
    {
      "epoch": 0.45408163265306123,
      "grad_norm": 0.5966089919793925,
      "learning_rate": 8.781787828102958e-06,
      "loss": 0.3415715217590332,
      "num_input_tokens_seen": 213883168,
      "step": 3560,
      "train_runtime": 45616.5553,
      "train_tokens_per_second": 4688.718
    },
    {
      "epoch": 0.45471938775510207,
      "grad_norm": 0.5861249272190848,
      "learning_rate": 8.778509305890069e-06,
      "loss": 0.3334629774093628,
      "num_input_tokens_seen": 214190912,
      "step": 3565,
      "train_runtime": 45675.1674,
      "train_tokens_per_second": 4689.439
    },
    {
      "epoch": 0.45535714285714285,
      "grad_norm": 0.6369980505919692,
      "learning_rate": 8.775226991679205e-06,
      "loss": 0.3384882926940918,
      "num_input_tokens_seen": 214502256,
      "step": 3570,
      "train_runtime": 45737.4637,
      "train_tokens_per_second": 4689.859
    },
    {
      "epoch": 0.4559948979591837,
      "grad_norm": 0.6964150252536505,
      "learning_rate": 8.77194088876439e-06,
      "loss": 0.3140512704849243,
      "num_input_tokens_seen": 214783328,
      "step": 3575,
      "train_runtime": 45802.4673,
      "train_tokens_per_second": 4689.34
    },
    {
      "epoch": 0.45663265306122447,
      "grad_norm": 0.7522468475076703,
      "learning_rate": 8.768651000443463e-06,
      "loss": 0.3444084644317627,
      "num_input_tokens_seen": 215087536,
      "step": 3580,
      "train_runtime": 45860.8405,
      "train_tokens_per_second": 4690.004
    },
    {
      "epoch": 0.4572704081632653,
      "grad_norm": 0.6162932046287327,
      "learning_rate": 8.765357330018056e-06,
      "loss": 0.3437525510787964,
      "num_input_tokens_seen": 215396736,
      "step": 3585,
      "train_runtime": 45922.62,
      "train_tokens_per_second": 4690.428
    },
    {
      "epoch": 0.45790816326530615,
      "grad_norm": 0.5843167708342699,
      "learning_rate": 8.762059880793595e-06,
      "loss": 0.32638962268829347,
      "num_input_tokens_seen": 215690808,
      "step": 3590,
      "train_runtime": 45983.7838,
      "train_tokens_per_second": 4690.584
    },
    {
      "epoch": 0.4585459183673469,
      "grad_norm": 0.5824123384380021,
      "learning_rate": 8.758758656079303e-06,
      "loss": 0.3440886974334717,
      "num_input_tokens_seen": 215988576,
      "step": 3595,
      "train_runtime": 46053.4432,
      "train_tokens_per_second": 4689.955
    },
    {
      "epoch": 0.45918367346938777,
      "grad_norm": 0.5574951426169432,
      "learning_rate": 8.755453659188186e-06,
      "loss": 0.33778557777404783,
      "num_input_tokens_seen": 216292144,
      "step": 3600,
      "train_runtime": 46114.0844,
      "train_tokens_per_second": 4690.371
    },
    {
      "epoch": 0.45982142857142855,
      "grad_norm": 0.5899376920985997,
      "learning_rate": 8.752144893437045e-06,
      "loss": 0.3352713108062744,
      "num_input_tokens_seen": 216593688,
      "step": 3605,
      "train_runtime": 46172.8945,
      "train_tokens_per_second": 4690.927
    },
    {
      "epoch": 0.4604591836734694,
      "grad_norm": 0.5847283713717142,
      "learning_rate": 8.748832362146454e-06,
      "loss": 0.3330048084259033,
      "num_input_tokens_seen": 216892048,
      "step": 3610,
      "train_runtime": 46246.9183,
      "train_tokens_per_second": 4689.87
    },
    {
      "epoch": 0.4610969387755102,
      "grad_norm": 0.6010388681667821,
      "learning_rate": 8.745516068640767e-06,
      "loss": 0.2907567977905273,
      "num_input_tokens_seen": 217182808,
      "step": 3615,
      "train_runtime": 46313.6167,
      "train_tokens_per_second": 4689.394
    },
    {
      "epoch": 0.461734693877551,
      "grad_norm": 0.6939045255363584,
      "learning_rate": 8.742196016248121e-06,
      "loss": 0.2981436014175415,
      "num_input_tokens_seen": 217476928,
      "step": 3620,
      "train_runtime": 46376.0369,
      "train_tokens_per_second": 4689.425
    },
    {
      "epoch": 0.46237244897959184,
      "grad_norm": 0.5563072773387618,
      "learning_rate": 8.738872208300417e-06,
      "loss": 0.31744837760925293,
      "num_input_tokens_seen": 217795992,
      "step": 3625,
      "train_runtime": 46434.5679,
      "train_tokens_per_second": 4690.385
    },
    {
      "epoch": 0.4630102040816326,
      "grad_norm": 0.6408152032899023,
      "learning_rate": 8.735544648133331e-06,
      "loss": 0.3405008316040039,
      "num_input_tokens_seen": 218097816,
      "step": 3630,
      "train_runtime": 46498.9249,
      "train_tokens_per_second": 4690.384
    },
    {
      "epoch": 0.46364795918367346,
      "grad_norm": 0.6246290179711698,
      "learning_rate": 8.7322133390863e-06,
      "loss": 0.34192686080932616,
      "num_input_tokens_seen": 218410720,
      "step": 3635,
      "train_runtime": 46567.0491,
      "train_tokens_per_second": 4690.242
    },
    {
      "epoch": 0.4642857142857143,
      "grad_norm": 0.5739391616187374,
      "learning_rate": 8.728878284502526e-06,
      "loss": 0.31751222610473634,
      "num_input_tokens_seen": 218711912,
      "step": 3640,
      "train_runtime": 46620.9326,
      "train_tokens_per_second": 4691.281
    },
    {
      "epoch": 0.4649234693877551,
      "grad_norm": 0.6632229939982521,
      "learning_rate": 8.725539487728969e-06,
      "loss": 0.3527590990066528,
      "num_input_tokens_seen": 219006304,
      "step": 3645,
      "train_runtime": 46682.8516,
      "train_tokens_per_second": 4691.365
    },
    {
      "epoch": 0.4655612244897959,
      "grad_norm": 0.6959421541032137,
      "learning_rate": 8.722196952116346e-06,
      "loss": 0.33302555084228513,
      "num_input_tokens_seen": 219299648,
      "step": 3650,
      "train_runtime": 46743.2716,
      "train_tokens_per_second": 4691.577
    },
    {
      "epoch": 0.46619897959183676,
      "grad_norm": 0.5985583649914814,
      "learning_rate": 8.718850681019125e-06,
      "loss": 0.33802356719970705,
      "num_input_tokens_seen": 219592736,
      "step": 3655,
      "train_runtime": 46813.6385,
      "train_tokens_per_second": 4690.785
    },
    {
      "epoch": 0.46683673469387754,
      "grad_norm": 0.6809419700700615,
      "learning_rate": 8.715500677795519e-06,
      "loss": 0.33800294399261477,
      "num_input_tokens_seen": 219904256,
      "step": 3660,
      "train_runtime": 46871.2824,
      "train_tokens_per_second": 4691.663
    },
    {
      "epoch": 0.4674744897959184,
      "grad_norm": 0.6257133112041434,
      "learning_rate": 8.712146945807494e-06,
      "loss": 0.3399499416351318,
      "num_input_tokens_seen": 220187592,
      "step": 3665,
      "train_runtime": 46940.6251,
      "train_tokens_per_second": 4690.768
    },
    {
      "epoch": 0.46811224489795916,
      "grad_norm": 0.5210991711236733,
      "learning_rate": 8.708789488420753e-06,
      "loss": 0.32831809520721433,
      "num_input_tokens_seen": 220488080,
      "step": 3670,
      "train_runtime": 46990.9104,
      "train_tokens_per_second": 4692.143
    },
    {
      "epoch": 0.46875,
      "grad_norm": 0.5852972272116405,
      "learning_rate": 8.705428309004737e-06,
      "loss": 0.3230716705322266,
      "num_input_tokens_seen": 220795200,
      "step": 3675,
      "train_runtime": 47058.8267,
      "train_tokens_per_second": 4691.898
    },
    {
      "epoch": 0.46938775510204084,
      "grad_norm": 0.5679929385782518,
      "learning_rate": 8.702063410932628e-06,
      "loss": 0.3512821912765503,
      "num_input_tokens_seen": 221091784,
      "step": 3680,
      "train_runtime": 47126.5318,
      "train_tokens_per_second": 4691.45
    },
    {
      "epoch": 0.4700255102040816,
      "grad_norm": 0.589604580913648,
      "learning_rate": 8.698694797581335e-06,
      "loss": 0.2926384210586548,
      "num_input_tokens_seen": 221399344,
      "step": 3685,
      "train_runtime": 47186.9126,
      "train_tokens_per_second": 4691.965
    },
    {
      "epoch": 0.47066326530612246,
      "grad_norm": 0.6807809442674404,
      "learning_rate": 8.695322472331497e-06,
      "loss": 0.3219421148300171,
      "num_input_tokens_seen": 221684384,
      "step": 3690,
      "train_runtime": 47251.7952,
      "train_tokens_per_second": 4691.555
    },
    {
      "epoch": 0.47130102040816324,
      "grad_norm": 0.5938210126388256,
      "learning_rate": 8.691946438567476e-06,
      "loss": 0.31500236988067626,
      "num_input_tokens_seen": 221997312,
      "step": 3695,
      "train_runtime": 47303.1611,
      "train_tokens_per_second": 4693.076
    },
    {
      "epoch": 0.4719387755102041,
      "grad_norm": 0.6250194018595292,
      "learning_rate": 8.688566699677362e-06,
      "loss": 0.35878853797912597,
      "num_input_tokens_seen": 222286616,
      "step": 3700,
      "train_runtime": 47372.0139,
      "train_tokens_per_second": 4692.362
    },
    {
      "epoch": 0.4725765306122449,
      "grad_norm": 0.5749015433831429,
      "learning_rate": 8.685183259052951e-06,
      "loss": 0.3441047668457031,
      "num_input_tokens_seen": 222584216,
      "step": 3705,
      "train_runtime": 47427.3118,
      "train_tokens_per_second": 4693.165
    },
    {
      "epoch": 0.4732142857142857,
      "grad_norm": 0.6419814308046566,
      "learning_rate": 8.68179612008977e-06,
      "loss": 0.3282122850418091,
      "num_input_tokens_seen": 222872640,
      "step": 3710,
      "train_runtime": 47497.5249,
      "train_tokens_per_second": 4692.3
    },
    {
      "epoch": 0.47385204081632654,
      "grad_norm": 0.5790820541061571,
      "learning_rate": 8.678405286187046e-06,
      "loss": 0.31385016441345215,
      "num_input_tokens_seen": 223165808,
      "step": 3715,
      "train_runtime": 47555.5361,
      "train_tokens_per_second": 4692.741
    },
    {
      "epoch": 0.4744897959183674,
      "grad_norm": 0.6269065031836492,
      "learning_rate": 8.675010760747717e-06,
      "loss": 0.3562500476837158,
      "num_input_tokens_seen": 223466440,
      "step": 3720,
      "train_runtime": 47632.67,
      "train_tokens_per_second": 4691.453
    },
    {
      "epoch": 0.47512755102040816,
      "grad_norm": 0.5956108833409146,
      "learning_rate": 8.671612547178428e-06,
      "loss": 0.2841916561126709,
      "num_input_tokens_seen": 223768640,
      "step": 3725,
      "train_runtime": 47692.5676,
      "train_tokens_per_second": 4691.898
    },
    {
      "epoch": 0.475765306122449,
      "grad_norm": 0.5753812107727605,
      "learning_rate": 8.668210648889523e-06,
      "loss": 0.3037557125091553,
      "num_input_tokens_seen": 224057208,
      "step": 3730,
      "train_runtime": 47762.3177,
      "train_tokens_per_second": 4691.087
    },
    {
      "epoch": 0.4764030612244898,
      "grad_norm": 0.6385507767296246,
      "learning_rate": 8.664805069295046e-06,
      "loss": 0.32805795669555665,
      "num_input_tokens_seen": 224342936,
      "step": 3735,
      "train_runtime": 47829.9512,
      "train_tokens_per_second": 4690.428
    },
    {
      "epoch": 0.4770408163265306,
      "grad_norm": 0.5394380313557849,
      "learning_rate": 8.661395811812732e-06,
      "loss": 0.3325211524963379,
      "num_input_tokens_seen": 224636952,
      "step": 3740,
      "train_runtime": 47894.1985,
      "train_tokens_per_second": 4690.275
    },
    {
      "epoch": 0.47767857142857145,
      "grad_norm": 0.6191117466541218,
      "learning_rate": 8.657982879864007e-06,
      "loss": 0.3126026391983032,
      "num_input_tokens_seen": 224933736,
      "step": 3745,
      "train_runtime": 47967.8341,
      "train_tokens_per_second": 4689.262
    },
    {
      "epoch": 0.47831632653061223,
      "grad_norm": 0.6409625329689695,
      "learning_rate": 8.654566276873992e-06,
      "loss": 0.3323666572570801,
      "num_input_tokens_seen": 225243288,
      "step": 3750,
      "train_runtime": 48041.5411,
      "train_tokens_per_second": 4688.511
    },
    {
      "epoch": 0.4789540816326531,
      "grad_norm": 0.6793527208485282,
      "learning_rate": 8.651146006271483e-06,
      "loss": 0.34303305149078367,
      "num_input_tokens_seen": 225532656,
      "step": 3755,
      "train_runtime": 48104.817,
      "train_tokens_per_second": 4688.359
    },
    {
      "epoch": 0.47959183673469385,
      "grad_norm": 0.6336706905015029,
      "learning_rate": 8.64772207148896e-06,
      "loss": 0.3112976312637329,
      "num_input_tokens_seen": 225828272,
      "step": 3760,
      "train_runtime": 48168.9398,
      "train_tokens_per_second": 4688.255
    },
    {
      "epoch": 0.4802295918367347,
      "grad_norm": 0.7415618167200674,
      "learning_rate": 8.644294475962583e-06,
      "loss": 0.3418597936630249,
      "num_input_tokens_seen": 226122576,
      "step": 3765,
      "train_runtime": 48231.1311,
      "train_tokens_per_second": 4688.312
    },
    {
      "epoch": 0.48086734693877553,
      "grad_norm": 0.6102114351372901,
      "learning_rate": 8.640863223132183e-06,
      "loss": 0.3238770246505737,
      "num_input_tokens_seen": 226425192,
      "step": 3770,
      "train_runtime": 48292.1185,
      "train_tokens_per_second": 4688.657
    },
    {
      "epoch": 0.4815051020408163,
      "grad_norm": 0.6071001149024883,
      "learning_rate": 8.63742831644126e-06,
      "loss": 0.3333350658416748,
      "num_input_tokens_seen": 226718696,
      "step": 3775,
      "train_runtime": 48353.2438,
      "train_tokens_per_second": 4688.8
    },
    {
      "epoch": 0.48214285714285715,
      "grad_norm": 0.5605538974785508,
      "learning_rate": 8.633989759336984e-06,
      "loss": 0.3035125732421875,
      "num_input_tokens_seen": 227007680,
      "step": 3780,
      "train_runtime": 48414.7454,
      "train_tokens_per_second": 4688.813
    },
    {
      "epoch": 0.48278061224489793,
      "grad_norm": 0.6273277143209608,
      "learning_rate": 8.630547555270187e-06,
      "loss": 0.34965779781341555,
      "num_input_tokens_seen": 227308672,
      "step": 3785,
      "train_runtime": 48482.0996,
      "train_tokens_per_second": 4688.507
    },
    {
      "epoch": 0.48341836734693877,
      "grad_norm": 0.6720150621973842,
      "learning_rate": 8.627101707695365e-06,
      "loss": 0.31407837867736815,
      "num_input_tokens_seen": 227613472,
      "step": 3790,
      "train_runtime": 48544.7444,
      "train_tokens_per_second": 4688.736
    },
    {
      "epoch": 0.4840561224489796,
      "grad_norm": 0.6009996028106223,
      "learning_rate": 8.62365222007066e-06,
      "loss": 0.3078362226486206,
      "num_input_tokens_seen": 227928160,
      "step": 3795,
      "train_runtime": 48609.3152,
      "train_tokens_per_second": 4688.981
    },
    {
      "epoch": 0.4846938775510204,
      "grad_norm": 0.6075814248709552,
      "learning_rate": 8.620199095857877e-06,
      "loss": 0.31285624504089354,
      "num_input_tokens_seen": 228223560,
      "step": 3800,
      "train_runtime": 48671.7948,
      "train_tokens_per_second": 4689.031
    },
    {
      "epoch": 0.48533163265306123,
      "grad_norm": 0.6516147255557245,
      "learning_rate": 8.616742338522467e-06,
      "loss": 0.3143020153045654,
      "num_input_tokens_seen": 228520104,
      "step": 3805,
      "train_runtime": 48728.592,
      "train_tokens_per_second": 4689.651
    },
    {
      "epoch": 0.48596938775510207,
      "grad_norm": 0.6460117416964749,
      "learning_rate": 8.613281951533527e-06,
      "loss": 0.3114283800125122,
      "num_input_tokens_seen": 228806096,
      "step": 3810,
      "train_runtime": 48802.2897,
      "train_tokens_per_second": 4688.43
    },
    {
      "epoch": 0.48660714285714285,
      "grad_norm": 0.5595261197912934,
      "learning_rate": 8.609817938363799e-06,
      "loss": 0.3336583852767944,
      "num_input_tokens_seen": 229106960,
      "step": 3815,
      "train_runtime": 48859.5373,
      "train_tokens_per_second": 4689.094
    },
    {
      "epoch": 0.4872448979591837,
      "grad_norm": 0.6269987048054863,
      "learning_rate": 8.60635030248966e-06,
      "loss": 0.3201381921768188,
      "num_input_tokens_seen": 229412248,
      "step": 3820,
      "train_runtime": 48929.9828,
      "train_tokens_per_second": 4688.582
    },
    {
      "epoch": 0.48788265306122447,
      "grad_norm": 0.6205237417180155,
      "learning_rate": 8.602879047391127e-06,
      "loss": 0.3380389451980591,
      "num_input_tokens_seen": 229711352,
      "step": 3825,
      "train_runtime": 48992.2172,
      "train_tokens_per_second": 4688.731
    },
    {
      "epoch": 0.4885204081632653,
      "grad_norm": 0.6368541432579827,
      "learning_rate": 8.599404176551843e-06,
      "loss": 0.31137685775756835,
      "num_input_tokens_seen": 229995312,
      "step": 3830,
      "train_runtime": 49045.3729,
      "train_tokens_per_second": 4689.44
    },
    {
      "epoch": 0.48915816326530615,
      "grad_norm": 0.588258055441562,
      "learning_rate": 8.595925693459086e-06,
      "loss": 0.33580193519592283,
      "num_input_tokens_seen": 230293248,
      "step": 3835,
      "train_runtime": 49116.6255,
      "train_tokens_per_second": 4688.703
    },
    {
      "epoch": 0.4897959183673469,
      "grad_norm": 0.6410596538528932,
      "learning_rate": 8.59244360160376e-06,
      "loss": 0.3307149887084961,
      "num_input_tokens_seen": 230581480,
      "step": 3840,
      "train_runtime": 49176.25,
      "train_tokens_per_second": 4688.879
    },
    {
      "epoch": 0.49043367346938777,
      "grad_norm": 0.6670839034570284,
      "learning_rate": 8.588957904480381e-06,
      "loss": 0.29323835372924806,
      "num_input_tokens_seen": 230880424,
      "step": 3845,
      "train_runtime": 49231.6709,
      "train_tokens_per_second": 4689.673
    },
    {
      "epoch": 0.49107142857142855,
      "grad_norm": 0.5555115612515904,
      "learning_rate": 8.585468605587094e-06,
      "loss": 0.342155909538269,
      "num_input_tokens_seen": 231184408,
      "step": 3850,
      "train_runtime": 49298.6318,
      "train_tokens_per_second": 4689.469
    },
    {
      "epoch": 0.4917091836734694,
      "grad_norm": 0.6073705320677633,
      "learning_rate": 8.581975708425652e-06,
      "loss": 0.33124098777770994,
      "num_input_tokens_seen": 231495880,
      "step": 3855,
      "train_runtime": 49356.435,
      "train_tokens_per_second": 4690.288
    },
    {
      "epoch": 0.4923469387755102,
      "grad_norm": 0.604541663972026,
      "learning_rate": 8.578479216501422e-06,
      "loss": 0.34891135692596437,
      "num_input_tokens_seen": 231786912,
      "step": 3860,
      "train_runtime": 49425.9884,
      "train_tokens_per_second": 4689.576
    },
    {
      "epoch": 0.492984693877551,
      "grad_norm": 0.5689662248411393,
      "learning_rate": 8.574979133323378e-06,
      "loss": 0.33214640617370605,
      "num_input_tokens_seen": 232092464,
      "step": 3865,
      "train_runtime": 49491.7501,
      "train_tokens_per_second": 4689.518
    },
    {
      "epoch": 0.49362244897959184,
      "grad_norm": 0.5835116464633966,
      "learning_rate": 8.571475462404098e-06,
      "loss": 0.31715946197509765,
      "num_input_tokens_seen": 232386200,
      "step": 3870,
      "train_runtime": 49566.4294,
      "train_tokens_per_second": 4688.379
    },
    {
      "epoch": 0.4942602040816326,
      "grad_norm": 0.5947108264754066,
      "learning_rate": 8.567968207259759e-06,
      "loss": 0.31617448329925535,
      "num_input_tokens_seen": 232683656,
      "step": 3875,
      "train_runtime": 49622.2812,
      "train_tokens_per_second": 4689.096
    },
    {
      "epoch": 0.49489795918367346,
      "grad_norm": 0.7234261523076563,
      "learning_rate": 8.564457371410139e-06,
      "loss": 0.32854447364807127,
      "num_input_tokens_seen": 232987152,
      "step": 3880,
      "train_runtime": 49693.4975,
      "train_tokens_per_second": 4688.484
    },
    {
      "epoch": 0.4955357142857143,
      "grad_norm": 0.5856843551656463,
      "learning_rate": 8.560942958378604e-06,
      "loss": 0.31505703926086426,
      "num_input_tokens_seen": 233274344,
      "step": 3885,
      "train_runtime": 49767.2794,
      "train_tokens_per_second": 4687.304
    },
    {
      "epoch": 0.4961734693877551,
      "grad_norm": 0.6173254846006689,
      "learning_rate": 8.557424971692117e-06,
      "loss": 0.3176628351211548,
      "num_input_tokens_seen": 233581288,
      "step": 3890,
      "train_runtime": 49824.5603,
      "train_tokens_per_second": 4688.075
    },
    {
      "epoch": 0.4968112244897959,
      "grad_norm": 0.6407797196291325,
      "learning_rate": 8.55390341488122e-06,
      "loss": 0.33142986297607424,
      "num_input_tokens_seen": 233883048,
      "step": 3895,
      "train_runtime": 49902.8684,
      "train_tokens_per_second": 4686.766
    },
    {
      "epoch": 0.49744897959183676,
      "grad_norm": 0.6809906279006612,
      "learning_rate": 8.550378291480041e-06,
      "loss": 0.3280193567276001,
      "num_input_tokens_seen": 234179512,
      "step": 3900,
      "train_runtime": 49966.9784,
      "train_tokens_per_second": 4686.685
    },
    {
      "epoch": 0.49808673469387754,
      "grad_norm": 0.5442065664869937,
      "learning_rate": 8.54684960502629e-06,
      "loss": 0.3108412027359009,
      "num_input_tokens_seen": 234481384,
      "step": 3905,
      "train_runtime": 50026.9708,
      "train_tokens_per_second": 4687.099
    },
    {
      "epoch": 0.4987244897959184,
      "grad_norm": 0.6001100539623434,
      "learning_rate": 8.543317359061247e-06,
      "loss": 0.31249351501464845,
      "num_input_tokens_seen": 234779032,
      "step": 3910,
      "train_runtime": 50094.8558,
      "train_tokens_per_second": 4686.689
    },
    {
      "epoch": 0.49936224489795916,
      "grad_norm": 0.608820557192486,
      "learning_rate": 8.539781557129772e-06,
      "loss": 0.3282254695892334,
      "num_input_tokens_seen": 235067024,
      "step": 3915,
      "train_runtime": 50163.2095,
      "train_tokens_per_second": 4686.044
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6134403949536948,
      "learning_rate": 8.536242202780285e-06,
      "loss": 0.3381755828857422,
      "num_input_tokens_seen": 235365120,
      "step": 3920,
      "train_runtime": 50225.6456,
      "train_tokens_per_second": 4686.154
    },
    {
      "epoch": 0.5006377551020408,
      "grad_norm": 0.529618085808314,
      "learning_rate": 8.532699299564777e-06,
      "loss": 0.30429079532623293,
      "num_input_tokens_seen": 235663552,
      "step": 3925,
      "train_runtime": 50283.826,
      "train_tokens_per_second": 4686.667
    },
    {
      "epoch": 0.5012755102040817,
      "grad_norm": 0.5401237694639268,
      "learning_rate": 8.529152851038798e-06,
      "loss": 0.3484530448913574,
      "num_input_tokens_seen": 235979680,
      "step": 3930,
      "train_runtime": 50345.9633,
      "train_tokens_per_second": 4687.162
    },
    {
      "epoch": 0.5019132653061225,
      "grad_norm": 0.5979224945384497,
      "learning_rate": 8.52560286076146e-06,
      "loss": 0.3274678230285645,
      "num_input_tokens_seen": 236281408,
      "step": 3935,
      "train_runtime": 50414.4229,
      "train_tokens_per_second": 4686.782
    },
    {
      "epoch": 0.5025510204081632,
      "grad_norm": 0.7674430914384112,
      "learning_rate": 8.522049332295421e-06,
      "loss": 0.3291932106018066,
      "num_input_tokens_seen": 236585680,
      "step": 3940,
      "train_runtime": 50473.1228,
      "train_tokens_per_second": 4687.36
    },
    {
      "epoch": 0.5031887755102041,
      "grad_norm": 0.6846798692593288,
      "learning_rate": 8.5184922692069e-06,
      "loss": 0.3128329277038574,
      "num_input_tokens_seen": 236896464,
      "step": 3945,
      "train_runtime": 50533.5785,
      "train_tokens_per_second": 4687.902
    },
    {
      "epoch": 0.5038265306122449,
      "grad_norm": 0.6282341631706102,
      "learning_rate": 8.514931675065654e-06,
      "loss": 0.338327956199646,
      "num_input_tokens_seen": 237201312,
      "step": 3950,
      "train_runtime": 50595.675,
      "train_tokens_per_second": 4688.174
    },
    {
      "epoch": 0.5044642857142857,
      "grad_norm": 0.5993769276365452,
      "learning_rate": 8.51136755344499e-06,
      "loss": 0.32868807315826415,
      "num_input_tokens_seen": 237482056,
      "step": 3955,
      "train_runtime": 50661.2703,
      "train_tokens_per_second": 4687.645
    },
    {
      "epoch": 0.5051020408163265,
      "grad_norm": 0.8772604828539141,
      "learning_rate": 8.507799907921754e-06,
      "loss": 0.30279438495635985,
      "num_input_tokens_seen": 237779776,
      "step": 3960,
      "train_runtime": 50733.6626,
      "train_tokens_per_second": 4686.825
    },
    {
      "epoch": 0.5057397959183674,
      "grad_norm": 0.5415278949758084,
      "learning_rate": 8.504228742076325e-06,
      "loss": 0.3054949283599854,
      "num_input_tokens_seen": 238075712,
      "step": 3965,
      "train_runtime": 50797.9224,
      "train_tokens_per_second": 4686.721
    },
    {
      "epoch": 0.5063775510204082,
      "grad_norm": 0.6364973916989575,
      "learning_rate": 8.500654059492618e-06,
      "loss": 0.338958215713501,
      "num_input_tokens_seen": 238371592,
      "step": 3970,
      "train_runtime": 50857.4995,
      "train_tokens_per_second": 4687.049
    },
    {
      "epoch": 0.5070153061224489,
      "grad_norm": 0.6041940008915029,
      "learning_rate": 8.497075863758079e-06,
      "loss": 0.32581939697265627,
      "num_input_tokens_seen": 238677336,
      "step": 3975,
      "train_runtime": 50923.371,
      "train_tokens_per_second": 4686.99
    },
    {
      "epoch": 0.5076530612244898,
      "grad_norm": 0.5422672998389819,
      "learning_rate": 8.493494158463674e-06,
      "loss": 0.32880821228027346,
      "num_input_tokens_seen": 238986112,
      "step": 3980,
      "train_runtime": 50997.1362,
      "train_tokens_per_second": 4686.265
    },
    {
      "epoch": 0.5082908163265306,
      "grad_norm": 0.594000297710505,
      "learning_rate": 8.489908947203897e-06,
      "loss": 0.34253578186035155,
      "num_input_tokens_seen": 239296208,
      "step": 3985,
      "train_runtime": 51065.1994,
      "train_tokens_per_second": 4686.092
    },
    {
      "epoch": 0.5089285714285714,
      "grad_norm": 0.6022810808139175,
      "learning_rate": 8.486320233576757e-06,
      "loss": 0.3009287595748901,
      "num_input_tokens_seen": 239582264,
      "step": 3990,
      "train_runtime": 51133.0334,
      "train_tokens_per_second": 4685.469
    },
    {
      "epoch": 0.5095663265306123,
      "grad_norm": 0.5739027753713752,
      "learning_rate": 8.482728021183777e-06,
      "loss": 0.2939309597015381,
      "num_input_tokens_seen": 239882328,
      "step": 3995,
      "train_runtime": 51206.1406,
      "train_tokens_per_second": 4684.64
    },
    {
      "epoch": 0.5102040816326531,
      "grad_norm": 0.6039637328340155,
      "learning_rate": 8.479132313629996e-06,
      "loss": 0.3285217761993408,
      "num_input_tokens_seen": 240167464,
      "step": 4000,
      "train_runtime": 51272.7174,
      "train_tokens_per_second": 4684.118
    },
    {
      "epoch": 0.5108418367346939,
      "grad_norm": 0.5853286775780016,
      "learning_rate": 8.475533114523954e-06,
      "loss": 0.3112363815307617,
      "num_input_tokens_seen": 240456352,
      "step": 4005,
      "train_runtime": 51337.5249,
      "train_tokens_per_second": 4683.832
    },
    {
      "epoch": 0.5114795918367347,
      "grad_norm": 0.5563056564871881,
      "learning_rate": 8.471930427477704e-06,
      "loss": 0.3293452739715576,
      "num_input_tokens_seen": 240753960,
      "step": 4010,
      "train_runtime": 51405.9346,
      "train_tokens_per_second": 4683.388
    },
    {
      "epoch": 0.5121173469387755,
      "grad_norm": 0.5897994472914004,
      "learning_rate": 8.468324256106789e-06,
      "loss": 0.32114386558532715,
      "num_input_tokens_seen": 241066184,
      "step": 4015,
      "train_runtime": 51470.099,
      "train_tokens_per_second": 4683.616
    },
    {
      "epoch": 0.5127551020408163,
      "grad_norm": 0.6492322193880778,
      "learning_rate": 8.464714604030256e-06,
      "loss": 0.30771784782409667,
      "num_input_tokens_seen": 241371416,
      "step": 4020,
      "train_runtime": 51539.3753,
      "train_tokens_per_second": 4683.243
    },
    {
      "epoch": 0.5133928571428571,
      "grad_norm": 0.6364867293398085,
      "learning_rate": 8.461101474870642e-06,
      "loss": 0.3258524894714355,
      "num_input_tokens_seen": 241671152,
      "step": 4025,
      "train_runtime": 51602.3985,
      "train_tokens_per_second": 4683.332
    },
    {
      "epoch": 0.514030612244898,
      "grad_norm": 0.6674651696176255,
      "learning_rate": 8.457484872253976e-06,
      "loss": 0.31948442459106446,
      "num_input_tokens_seen": 241978096,
      "step": 4030,
      "train_runtime": 51661.0997,
      "train_tokens_per_second": 4683.952
    },
    {
      "epoch": 0.5146683673469388,
      "grad_norm": 0.5973739586952436,
      "learning_rate": 8.453864799809772e-06,
      "loss": 0.3342707633972168,
      "num_input_tokens_seen": 242285752,
      "step": 4035,
      "train_runtime": 51729.6137,
      "train_tokens_per_second": 4683.695
    },
    {
      "epoch": 0.5153061224489796,
      "grad_norm": 0.5616265387147885,
      "learning_rate": 8.450241261171022e-06,
      "loss": 0.31052722930908205,
      "num_input_tokens_seen": 242579024,
      "step": 4040,
      "train_runtime": 51804.5156,
      "train_tokens_per_second": 4682.585
    },
    {
      "epoch": 0.5159438775510204,
      "grad_norm": 0.5969278945501028,
      "learning_rate": 8.446614259974201e-06,
      "loss": 0.30537512302398684,
      "num_input_tokens_seen": 242870944,
      "step": 4045,
      "train_runtime": 51877.7153,
      "train_tokens_per_second": 4681.604
    },
    {
      "epoch": 0.5165816326530612,
      "grad_norm": 0.6356334614782454,
      "learning_rate": 8.44298379985926e-06,
      "loss": 0.3459012031555176,
      "num_input_tokens_seen": 243173728,
      "step": 4050,
      "train_runtime": 51941.9346,
      "train_tokens_per_second": 4681.646
    },
    {
      "epoch": 0.517219387755102,
      "grad_norm": 0.6497022024172072,
      "learning_rate": 8.439349884469621e-06,
      "loss": 0.31816725730895995,
      "num_input_tokens_seen": 243482640,
      "step": 4055,
      "train_runtime": 52005.9855,
      "train_tokens_per_second": 4681.82
    },
    {
      "epoch": 0.5178571428571429,
      "grad_norm": 0.6731650154036644,
      "learning_rate": 8.435712517452167e-06,
      "loss": 0.3488030910491943,
      "num_input_tokens_seen": 243787968,
      "step": 4060,
      "train_runtime": 52071.1423,
      "train_tokens_per_second": 4681.825
    },
    {
      "epoch": 0.5184948979591837,
      "grad_norm": 0.5692312417385954,
      "learning_rate": 8.432071702457253e-06,
      "loss": 0.3288985013961792,
      "num_input_tokens_seen": 244092336,
      "step": 4065,
      "train_runtime": 52142.0306,
      "train_tokens_per_second": 4681.297
    },
    {
      "epoch": 0.5191326530612245,
      "grad_norm": 0.5408079660399894,
      "learning_rate": 8.428427443138689e-06,
      "loss": 0.3070178985595703,
      "num_input_tokens_seen": 244379936,
      "step": 4070,
      "train_runtime": 52196.2794,
      "train_tokens_per_second": 4681.942
    },
    {
      "epoch": 0.5197704081632653,
      "grad_norm": 0.6145659542390675,
      "learning_rate": 8.424779743153747e-06,
      "loss": 0.30741429328918457,
      "num_input_tokens_seen": 244662840,
      "step": 4075,
      "train_runtime": 52253.7032,
      "train_tokens_per_second": 4682.211
    },
    {
      "epoch": 0.5204081632653061,
      "grad_norm": 0.6353161518584013,
      "learning_rate": 8.421128606163145e-06,
      "loss": 0.3511251926422119,
      "num_input_tokens_seen": 244966000,
      "step": 4080,
      "train_runtime": 52317.8245,
      "train_tokens_per_second": 4682.267
    },
    {
      "epoch": 0.5210459183673469,
      "grad_norm": 0.6094802701256447,
      "learning_rate": 8.417474035831055e-06,
      "loss": 0.2934889316558838,
      "num_input_tokens_seen": 245260784,
      "step": 4085,
      "train_runtime": 52383.4783,
      "train_tokens_per_second": 4682.026
    },
    {
      "epoch": 0.5216836734693877,
      "grad_norm": 0.6727496823709653,
      "learning_rate": 8.413816035825095e-06,
      "loss": 0.3110150098800659,
      "num_input_tokens_seen": 245550280,
      "step": 4090,
      "train_runtime": 52457.2366,
      "train_tokens_per_second": 4680.961
    },
    {
      "epoch": 0.5223214285714286,
      "grad_norm": 0.5485785769866987,
      "learning_rate": 8.410154609816324e-06,
      "loss": 0.296339225769043,
      "num_input_tokens_seen": 245839128,
      "step": 4095,
      "train_runtime": 52526.9748,
      "train_tokens_per_second": 4680.245
    },
    {
      "epoch": 0.5229591836734694,
      "grad_norm": 0.5557512696798947,
      "learning_rate": 8.406489761479235e-06,
      "loss": 0.33946802616119387,
      "num_input_tokens_seen": 246127200,
      "step": 4100,
      "train_runtime": 52577.7259,
      "train_tokens_per_second": 4681.207
    },
    {
      "epoch": 0.5235969387755102,
      "grad_norm": 0.6014225056580619,
      "learning_rate": 8.402821494491762e-06,
      "loss": 0.30245676040649416,
      "num_input_tokens_seen": 246421424,
      "step": 4105,
      "train_runtime": 52647.4962,
      "train_tokens_per_second": 4680.592
    },
    {
      "epoch": 0.5242346938775511,
      "grad_norm": 0.6246767533912915,
      "learning_rate": 8.399149812535269e-06,
      "loss": 0.33345613479614256,
      "num_input_tokens_seen": 246735904,
      "step": 4110,
      "train_runtime": 52702.3016,
      "train_tokens_per_second": 4681.691
    },
    {
      "epoch": 0.5248724489795918,
      "grad_norm": 0.5961399193010901,
      "learning_rate": 8.395474719294541e-06,
      "loss": 0.31256182193756105,
      "num_input_tokens_seen": 247038904,
      "step": 4115,
      "train_runtime": 52767.868,
      "train_tokens_per_second": 4681.616
    },
    {
      "epoch": 0.5255102040816326,
      "grad_norm": 0.6732978417816181,
      "learning_rate": 8.391796218457794e-06,
      "loss": 0.3450355052947998,
      "num_input_tokens_seen": 247343624,
      "step": 4120,
      "train_runtime": 52834.4009,
      "train_tokens_per_second": 4681.488
    },
    {
      "epoch": 0.5261479591836735,
      "grad_norm": 0.6034276666979609,
      "learning_rate": 8.388114313716658e-06,
      "loss": 0.35344223976135253,
      "num_input_tokens_seen": 247636408,
      "step": 4125,
      "train_runtime": 52906.1553,
      "train_tokens_per_second": 4680.673
    },
    {
      "epoch": 0.5267857142857143,
      "grad_norm": 0.5822756790678641,
      "learning_rate": 8.384429008766183e-06,
      "loss": 0.34767148494720457,
      "num_input_tokens_seen": 247943096,
      "step": 4130,
      "train_runtime": 52972.1639,
      "train_tokens_per_second": 4680.63
    },
    {
      "epoch": 0.5274234693877551,
      "grad_norm": 0.5534471990850092,
      "learning_rate": 8.380740307304831e-06,
      "loss": 0.28744490146636964,
      "num_input_tokens_seen": 248250080,
      "step": 4135,
      "train_runtime": 53046.5905,
      "train_tokens_per_second": 4679.85
    },
    {
      "epoch": 0.5280612244897959,
      "grad_norm": 0.6180102089530837,
      "learning_rate": 8.377048213034469e-06,
      "loss": 0.3034853219985962,
      "num_input_tokens_seen": 248551000,
      "step": 4140,
      "train_runtime": 53104.4224,
      "train_tokens_per_second": 4680.42
    },
    {
      "epoch": 0.5286989795918368,
      "grad_norm": 0.6340302009964914,
      "learning_rate": 8.373352729660373e-06,
      "loss": 0.34547905921936034,
      "num_input_tokens_seen": 248853736,
      "step": 4145,
      "train_runtime": 53167.5911,
      "train_tokens_per_second": 4680.553
    },
    {
      "epoch": 0.5293367346938775,
      "grad_norm": 0.580590114831241,
      "learning_rate": 8.369653860891218e-06,
      "loss": 0.335661792755127,
      "num_input_tokens_seen": 249153448,
      "step": 4150,
      "train_runtime": 53235.7742,
      "train_tokens_per_second": 4680.188
    },
    {
      "epoch": 0.5299744897959183,
      "grad_norm": 0.8148059899498963,
      "learning_rate": 8.365951610439078e-06,
      "loss": 0.3232774257659912,
      "num_input_tokens_seen": 249454664,
      "step": 4155,
      "train_runtime": 53310.5359,
      "train_tokens_per_second": 4679.275
    },
    {
      "epoch": 0.5306122448979592,
      "grad_norm": 0.5925809138419219,
      "learning_rate": 8.362245982019419e-06,
      "loss": 0.321329402923584,
      "num_input_tokens_seen": 249759488,
      "step": 4160,
      "train_runtime": 53364.5393,
      "train_tokens_per_second": 4680.252
    },
    {
      "epoch": 0.53125,
      "grad_norm": 0.5797856049342134,
      "learning_rate": 8.358536979351098e-06,
      "loss": 0.34303817749023435,
      "num_input_tokens_seen": 250052816,
      "step": 4165,
      "train_runtime": 53435.8316,
      "train_tokens_per_second": 4679.497
    },
    {
      "epoch": 0.5318877551020408,
      "grad_norm": 0.6059898471049472,
      "learning_rate": 8.35482460615636e-06,
      "loss": 0.31549644470214844,
      "num_input_tokens_seen": 250362600,
      "step": 4170,
      "train_runtime": 53500.0538,
      "train_tokens_per_second": 4679.67
    },
    {
      "epoch": 0.5325255102040817,
      "grad_norm": 0.6307640947510069,
      "learning_rate": 8.351108866160827e-06,
      "loss": 0.32418289184570315,
      "num_input_tokens_seen": 250653216,
      "step": 4175,
      "train_runtime": 53560.4576,
      "train_tokens_per_second": 4679.818
    },
    {
      "epoch": 0.5331632653061225,
      "grad_norm": 0.63557884154035,
      "learning_rate": 8.347389763093507e-06,
      "loss": 0.330997633934021,
      "num_input_tokens_seen": 250938992,
      "step": 4180,
      "train_runtime": 53624.5737,
      "train_tokens_per_second": 4679.552
    },
    {
      "epoch": 0.5338010204081632,
      "grad_norm": 0.6664096437044886,
      "learning_rate": 8.34366730068678e-06,
      "loss": 0.34677467346191404,
      "num_input_tokens_seen": 251237216,
      "step": 4185,
      "train_runtime": 53695.991,
      "train_tokens_per_second": 4678.882
    },
    {
      "epoch": 0.5344387755102041,
      "grad_norm": 0.5452953894462239,
      "learning_rate": 8.339941482676394e-06,
      "loss": 0.329576301574707,
      "num_input_tokens_seen": 251540176,
      "step": 4190,
      "train_runtime": 53768.7052,
      "train_tokens_per_second": 4678.189
    },
    {
      "epoch": 0.5350765306122449,
      "grad_norm": 0.5926858088338427,
      "learning_rate": 8.336212312801469e-06,
      "loss": 0.3385266542434692,
      "num_input_tokens_seen": 251846328,
      "step": 4195,
      "train_runtime": 53836.4095,
      "train_tokens_per_second": 4677.993
    },
    {
      "epoch": 0.5357142857142857,
      "grad_norm": 0.631641679615256,
      "learning_rate": 8.332479794804489e-06,
      "loss": 0.32497210502624513,
      "num_input_tokens_seen": 252149328,
      "step": 4200,
      "train_runtime": 53895.9305,
      "train_tokens_per_second": 4678.448
    },
    {
      "epoch": 0.5363520408163265,
      "grad_norm": 0.6191889858674562,
      "learning_rate": 8.328743932431293e-06,
      "loss": 0.33168396949768064,
      "num_input_tokens_seen": 252469872,
      "step": 4205,
      "train_runtime": 53960.4025,
      "train_tokens_per_second": 4678.799
    },
    {
      "epoch": 0.5369897959183674,
      "grad_norm": 0.6146660611071938,
      "learning_rate": 8.325004729431083e-06,
      "loss": 0.3352448225021362,
      "num_input_tokens_seen": 252767728,
      "step": 4210,
      "train_runtime": 54027.2317,
      "train_tokens_per_second": 4678.525
    },
    {
      "epoch": 0.5376275510204082,
      "grad_norm": 0.5967061292516193,
      "learning_rate": 8.32126218955641e-06,
      "loss": 0.30342822074890136,
      "num_input_tokens_seen": 253064680,
      "step": 4215,
      "train_runtime": 54082.7429,
      "train_tokens_per_second": 4679.213
    },
    {
      "epoch": 0.5382653061224489,
      "grad_norm": 0.5643048497659794,
      "learning_rate": 8.317516316563172e-06,
      "loss": 0.3140439510345459,
      "num_input_tokens_seen": 253377952,
      "step": 4220,
      "train_runtime": 54143.3604,
      "train_tokens_per_second": 4679.76
    },
    {
      "epoch": 0.5389030612244898,
      "grad_norm": 0.6708561525924382,
      "learning_rate": 8.313767114210615e-06,
      "loss": 0.3102859020233154,
      "num_input_tokens_seen": 253651992,
      "step": 4225,
      "train_runtime": 54208.0086,
      "train_tokens_per_second": 4679.235
    },
    {
      "epoch": 0.5395408163265306,
      "grad_norm": 0.6462374542167583,
      "learning_rate": 8.310014586261329e-06,
      "loss": 0.3332800388336182,
      "num_input_tokens_seen": 253945136,
      "step": 4230,
      "train_runtime": 54273.4829,
      "train_tokens_per_second": 4678.991
    },
    {
      "epoch": 0.5401785714285714,
      "grad_norm": 0.6087934695641831,
      "learning_rate": 8.306258736481231e-06,
      "loss": 0.30633761882781985,
      "num_input_tokens_seen": 254258168,
      "step": 4235,
      "train_runtime": 54327.249,
      "train_tokens_per_second": 4680.122
    },
    {
      "epoch": 0.5408163265306123,
      "grad_norm": 0.581317239680331,
      "learning_rate": 8.302499568639583e-06,
      "loss": 0.33106720447540283,
      "num_input_tokens_seen": 254560128,
      "step": 4240,
      "train_runtime": 54385.9692,
      "train_tokens_per_second": 4680.621
    },
    {
      "epoch": 0.5414540816326531,
      "grad_norm": 0.5832690040111503,
      "learning_rate": 8.298737086508973e-06,
      "loss": 0.34075093269348145,
      "num_input_tokens_seen": 254854240,
      "step": 4245,
      "train_runtime": 54452.3276,
      "train_tokens_per_second": 4680.319
    },
    {
      "epoch": 0.5420918367346939,
      "grad_norm": 0.6479127787282017,
      "learning_rate": 8.294971293865315e-06,
      "loss": 0.3228892803192139,
      "num_input_tokens_seen": 255160192,
      "step": 4250,
      "train_runtime": 54502.7329,
      "train_tokens_per_second": 4681.604
    },
    {
      "epoch": 0.5427295918367347,
      "grad_norm": 0.549889966578858,
      "learning_rate": 8.29120219448784e-06,
      "loss": 0.2891772031784058,
      "num_input_tokens_seen": 255452128,
      "step": 4255,
      "train_runtime": 54577.2321,
      "train_tokens_per_second": 4680.562
    },
    {
      "epoch": 0.5433673469387755,
      "grad_norm": 0.6029598755497491,
      "learning_rate": 8.287429792159108e-06,
      "loss": 0.30979280471801757,
      "num_input_tokens_seen": 255744456,
      "step": 4260,
      "train_runtime": 54636.473,
      "train_tokens_per_second": 4680.838
    },
    {
      "epoch": 0.5440051020408163,
      "grad_norm": 0.5916236217514762,
      "learning_rate": 8.283654090664986e-06,
      "loss": 0.3161670207977295,
      "num_input_tokens_seen": 256054536,
      "step": 4265,
      "train_runtime": 54698.8425,
      "train_tokens_per_second": 4681.169
    },
    {
      "epoch": 0.5446428571428571,
      "grad_norm": 0.6367433641413873,
      "learning_rate": 8.279875093794652e-06,
      "loss": 0.30004215240478516,
      "num_input_tokens_seen": 256333248,
      "step": 4270,
      "train_runtime": 54769.6234,
      "train_tokens_per_second": 4680.208
    },
    {
      "epoch": 0.545280612244898,
      "grad_norm": 0.5042579518790996,
      "learning_rate": 8.276092805340596e-06,
      "loss": 0.29477152824401853,
      "num_input_tokens_seen": 256626592,
      "step": 4275,
      "train_runtime": 54842.3787,
      "train_tokens_per_second": 4679.348
    },
    {
      "epoch": 0.5459183673469388,
      "grad_norm": 1.1128245768635054,
      "learning_rate": 8.27230722909861e-06,
      "loss": 0.32368955612182615,
      "num_input_tokens_seen": 256933704,
      "step": 4280,
      "train_runtime": 54917.0586,
      "train_tokens_per_second": 4678.577
    },
    {
      "epoch": 0.5465561224489796,
      "grad_norm": 0.5968864271196218,
      "learning_rate": 8.268518368867781e-06,
      "loss": 0.3586747646331787,
      "num_input_tokens_seen": 257231712,
      "step": 4285,
      "train_runtime": 54977.612,
      "train_tokens_per_second": 4678.845
    },
    {
      "epoch": 0.5471938775510204,
      "grad_norm": 0.5998615928387417,
      "learning_rate": 8.264726228450495e-06,
      "loss": 0.3417146444320679,
      "num_input_tokens_seen": 257525104,
      "step": 4290,
      "train_runtime": 55031.7486,
      "train_tokens_per_second": 4679.573
    },
    {
      "epoch": 0.5478316326530612,
      "grad_norm": 0.6128832260840915,
      "learning_rate": 8.260930811652433e-06,
      "loss": 0.3421389102935791,
      "num_input_tokens_seen": 257824568,
      "step": 4295,
      "train_runtime": 55093.1213,
      "train_tokens_per_second": 4679.796
    },
    {
      "epoch": 0.548469387755102,
      "grad_norm": 0.6418323436502023,
      "learning_rate": 8.257132122282557e-06,
      "loss": 0.3121157646179199,
      "num_input_tokens_seen": 258118576,
      "step": 4300,
      "train_runtime": 55152.2017,
      "train_tokens_per_second": 4680.114
    },
    {
      "epoch": 0.5491071428571429,
      "grad_norm": 0.5842052529896943,
      "learning_rate": 8.253330164153118e-06,
      "loss": 0.30317847728729247,
      "num_input_tokens_seen": 258433376,
      "step": 4305,
      "train_runtime": 55213.9357,
      "train_tokens_per_second": 4680.582
    },
    {
      "epoch": 0.5497448979591837,
      "grad_norm": 0.5719238330795828,
      "learning_rate": 8.24952494107965e-06,
      "loss": 0.3032562255859375,
      "num_input_tokens_seen": 258723184,
      "step": 4310,
      "train_runtime": 55289.11,
      "train_tokens_per_second": 4679.46
    },
    {
      "epoch": 0.5503826530612245,
      "grad_norm": 0.5999036755436128,
      "learning_rate": 8.245716456880954e-06,
      "loss": 0.32174317836761473,
      "num_input_tokens_seen": 259015328,
      "step": 4315,
      "train_runtime": 55358.7971,
      "train_tokens_per_second": 4678.847
    },
    {
      "epoch": 0.5510204081632653,
      "grad_norm": 0.6097436452809384,
      "learning_rate": 8.241904715379115e-06,
      "loss": 0.3356198310852051,
      "num_input_tokens_seen": 259311400,
      "step": 4320,
      "train_runtime": 55419.4483,
      "train_tokens_per_second": 4679.069
    },
    {
      "epoch": 0.5516581632653061,
      "grad_norm": 0.6063697057688631,
      "learning_rate": 8.23808972039948e-06,
      "loss": 0.3257925033569336,
      "num_input_tokens_seen": 259601608,
      "step": 4325,
      "train_runtime": 55487.9113,
      "train_tokens_per_second": 4678.526
    },
    {
      "epoch": 0.5522959183673469,
      "grad_norm": 0.5822107839726435,
      "learning_rate": 8.234271475770662e-06,
      "loss": 0.3142188549041748,
      "num_input_tokens_seen": 259907824,
      "step": 4330,
      "train_runtime": 55564.8605,
      "train_tokens_per_second": 4677.557
    },
    {
      "epoch": 0.5529336734693877,
      "grad_norm": 0.5330086736550768,
      "learning_rate": 8.230449985324538e-06,
      "loss": 0.3134969711303711,
      "num_input_tokens_seen": 260201456,
      "step": 4335,
      "train_runtime": 55629.7575,
      "train_tokens_per_second": 4677.379
    },
    {
      "epoch": 0.5535714285714286,
      "grad_norm": 0.5481024736599419,
      "learning_rate": 8.226625252896239e-06,
      "loss": 0.32528071403503417,
      "num_input_tokens_seen": 260510920,
      "step": 4340,
      "train_runtime": 55695.1674,
      "train_tokens_per_second": 4677.442
    },
    {
      "epoch": 0.5542091836734694,
      "grad_norm": 0.7373651181598547,
      "learning_rate": 8.222797282324152e-06,
      "loss": 0.34186592102050783,
      "num_input_tokens_seen": 260802744,
      "step": 4345,
      "train_runtime": 55766.6206,
      "train_tokens_per_second": 4676.682
    },
    {
      "epoch": 0.5548469387755102,
      "grad_norm": 0.5905825099069442,
      "learning_rate": 8.218966077449912e-06,
      "loss": 0.31566557884216306,
      "num_input_tokens_seen": 261105232,
      "step": 4350,
      "train_runtime": 55827.3125,
      "train_tokens_per_second": 4677.016
    },
    {
      "epoch": 0.5554846938775511,
      "grad_norm": 0.6507514232075635,
      "learning_rate": 8.215131642118401e-06,
      "loss": 0.3226027488708496,
      "num_input_tokens_seen": 261415160,
      "step": 4355,
      "train_runtime": 55893.593,
      "train_tokens_per_second": 4677.015
    },
    {
      "epoch": 0.5561224489795918,
      "grad_norm": 0.6227011582604052,
      "learning_rate": 8.211293980177744e-06,
      "loss": 0.2847200632095337,
      "num_input_tokens_seen": 261699240,
      "step": 4360,
      "train_runtime": 55954.2652,
      "train_tokens_per_second": 4677.02
    },
    {
      "epoch": 0.5567602040816326,
      "grad_norm": 0.5688769001803791,
      "learning_rate": 8.2074530954793e-06,
      "loss": 0.2905887603759766,
      "num_input_tokens_seen": 262010480,
      "step": 4365,
      "train_runtime": 56021.1398,
      "train_tokens_per_second": 4676.993
    },
    {
      "epoch": 0.5573979591836735,
      "grad_norm": 0.6077548704289386,
      "learning_rate": 8.203608991877669e-06,
      "loss": 0.35893514156341555,
      "num_input_tokens_seen": 262308904,
      "step": 4370,
      "train_runtime": 56083.6474,
      "train_tokens_per_second": 4677.101
    },
    {
      "epoch": 0.5580357142857143,
      "grad_norm": 0.608483822828799,
      "learning_rate": 8.199761673230674e-06,
      "loss": 0.29238204956054686,
      "num_input_tokens_seen": 262616976,
      "step": 4375,
      "train_runtime": 56154.2218,
      "train_tokens_per_second": 4676.709
    },
    {
      "epoch": 0.5586734693877551,
      "grad_norm": 0.5830624962518477,
      "learning_rate": 8.195911143399368e-06,
      "loss": 0.3076601982116699,
      "num_input_tokens_seen": 262909384,
      "step": 4380,
      "train_runtime": 56221.6719,
      "train_tokens_per_second": 4676.3
    },
    {
      "epoch": 0.5593112244897959,
      "grad_norm": 0.673049126211511,
      "learning_rate": 8.192057406248028e-06,
      "loss": 0.3068440675735474,
      "num_input_tokens_seen": 263213760,
      "step": 4385,
      "train_runtime": 56295.113,
      "train_tokens_per_second": 4675.606
    },
    {
      "epoch": 0.5599489795918368,
      "grad_norm": 0.5736353161658411,
      "learning_rate": 8.18820046564415e-06,
      "loss": 0.32100448608398435,
      "num_input_tokens_seen": 263515416,
      "step": 4390,
      "train_runtime": 56361.6434,
      "train_tokens_per_second": 4675.439
    },
    {
      "epoch": 0.5605867346938775,
      "grad_norm": 0.593170776988159,
      "learning_rate": 8.184340325458439e-06,
      "loss": 0.30128042697906493,
      "num_input_tokens_seen": 263822760,
      "step": 4395,
      "train_runtime": 56427.6086,
      "train_tokens_per_second": 4675.42
    },
    {
      "epoch": 0.5612244897959183,
      "grad_norm": 0.6292006445058823,
      "learning_rate": 8.180476989564818e-06,
      "loss": 0.2959024429321289,
      "num_input_tokens_seen": 264122448,
      "step": 4400,
      "train_runtime": 56493.3571,
      "train_tokens_per_second": 4675.283
    },
    {
      "epoch": 0.5618622448979592,
      "grad_norm": 0.5821519726630247,
      "learning_rate": 8.176610461840415e-06,
      "loss": 0.3054521083831787,
      "num_input_tokens_seen": 264436904,
      "step": 4405,
      "train_runtime": 56557.7219,
      "train_tokens_per_second": 4675.523
    },
    {
      "epoch": 0.5625,
      "grad_norm": 0.668053801557447,
      "learning_rate": 8.17274074616556e-06,
      "loss": 0.33715431690216063,
      "num_input_tokens_seen": 264738224,
      "step": 4410,
      "train_runtime": 56620.5839,
      "train_tokens_per_second": 4675.653
    },
    {
      "epoch": 0.5631377551020408,
      "grad_norm": 0.6303997378853025,
      "learning_rate": 8.16886784642378e-06,
      "loss": 0.31183786392211915,
      "num_input_tokens_seen": 265031064,
      "step": 4415,
      "train_runtime": 56686.7773,
      "train_tokens_per_second": 4675.36
    },
    {
      "epoch": 0.5637755102040817,
      "grad_norm": 0.5962982226170522,
      "learning_rate": 8.164991766501804e-06,
      "loss": 0.3513195037841797,
      "num_input_tokens_seen": 265310864,
      "step": 4420,
      "train_runtime": 56744.2799,
      "train_tokens_per_second": 4675.553
    },
    {
      "epoch": 0.5644132653061225,
      "grad_norm": 0.6182195185195032,
      "learning_rate": 8.16111251028955e-06,
      "loss": 0.31078166961669923,
      "num_input_tokens_seen": 265611584,
      "step": 4425,
      "train_runtime": 56807.1546,
      "train_tokens_per_second": 4675.671
    },
    {
      "epoch": 0.5650510204081632,
      "grad_norm": 0.4950459126655072,
      "learning_rate": 8.157230081680118e-06,
      "loss": 0.31040363311767577,
      "num_input_tokens_seen": 265908936,
      "step": 4430,
      "train_runtime": 56878.4901,
      "train_tokens_per_second": 4675.035
    },
    {
      "epoch": 0.5656887755102041,
      "grad_norm": 0.625629747375622,
      "learning_rate": 8.153344484569798e-06,
      "loss": 0.33584554195404054,
      "num_input_tokens_seen": 266214216,
      "step": 4435,
      "train_runtime": 56944.8151,
      "train_tokens_per_second": 4674.951
    },
    {
      "epoch": 0.5663265306122449,
      "grad_norm": 0.5575423240531477,
      "learning_rate": 8.14945572285806e-06,
      "loss": 0.2880943775177002,
      "num_input_tokens_seen": 266521944,
      "step": 4440,
      "train_runtime": 57011.7003,
      "train_tokens_per_second": 4674.864
    },
    {
      "epoch": 0.5669642857142857,
      "grad_norm": 0.6377758398703642,
      "learning_rate": 8.145563800447547e-06,
      "loss": 0.32821414470672605,
      "num_input_tokens_seen": 266821112,
      "step": 4445,
      "train_runtime": 57080.3068,
      "train_tokens_per_second": 4674.486
    },
    {
      "epoch": 0.5676020408163265,
      "grad_norm": 0.6990364683606246,
      "learning_rate": 8.141668721244073e-06,
      "loss": 0.33448290824890137,
      "num_input_tokens_seen": 267110056,
      "step": 4450,
      "train_runtime": 57135.8468,
      "train_tokens_per_second": 4675.0
    },
    {
      "epoch": 0.5682397959183674,
      "grad_norm": 0.5582537167645563,
      "learning_rate": 8.137770489156624e-06,
      "loss": 0.32361855506896975,
      "num_input_tokens_seen": 267406768,
      "step": 4455,
      "train_runtime": 57201.0749,
      "train_tokens_per_second": 4674.856
    },
    {
      "epoch": 0.5688775510204082,
      "grad_norm": 0.6190981061148347,
      "learning_rate": 8.133869108097349e-06,
      "loss": 0.3138890743255615,
      "num_input_tokens_seen": 267703584,
      "step": 4460,
      "train_runtime": 57268.5972,
      "train_tokens_per_second": 4674.527
    },
    {
      "epoch": 0.5695153061224489,
      "grad_norm": 0.5847490108749284,
      "learning_rate": 8.129964581981554e-06,
      "loss": 0.32745659351348877,
      "num_input_tokens_seen": 267999824,
      "step": 4465,
      "train_runtime": 57337.5739,
      "train_tokens_per_second": 4674.07
    },
    {
      "epoch": 0.5701530612244898,
      "grad_norm": 0.5755116016825046,
      "learning_rate": 8.126056914727705e-06,
      "loss": 0.32536706924438474,
      "num_input_tokens_seen": 268308056,
      "step": 4470,
      "train_runtime": 57396.0366,
      "train_tokens_per_second": 4674.679
    },
    {
      "epoch": 0.5707908163265306,
      "grad_norm": 0.6154693294509455,
      "learning_rate": 8.122146110257419e-06,
      "loss": 0.31456241607666013,
      "num_input_tokens_seen": 268596848,
      "step": 4475,
      "train_runtime": 57462.4776,
      "train_tokens_per_second": 4674.3
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.5502179221099155,
      "learning_rate": 8.11823217249546e-06,
      "loss": 0.310486888885498,
      "num_input_tokens_seen": 268895872,
      "step": 4480,
      "train_runtime": 57524.0889,
      "train_tokens_per_second": 4674.492
    },
    {
      "epoch": 0.5720663265306123,
      "grad_norm": 0.5802016769840421,
      "learning_rate": 8.114315105369739e-06,
      "loss": 0.30334928035736086,
      "num_input_tokens_seen": 269201352,
      "step": 4485,
      "train_runtime": 57586.6413,
      "train_tokens_per_second": 4674.719
    },
    {
      "epoch": 0.5727040816326531,
      "grad_norm": 0.5883515183462985,
      "learning_rate": 8.110394912811308e-06,
      "loss": 0.30872986316680906,
      "num_input_tokens_seen": 269491536,
      "step": 4490,
      "train_runtime": 57656.0337,
      "train_tokens_per_second": 4674.125
    },
    {
      "epoch": 0.5733418367346939,
      "grad_norm": 0.649324643030826,
      "learning_rate": 8.106471598754353e-06,
      "loss": 0.33896265029907224,
      "num_input_tokens_seen": 269792872,
      "step": 4495,
      "train_runtime": 57712.569,
      "train_tokens_per_second": 4674.768
    },
    {
      "epoch": 0.5739795918367347,
      "grad_norm": 0.627166759763884,
      "learning_rate": 8.102545167136191e-06,
      "loss": 0.3339379787445068,
      "num_input_tokens_seen": 270112504,
      "step": 4500,
      "train_runtime": 57770.4755,
      "train_tokens_per_second": 4675.615
    },
    {
      "epoch": 0.5746173469387755,
      "grad_norm": 0.5896976553712885,
      "learning_rate": 8.098615621897272e-06,
      "loss": 0.34661197662353516,
      "num_input_tokens_seen": 270419208,
      "step": 4505,
      "train_runtime": 57835.7103,
      "train_tokens_per_second": 4675.644
    },
    {
      "epoch": 0.5752551020408163,
      "grad_norm": 0.5560755207064892,
      "learning_rate": 8.094682966981172e-06,
      "loss": 0.32341957092285156,
      "num_input_tokens_seen": 270709976,
      "step": 4510,
      "train_runtime": 57899.4109,
      "train_tokens_per_second": 4675.522
    },
    {
      "epoch": 0.5758928571428571,
      "grad_norm": 0.6455684435641715,
      "learning_rate": 8.090747206334582e-06,
      "loss": 0.3314667224884033,
      "num_input_tokens_seen": 270992920,
      "step": 4515,
      "train_runtime": 57957.1139,
      "train_tokens_per_second": 4675.749
    },
    {
      "epoch": 0.576530612244898,
      "grad_norm": 0.5887125092425513,
      "learning_rate": 8.086808343907313e-06,
      "loss": 0.3472761869430542,
      "num_input_tokens_seen": 271290696,
      "step": 4520,
      "train_runtime": 58015.5676,
      "train_tokens_per_second": 4676.171
    },
    {
      "epoch": 0.5771683673469388,
      "grad_norm": 0.6918344194685301,
      "learning_rate": 8.082866383652288e-06,
      "loss": 0.35142107009887696,
      "num_input_tokens_seen": 271602944,
      "step": 4525,
      "train_runtime": 58073.2704,
      "train_tokens_per_second": 4676.901
    },
    {
      "epoch": 0.5778061224489796,
      "grad_norm": 0.5656574815084812,
      "learning_rate": 8.07892132952554e-06,
      "loss": 0.31038622856140136,
      "num_input_tokens_seen": 271903088,
      "step": 4530,
      "train_runtime": 58129.2517,
      "train_tokens_per_second": 4677.56
    },
    {
      "epoch": 0.5784438775510204,
      "grad_norm": 0.5842917315653896,
      "learning_rate": 8.074973185486206e-06,
      "loss": 0.30137619972229,
      "num_input_tokens_seen": 272210992,
      "step": 4535,
      "train_runtime": 58188.5757,
      "train_tokens_per_second": 4678.083
    },
    {
      "epoch": 0.5790816326530612,
      "grad_norm": 0.5949065684239946,
      "learning_rate": 8.071021955496526e-06,
      "loss": 0.32496907711029055,
      "num_input_tokens_seen": 272530424,
      "step": 4540,
      "train_runtime": 58258.679,
      "train_tokens_per_second": 4677.937
    },
    {
      "epoch": 0.579719387755102,
      "grad_norm": 0.6225068609990876,
      "learning_rate": 8.067067643521834e-06,
      "loss": 0.3328864574432373,
      "num_input_tokens_seen": 272843728,
      "step": 4545,
      "train_runtime": 58321.4303,
      "train_tokens_per_second": 4678.276
    },
    {
      "epoch": 0.5803571428571429,
      "grad_norm": 0.5864685523092161,
      "learning_rate": 8.063110253530557e-06,
      "loss": 0.3262446403503418,
      "num_input_tokens_seen": 273157344,
      "step": 4550,
      "train_runtime": 58386.701,
      "train_tokens_per_second": 4678.417
    },
    {
      "epoch": 0.5809948979591837,
      "grad_norm": 0.5987457013718706,
      "learning_rate": 8.059149789494215e-06,
      "loss": 0.3453367233276367,
      "num_input_tokens_seen": 273462168,
      "step": 4555,
      "train_runtime": 58454.6399,
      "train_tokens_per_second": 4678.194
    },
    {
      "epoch": 0.5816326530612245,
      "grad_norm": 0.5917895077389059,
      "learning_rate": 8.055186255387409e-06,
      "loss": 0.326198148727417,
      "num_input_tokens_seen": 273774352,
      "step": 4560,
      "train_runtime": 58517.3721,
      "train_tokens_per_second": 4678.514
    },
    {
      "epoch": 0.5822704081632653,
      "grad_norm": 0.6515604444187467,
      "learning_rate": 8.051219655187818e-06,
      "loss": 0.32740585803985595,
      "num_input_tokens_seen": 274066568,
      "step": 4565,
      "train_runtime": 58586.4292,
      "train_tokens_per_second": 4677.987
    },
    {
      "epoch": 0.5829081632653061,
      "grad_norm": 0.5795418883268696,
      "learning_rate": 8.04724999287621e-06,
      "loss": 0.3357175350189209,
      "num_input_tokens_seen": 274380280,
      "step": 4570,
      "train_runtime": 58643.7348,
      "train_tokens_per_second": 4678.765
    },
    {
      "epoch": 0.5835459183673469,
      "grad_norm": 0.6632735372633155,
      "learning_rate": 8.043277272436415e-06,
      "loss": 0.3141871452331543,
      "num_input_tokens_seen": 274676088,
      "step": 4575,
      "train_runtime": 58708.7292,
      "train_tokens_per_second": 4678.624
    },
    {
      "epoch": 0.5841836734693877,
      "grad_norm": 0.54087084864485,
      "learning_rate": 8.039301497855331e-06,
      "loss": 0.33591418266296386,
      "num_input_tokens_seen": 274981048,
      "step": 4580,
      "train_runtime": 58762.902,
      "train_tokens_per_second": 4679.501
    },
    {
      "epoch": 0.5848214285714286,
      "grad_norm": 0.6427792175214475,
      "learning_rate": 8.035322673122934e-06,
      "loss": 0.3060802459716797,
      "num_input_tokens_seen": 275282680,
      "step": 4585,
      "train_runtime": 58817.932,
      "train_tokens_per_second": 4680.251
    },
    {
      "epoch": 0.5854591836734694,
      "grad_norm": 0.6031025817377057,
      "learning_rate": 8.031340802232246e-06,
      "loss": 0.33022494316101075,
      "num_input_tokens_seen": 275578376,
      "step": 4590,
      "train_runtime": 58886.4379,
      "train_tokens_per_second": 4679.828
    },
    {
      "epoch": 0.5860969387755102,
      "grad_norm": 0.5225939761001044,
      "learning_rate": 8.027355889179356e-06,
      "loss": 0.35036659240722656,
      "num_input_tokens_seen": 275890552,
      "step": 4595,
      "train_runtime": 58956.8332,
      "train_tokens_per_second": 4679.535
    },
    {
      "epoch": 0.5867346938775511,
      "grad_norm": 0.6038568442283259,
      "learning_rate": 8.0233679379634e-06,
      "loss": 0.3158730506896973,
      "num_input_tokens_seen": 276186592,
      "step": 4600,
      "train_runtime": 59016.9952,
      "train_tokens_per_second": 4679.781
    },
    {
      "epoch": 0.5873724489795918,
      "grad_norm": 0.5418728232199929,
      "learning_rate": 8.019376952586567e-06,
      "loss": 0.30447533130645754,
      "num_input_tokens_seen": 276493112,
      "step": 4605,
      "train_runtime": 59087.1949,
      "train_tokens_per_second": 4679.408
    },
    {
      "epoch": 0.5880102040816326,
      "grad_norm": 0.6604770163281342,
      "learning_rate": 8.01538293705409e-06,
      "loss": 0.29389081001281736,
      "num_input_tokens_seen": 276778280,
      "step": 4610,
      "train_runtime": 59156.0214,
      "train_tokens_per_second": 4678.785
    },
    {
      "epoch": 0.5886479591836735,
      "grad_norm": 0.5910960639532645,
      "learning_rate": 8.011385895374239e-06,
      "loss": 0.31752629280090333,
      "num_input_tokens_seen": 277079264,
      "step": 4615,
      "train_runtime": 59217.2144,
      "train_tokens_per_second": 4679.032
    },
    {
      "epoch": 0.5892857142857143,
      "grad_norm": 0.5874338866486604,
      "learning_rate": 8.007385831558329e-06,
      "loss": 0.3469626665115356,
      "num_input_tokens_seen": 277388912,
      "step": 4620,
      "train_runtime": 59285.0545,
      "train_tokens_per_second": 4678.901
    },
    {
      "epoch": 0.5899234693877551,
      "grad_norm": 0.6283317385335976,
      "learning_rate": 8.003382749620704e-06,
      "loss": 0.3177196741104126,
      "num_input_tokens_seen": 277684968,
      "step": 4625,
      "train_runtime": 59352.4811,
      "train_tokens_per_second": 4678.574
    },
    {
      "epoch": 0.5905612244897959,
      "grad_norm": 0.6337329309218752,
      "learning_rate": 7.99937665357873e-06,
      "loss": 0.3333481788635254,
      "num_input_tokens_seen": 277989152,
      "step": 4630,
      "train_runtime": 59424.6485,
      "train_tokens_per_second": 4678.011
    },
    {
      "epoch": 0.5911989795918368,
      "grad_norm": 0.6077936632271252,
      "learning_rate": 7.995367547452811e-06,
      "loss": 0.29072420597076415,
      "num_input_tokens_seen": 278272968,
      "step": 4635,
      "train_runtime": 59483.0749,
      "train_tokens_per_second": 4678.187
    },
    {
      "epoch": 0.5918367346938775,
      "grad_norm": 0.615433923915632,
      "learning_rate": 7.991355435266362e-06,
      "loss": 0.2903087615966797,
      "num_input_tokens_seen": 278571200,
      "step": 4640,
      "train_runtime": 59545.8982,
      "train_tokens_per_second": 4678.26
    },
    {
      "epoch": 0.5924744897959183,
      "grad_norm": 0.6792375997945723,
      "learning_rate": 7.987340321045816e-06,
      "loss": 0.30604982376098633,
      "num_input_tokens_seen": 278866808,
      "step": 4645,
      "train_runtime": 59618.4852,
      "train_tokens_per_second": 4677.523
    },
    {
      "epoch": 0.5931122448979592,
      "grad_norm": 0.6909830234790262,
      "learning_rate": 7.983322208820624e-06,
      "loss": 0.32366094589233396,
      "num_input_tokens_seen": 279147960,
      "step": 4650,
      "train_runtime": 59677.0057,
      "train_tokens_per_second": 4677.647
    },
    {
      "epoch": 0.59375,
      "grad_norm": 0.5682240529995807,
      "learning_rate": 7.97930110262324e-06,
      "loss": 0.30706181526184084,
      "num_input_tokens_seen": 279457336,
      "step": 4655,
      "train_runtime": 59747.2249,
      "train_tokens_per_second": 4677.327
    },
    {
      "epoch": 0.5943877551020408,
      "grad_norm": 0.5855001978640996,
      "learning_rate": 7.975277006489126e-06,
      "loss": 0.31863577365875245,
      "num_input_tokens_seen": 279739376,
      "step": 4660,
      "train_runtime": 59808.0774,
      "train_tokens_per_second": 4677.284
    },
    {
      "epoch": 0.5950255102040817,
      "grad_norm": 0.6283309046222093,
      "learning_rate": 7.971249924456742e-06,
      "loss": 0.3421057462692261,
      "num_input_tokens_seen": 280055896,
      "step": 4665,
      "train_runtime": 59876.8153,
      "train_tokens_per_second": 4677.201
    },
    {
      "epoch": 0.5956632653061225,
      "grad_norm": 0.6852848428090714,
      "learning_rate": 7.967219860567547e-06,
      "loss": 0.3344215631484985,
      "num_input_tokens_seen": 280344680,
      "step": 4670,
      "train_runtime": 59939.7728,
      "train_tokens_per_second": 4677.106
    },
    {
      "epoch": 0.5963010204081632,
      "grad_norm": 0.5976999232776021,
      "learning_rate": 7.963186818865993e-06,
      "loss": 0.31830806732177735,
      "num_input_tokens_seen": 280641200,
      "step": 4675,
      "train_runtime": 59997.9658,
      "train_tokens_per_second": 4677.512
    },
    {
      "epoch": 0.5969387755102041,
      "grad_norm": 0.6595538383558863,
      "learning_rate": 7.959150803399519e-06,
      "loss": 0.3340503215789795,
      "num_input_tokens_seen": 280944136,
      "step": 4680,
      "train_runtime": 60073.8834,
      "train_tokens_per_second": 4676.643
    },
    {
      "epoch": 0.5975765306122449,
      "grad_norm": 0.5490632590212808,
      "learning_rate": 7.955111818218544e-06,
      "loss": 0.2980647563934326,
      "num_input_tokens_seen": 281248600,
      "step": 4685,
      "train_runtime": 60131.2907,
      "train_tokens_per_second": 4677.242
    },
    {
      "epoch": 0.5982142857142857,
      "grad_norm": 0.6775033404877127,
      "learning_rate": 7.951069867376477e-06,
      "loss": 0.335029673576355,
      "num_input_tokens_seen": 281561744,
      "step": 4690,
      "train_runtime": 60195.349,
      "train_tokens_per_second": 4677.467
    },
    {
      "epoch": 0.5988520408163265,
      "grad_norm": 0.6153161340577447,
      "learning_rate": 7.947024954929696e-06,
      "loss": 0.3289950370788574,
      "num_input_tokens_seen": 281867880,
      "step": 4695,
      "train_runtime": 60271.5282,
      "train_tokens_per_second": 4676.634
    },
    {
      "epoch": 0.5994897959183674,
      "grad_norm": 0.6146139115004825,
      "learning_rate": 7.942977084937552e-06,
      "loss": 0.3104766845703125,
      "num_input_tokens_seen": 282165104,
      "step": 4700,
      "train_runtime": 60344.3141,
      "train_tokens_per_second": 4675.919
    },
    {
      "epoch": 0.6001275510204082,
      "grad_norm": 0.6597846232204004,
      "learning_rate": 7.938926261462366e-06,
      "loss": 0.32243638038635253,
      "num_input_tokens_seen": 282447688,
      "step": 4705,
      "train_runtime": 60408.1953,
      "train_tokens_per_second": 4675.652
    },
    {
      "epoch": 0.6007653061224489,
      "grad_norm": 0.6147432109350911,
      "learning_rate": 7.934872488569423e-06,
      "loss": 0.328096604347229,
      "num_input_tokens_seen": 282739992,
      "step": 4710,
      "train_runtime": 60473.6074,
      "train_tokens_per_second": 4675.428
    },
    {
      "epoch": 0.6014030612244898,
      "grad_norm": 0.6242896507087882,
      "learning_rate": 7.930815770326968e-06,
      "loss": 0.3395079135894775,
      "num_input_tokens_seen": 283026552,
      "step": 4715,
      "train_runtime": 60528.8065,
      "train_tokens_per_second": 4675.898
    },
    {
      "epoch": 0.6020408163265306,
      "grad_norm": 0.6249277645914738,
      "learning_rate": 7.926756110806197e-06,
      "loss": 0.3366846084594727,
      "num_input_tokens_seen": 283322448,
      "step": 4720,
      "train_runtime": 60592.3677,
      "train_tokens_per_second": 4675.877
    },
    {
      "epoch": 0.6026785714285714,
      "grad_norm": 0.629573578510475,
      "learning_rate": 7.922693514081267e-06,
      "loss": 0.33675575256347656,
      "num_input_tokens_seen": 283632392,
      "step": 4725,
      "train_runtime": 60668.4159,
      "train_tokens_per_second": 4675.124
    },
    {
      "epoch": 0.6033163265306123,
      "grad_norm": 0.6909843984830276,
      "learning_rate": 7.918627984229274e-06,
      "loss": 0.3276346206665039,
      "num_input_tokens_seen": 283930896,
      "step": 4730,
      "train_runtime": 60736.3279,
      "train_tokens_per_second": 4674.812
    },
    {
      "epoch": 0.6039540816326531,
      "grad_norm": 0.556604652718109,
      "learning_rate": 7.914559525330262e-06,
      "loss": 0.30612616539001464,
      "num_input_tokens_seen": 284233336,
      "step": 4735,
      "train_runtime": 60810.0437,
      "train_tokens_per_second": 4674.118
    },
    {
      "epoch": 0.6045918367346939,
      "grad_norm": 0.5489958640080462,
      "learning_rate": 7.910488141467215e-06,
      "loss": 0.30764472484588623,
      "num_input_tokens_seen": 284535840,
      "step": 4740,
      "train_runtime": 60870.9018,
      "train_tokens_per_second": 4674.415
    },
    {
      "epoch": 0.6052295918367347,
      "grad_norm": 0.6349237008579836,
      "learning_rate": 7.906413836726049e-06,
      "loss": 0.3160310029983521,
      "num_input_tokens_seen": 284842016,
      "step": 4745,
      "train_runtime": 60928.9895,
      "train_tokens_per_second": 4674.983
    },
    {
      "epoch": 0.6058673469387755,
      "grad_norm": 0.663809716098464,
      "learning_rate": 7.902336615195614e-06,
      "loss": 0.30224664211273194,
      "num_input_tokens_seen": 285134624,
      "step": 4750,
      "train_runtime": 60986.9207,
      "train_tokens_per_second": 4675.341
    },
    {
      "epoch": 0.6065051020408163,
      "grad_norm": 0.6536914593460793,
      "learning_rate": 7.89825648096769e-06,
      "loss": 0.3554865837097168,
      "num_input_tokens_seen": 285448544,
      "step": 4755,
      "train_runtime": 61046.5608,
      "train_tokens_per_second": 4675.915
    },
    {
      "epoch": 0.6071428571428571,
      "grad_norm": 0.5566778191411491,
      "learning_rate": 7.89417343813697e-06,
      "loss": 0.30611572265625,
      "num_input_tokens_seen": 285755440,
      "step": 4760,
      "train_runtime": 61122.2035,
      "train_tokens_per_second": 4675.15
    },
    {
      "epoch": 0.607780612244898,
      "grad_norm": 0.5449701216835566,
      "learning_rate": 7.890087490801077e-06,
      "loss": 0.30018014907836915,
      "num_input_tokens_seen": 286052824,
      "step": 4765,
      "train_runtime": 61185.6008,
      "train_tokens_per_second": 4675.166
    },
    {
      "epoch": 0.6084183673469388,
      "grad_norm": 0.6464923010268058,
      "learning_rate": 7.885998643060544e-06,
      "loss": 0.2942426443099976,
      "num_input_tokens_seen": 286352088,
      "step": 4770,
      "train_runtime": 61244.7406,
      "train_tokens_per_second": 4675.538
    },
    {
      "epoch": 0.6090561224489796,
      "grad_norm": 0.5962819806881094,
      "learning_rate": 7.881906899018815e-06,
      "loss": 0.3343794822692871,
      "num_input_tokens_seen": 286647208,
      "step": 4775,
      "train_runtime": 61310.5843,
      "train_tokens_per_second": 4675.33
    },
    {
      "epoch": 0.6096938775510204,
      "grad_norm": 0.5805268901725466,
      "learning_rate": 7.87781226278224e-06,
      "loss": 0.30678706169128417,
      "num_input_tokens_seen": 286937328,
      "step": 4780,
      "train_runtime": 61362.5036,
      "train_tokens_per_second": 4676.102
    },
    {
      "epoch": 0.6103316326530612,
      "grad_norm": 0.6437899436315898,
      "learning_rate": 7.873714738460075e-06,
      "loss": 0.3248507022857666,
      "num_input_tokens_seen": 287226808,
      "step": 4785,
      "train_runtime": 61443.8801,
      "train_tokens_per_second": 4674.62
    },
    {
      "epoch": 0.610969387755102,
      "grad_norm": 0.5852781947619063,
      "learning_rate": 7.86961433016447e-06,
      "loss": 0.32093214988708496,
      "num_input_tokens_seen": 287538528,
      "step": 4790,
      "train_runtime": 61516.2233,
      "train_tokens_per_second": 4674.19
    },
    {
      "epoch": 0.6116071428571429,
      "grad_norm": 0.6754745796276539,
      "learning_rate": 7.865511042010469e-06,
      "loss": 0.3276777505874634,
      "num_input_tokens_seen": 287840048,
      "step": 4795,
      "train_runtime": 61583.222,
      "train_tokens_per_second": 4674.001
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 0.626323586120589,
      "learning_rate": 7.861404878116014e-06,
      "loss": 0.3292057037353516,
      "num_input_tokens_seen": 288140544,
      "step": 4800,
      "train_runtime": 61637.4782,
      "train_tokens_per_second": 4674.762
    },
    {
      "epoch": 0.6128826530612245,
      "grad_norm": 0.7520627783549967,
      "learning_rate": 7.857295842601921e-06,
      "loss": 0.3104045867919922,
      "num_input_tokens_seen": 288449832,
      "step": 4805,
      "train_runtime": 61692.264,
      "train_tokens_per_second": 4675.624
    },
    {
      "epoch": 0.6135204081632653,
      "grad_norm": 0.5664426864627804,
      "learning_rate": 7.853183939591898e-06,
      "loss": 0.3004433631896973,
      "num_input_tokens_seen": 288740632,
      "step": 4810,
      "train_runtime": 61750.7635,
      "train_tokens_per_second": 4675.904
    },
    {
      "epoch": 0.6141581632653061,
      "grad_norm": 0.600372992492814,
      "learning_rate": 7.849069173212526e-06,
      "loss": 0.30824880599975585,
      "num_input_tokens_seen": 289036488,
      "step": 4815,
      "train_runtime": 61809.6933,
      "train_tokens_per_second": 4676.232
    },
    {
      "epoch": 0.6147959183673469,
      "grad_norm": 0.6426046266226517,
      "learning_rate": 7.844951547593261e-06,
      "loss": 0.3388571262359619,
      "num_input_tokens_seen": 289337856,
      "step": 4820,
      "train_runtime": 61877.2968,
      "train_tokens_per_second": 4675.994
    },
    {
      "epoch": 0.6154336734693877,
      "grad_norm": 0.6572670897112,
      "learning_rate": 7.840831066866423e-06,
      "loss": 0.3076493740081787,
      "num_input_tokens_seen": 289627360,
      "step": 4825,
      "train_runtime": 61950.161,
      "train_tokens_per_second": 4675.167
    },
    {
      "epoch": 0.6160714285714286,
      "grad_norm": 0.518151383501858,
      "learning_rate": 7.83670773516721e-06,
      "loss": 0.3006767749786377,
      "num_input_tokens_seen": 289937744,
      "step": 4830,
      "train_runtime": 62006.6492,
      "train_tokens_per_second": 4675.914
    },
    {
      "epoch": 0.6167091836734694,
      "grad_norm": 0.6708194572562727,
      "learning_rate": 7.83258155663367e-06,
      "loss": 0.3451414108276367,
      "num_input_tokens_seen": 290220600,
      "step": 4835,
      "train_runtime": 62076.1159,
      "train_tokens_per_second": 4675.238
    },
    {
      "epoch": 0.6173469387755102,
      "grad_norm": 0.5400032632095155,
      "learning_rate": 7.828452535406707e-06,
      "loss": 0.3213111400604248,
      "num_input_tokens_seen": 290525632,
      "step": 4840,
      "train_runtime": 62137.0817,
      "train_tokens_per_second": 4675.56
    },
    {
      "epoch": 0.6179846938775511,
      "grad_norm": 0.6390266182613682,
      "learning_rate": 7.82432067563009e-06,
      "loss": 0.32868859767913816,
      "num_input_tokens_seen": 290813016,
      "step": 4845,
      "train_runtime": 62210.2044,
      "train_tokens_per_second": 4674.684
    },
    {
      "epoch": 0.6186224489795918,
      "grad_norm": 0.5639678226521079,
      "learning_rate": 7.820185981450422e-06,
      "loss": 0.3112217903137207,
      "num_input_tokens_seen": 291116336,
      "step": 4850,
      "train_runtime": 62271.583,
      "train_tokens_per_second": 4674.947
    },
    {
      "epoch": 0.6192602040816326,
      "grad_norm": 0.5168941595517222,
      "learning_rate": 7.816048457017163e-06,
      "loss": 0.32264735698699953,
      "num_input_tokens_seen": 291425128,
      "step": 4855,
      "train_runtime": 62334.2272,
      "train_tokens_per_second": 4675.202
    },
    {
      "epoch": 0.6198979591836735,
      "grad_norm": 0.6172682257231668,
      "learning_rate": 7.811908106482601e-06,
      "loss": 0.31756374835968015,
      "num_input_tokens_seen": 291729000,
      "step": 4860,
      "train_runtime": 62392.5053,
      "train_tokens_per_second": 4675.706
    },
    {
      "epoch": 0.6205357142857143,
      "grad_norm": 0.5892438876269591,
      "learning_rate": 7.807764934001875e-06,
      "loss": 0.3277350187301636,
      "num_input_tokens_seen": 292033976,
      "step": 4865,
      "train_runtime": 62459.489,
      "train_tokens_per_second": 4675.574
    },
    {
      "epoch": 0.6211734693877551,
      "grad_norm": 0.6457575102508739,
      "learning_rate": 7.803618943732943e-06,
      "loss": 0.3173250675201416,
      "num_input_tokens_seen": 292325616,
      "step": 4870,
      "train_runtime": 62524.1033,
      "train_tokens_per_second": 4675.407
    },
    {
      "epoch": 0.6218112244897959,
      "grad_norm": 0.5758951210097139,
      "learning_rate": 7.799470139836593e-06,
      "loss": 0.31996634006500246,
      "num_input_tokens_seen": 292634232,
      "step": 4875,
      "train_runtime": 62597.9942,
      "train_tokens_per_second": 4674.818
    },
    {
      "epoch": 0.6224489795918368,
      "grad_norm": 0.5909826249235933,
      "learning_rate": 7.795318526476448e-06,
      "loss": 0.33882918357849123,
      "num_input_tokens_seen": 292931376,
      "step": 4880,
      "train_runtime": 62667.1691,
      "train_tokens_per_second": 4674.399
    },
    {
      "epoch": 0.6230867346938775,
      "grad_norm": 0.6086976638220172,
      "learning_rate": 7.791164107818933e-06,
      "loss": 0.31146893501281736,
      "num_input_tokens_seen": 293224304,
      "step": 4885,
      "train_runtime": 62743.5085,
      "train_tokens_per_second": 4673.381
    },
    {
      "epoch": 0.6237244897959183,
      "grad_norm": 0.580480051408316,
      "learning_rate": 7.787006888033304e-06,
      "loss": 0.3144942283630371,
      "num_input_tokens_seen": 293523048,
      "step": 4890,
      "train_runtime": 62808.265,
      "train_tokens_per_second": 4673.319
    },
    {
      "epoch": 0.6243622448979592,
      "grad_norm": 0.6371609435151778,
      "learning_rate": 7.782846871291619e-06,
      "loss": 0.3092595100402832,
      "num_input_tokens_seen": 293821216,
      "step": 4895,
      "train_runtime": 62870.7807,
      "train_tokens_per_second": 4673.414
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.5499171585445445,
      "learning_rate": 7.778684061768744e-06,
      "loss": 0.2991960048675537,
      "num_input_tokens_seen": 294107864,
      "step": 4900,
      "train_runtime": 62934.3929,
      "train_tokens_per_second": 4673.245
    },
    {
      "epoch": 0.6256377551020408,
      "grad_norm": 0.5501051787913646,
      "learning_rate": 7.774518463642351e-06,
      "loss": 0.30958659648895265,
      "num_input_tokens_seen": 294414864,
      "step": 4905,
      "train_runtime": 62992.8264,
      "train_tokens_per_second": 4673.784
    },
    {
      "epoch": 0.6262755102040817,
      "grad_norm": 0.6533886515282153,
      "learning_rate": 7.770350081092906e-06,
      "loss": 0.32001142501831054,
      "num_input_tokens_seen": 294714672,
      "step": 4910,
      "train_runtime": 63052.3152,
      "train_tokens_per_second": 4674.129
    },
    {
      "epoch": 0.6269132653061225,
      "grad_norm": 0.5976532218295886,
      "learning_rate": 7.766178918303675e-06,
      "loss": 0.33820013999938964,
      "num_input_tokens_seen": 295023040,
      "step": 4915,
      "train_runtime": 63110.2752,
      "train_tokens_per_second": 4674.723
    },
    {
      "epoch": 0.6275510204081632,
      "grad_norm": 0.6125342171002576,
      "learning_rate": 7.762004979460711e-06,
      "loss": 0.3244414567947388,
      "num_input_tokens_seen": 295323616,
      "step": 4920,
      "train_runtime": 63184.0344,
      "train_tokens_per_second": 4674.023
    },
    {
      "epoch": 0.6281887755102041,
      "grad_norm": 0.5924976362043458,
      "learning_rate": 7.75782826875285e-06,
      "loss": 0.3276236057281494,
      "num_input_tokens_seen": 295615896,
      "step": 4925,
      "train_runtime": 63249.4879,
      "train_tokens_per_second": 4673.807
    },
    {
      "epoch": 0.6288265306122449,
      "grad_norm": 0.6082869176992618,
      "learning_rate": 7.753648790371716e-06,
      "loss": 0.33316860198974607,
      "num_input_tokens_seen": 295919816,
      "step": 4930,
      "train_runtime": 63318.2125,
      "train_tokens_per_second": 4673.534
    },
    {
      "epoch": 0.6294642857142857,
      "grad_norm": 0.7002911402149473,
      "learning_rate": 7.749466548511706e-06,
      "loss": 0.3236359119415283,
      "num_input_tokens_seen": 296198432,
      "step": 4935,
      "train_runtime": 63378.1008,
      "train_tokens_per_second": 4673.514
    },
    {
      "epoch": 0.6301020408163265,
      "grad_norm": 0.5816219463264141,
      "learning_rate": 7.745281547369992e-06,
      "loss": 0.3121765375137329,
      "num_input_tokens_seen": 296495328,
      "step": 4940,
      "train_runtime": 63443.5318,
      "train_tokens_per_second": 4673.374
    },
    {
      "epoch": 0.6307397959183674,
      "grad_norm": 0.6022866691062141,
      "learning_rate": 7.741093791146517e-06,
      "loss": 0.3518125057220459,
      "num_input_tokens_seen": 296795576,
      "step": 4945,
      "train_runtime": 63510.0051,
      "train_tokens_per_second": 4673.21
    },
    {
      "epoch": 0.6313775510204082,
      "grad_norm": 0.5766450882516254,
      "learning_rate": 7.736903284043985e-06,
      "loss": 0.30551507472991946,
      "num_input_tokens_seen": 297096360,
      "step": 4950,
      "train_runtime": 63573.7966,
      "train_tokens_per_second": 4673.252
    },
    {
      "epoch": 0.6320153061224489,
      "grad_norm": 0.6809684572369225,
      "learning_rate": 7.73271003026786e-06,
      "loss": 0.28991546630859377,
      "num_input_tokens_seen": 297396616,
      "step": 4955,
      "train_runtime": 63641.4024,
      "train_tokens_per_second": 4673.005
    },
    {
      "epoch": 0.6326530612244898,
      "grad_norm": 0.7136593430658471,
      "learning_rate": 7.72851403402637e-06,
      "loss": 0.3405500888824463,
      "num_input_tokens_seen": 297702568,
      "step": 4960,
      "train_runtime": 63703.3727,
      "train_tokens_per_second": 4673.262
    },
    {
      "epoch": 0.6332908163265306,
      "grad_norm": 0.6915335746408366,
      "learning_rate": 7.724315299530485e-06,
      "loss": 0.3068784236907959,
      "num_input_tokens_seen": 298000528,
      "step": 4965,
      "train_runtime": 63761.4508,
      "train_tokens_per_second": 4673.679
    },
    {
      "epoch": 0.6339285714285714,
      "grad_norm": 0.578644485028179,
      "learning_rate": 7.720113830993934e-06,
      "loss": 0.2965397357940674,
      "num_input_tokens_seen": 298288560,
      "step": 4970,
      "train_runtime": 63820.1295,
      "train_tokens_per_second": 4673.895
    },
    {
      "epoch": 0.6345663265306123,
      "grad_norm": 0.6109386315094099,
      "learning_rate": 7.71590963263318e-06,
      "loss": 0.282192587852478,
      "num_input_tokens_seen": 298589976,
      "step": 4975,
      "train_runtime": 63890.578,
      "train_tokens_per_second": 4673.459
    },
    {
      "epoch": 0.6352040816326531,
      "grad_norm": 0.6330384270424209,
      "learning_rate": 7.711702708667432e-06,
      "loss": 0.30868093967437743,
      "num_input_tokens_seen": 298884408,
      "step": 4980,
      "train_runtime": 63958.552,
      "train_tokens_per_second": 4673.095
    },
    {
      "epoch": 0.6358418367346939,
      "grad_norm": 0.6484254325873305,
      "learning_rate": 7.70749306331863e-06,
      "loss": 0.32778377532958985,
      "num_input_tokens_seen": 299174816,
      "step": 4985,
      "train_runtime": 64026.0495,
      "train_tokens_per_second": 4672.705
    },
    {
      "epoch": 0.6364795918367347,
      "grad_norm": 0.6253637143526166,
      "learning_rate": 7.703280700811447e-06,
      "loss": 0.31407947540283204,
      "num_input_tokens_seen": 299479872,
      "step": 4990,
      "train_runtime": 64095.8976,
      "train_tokens_per_second": 4672.372
    },
    {
      "epoch": 0.6371173469387755,
      "grad_norm": 0.6863215067239808,
      "learning_rate": 7.699065625373285e-06,
      "loss": 0.3558173894882202,
      "num_input_tokens_seen": 299788304,
      "step": 4995,
      "train_runtime": 64170.0961,
      "train_tokens_per_second": 4671.776
    },
    {
      "epoch": 0.6377551020408163,
      "grad_norm": 0.6250919444850997,
      "learning_rate": 7.694847841234268e-06,
      "loss": 0.3238536834716797,
      "num_input_tokens_seen": 300093128,
      "step": 5000,
      "train_runtime": 64232.8458,
      "train_tokens_per_second": 4671.958
    },
    {
      "epoch": 0.6383928571428571,
      "grad_norm": 0.5664256225168005,
      "learning_rate": 7.690627352627232e-06,
      "loss": 0.3008042573928833,
      "num_input_tokens_seen": 300381680,
      "step": 5005,
      "train_runtime": 64295.0871,
      "train_tokens_per_second": 4671.923
    },
    {
      "epoch": 0.639030612244898,
      "grad_norm": 0.5663882388947498,
      "learning_rate": 7.686404163787734e-06,
      "loss": 0.3397226810455322,
      "num_input_tokens_seen": 300692048,
      "step": 5010,
      "train_runtime": 64362.2952,
      "train_tokens_per_second": 4671.866
    },
    {
      "epoch": 0.6396683673469388,
      "grad_norm": 0.6387714807175179,
      "learning_rate": 7.68217827895404e-06,
      "loss": 0.3287776708602905,
      "num_input_tokens_seen": 301001184,
      "step": 5015,
      "train_runtime": 64439.2668,
      "train_tokens_per_second": 4671.083
    },
    {
      "epoch": 0.6403061224489796,
      "grad_norm": 0.5376293974091604,
      "learning_rate": 7.677949702367122e-06,
      "loss": 0.29458041191101075,
      "num_input_tokens_seen": 301303224,
      "step": 5020,
      "train_runtime": 64505.8927,
      "train_tokens_per_second": 4670.941
    },
    {
      "epoch": 0.6409438775510204,
      "grad_norm": 0.6333155482226666,
      "learning_rate": 7.673718438270649e-06,
      "loss": 0.3165082216262817,
      "num_input_tokens_seen": 301604056,
      "step": 5025,
      "train_runtime": 64566.5956,
      "train_tokens_per_second": 4671.209
    },
    {
      "epoch": 0.6415816326530612,
      "grad_norm": 0.6056866270349751,
      "learning_rate": 7.66948449091099e-06,
      "loss": 0.34581298828125,
      "num_input_tokens_seen": 301916416,
      "step": 5030,
      "train_runtime": 64635.2437,
      "train_tokens_per_second": 4671.08
    },
    {
      "epoch": 0.642219387755102,
      "grad_norm": 0.6205809220255843,
      "learning_rate": 7.66524786453721e-06,
      "loss": 0.31554408073425294,
      "num_input_tokens_seen": 302230184,
      "step": 5035,
      "train_runtime": 64704.3763,
      "train_tokens_per_second": 4670.939
    },
    {
      "epoch": 0.6428571428571429,
      "grad_norm": 0.5971419512432528,
      "learning_rate": 7.661008563401056e-06,
      "loss": 0.3280364513397217,
      "num_input_tokens_seen": 302539664,
      "step": 5040,
      "train_runtime": 64766.6388,
      "train_tokens_per_second": 4671.227
    },
    {
      "epoch": 0.6434948979591837,
      "grad_norm": 0.7949296375737326,
      "learning_rate": 7.656766591756964e-06,
      "loss": 0.31493797302246096,
      "num_input_tokens_seen": 302838568,
      "step": 5045,
      "train_runtime": 64827.9024,
      "train_tokens_per_second": 4671.423
    },
    {
      "epoch": 0.6441326530612245,
      "grad_norm": 0.5739689987719244,
      "learning_rate": 7.652521953862051e-06,
      "loss": 0.31455976963043214,
      "num_input_tokens_seen": 303146840,
      "step": 5050,
      "train_runtime": 64896.9355,
      "train_tokens_per_second": 4671.204
    },
    {
      "epoch": 0.6447704081632653,
      "grad_norm": 0.6820974189995908,
      "learning_rate": 7.648274653976102e-06,
      "loss": 0.3152459621429443,
      "num_input_tokens_seen": 303426728,
      "step": 5055,
      "train_runtime": 64967.5737,
      "train_tokens_per_second": 4670.433
    },
    {
      "epoch": 0.6454081632653061,
      "grad_norm": 0.625971283453452,
      "learning_rate": 7.644024696361587e-06,
      "loss": 0.3105034828186035,
      "num_input_tokens_seen": 303722768,
      "step": 5060,
      "train_runtime": 65036.4977,
      "train_tokens_per_second": 4670.036
    },
    {
      "epoch": 0.6460459183673469,
      "grad_norm": 0.5668267723732782,
      "learning_rate": 7.639772085283629e-06,
      "loss": 0.26640992164611815,
      "num_input_tokens_seen": 304019360,
      "step": 5065,
      "train_runtime": 65101.5581,
      "train_tokens_per_second": 4669.924
    },
    {
      "epoch": 0.6466836734693877,
      "grad_norm": 0.583395623415064,
      "learning_rate": 7.635516825010022e-06,
      "loss": 0.3122267246246338,
      "num_input_tokens_seen": 304326912,
      "step": 5070,
      "train_runtime": 65162.5529,
      "train_tokens_per_second": 4670.273
    },
    {
      "epoch": 0.6473214285714286,
      "grad_norm": 0.5322851635438068,
      "learning_rate": 7.631258919811215e-06,
      "loss": 0.30737671852111814,
      "num_input_tokens_seen": 304625808,
      "step": 5075,
      "train_runtime": 65229.3484,
      "train_tokens_per_second": 4670.073
    },
    {
      "epoch": 0.6479591836734694,
      "grad_norm": 0.5258592000351716,
      "learning_rate": 7.626998373960317e-06,
      "loss": 0.3087155342102051,
      "num_input_tokens_seen": 304932568,
      "step": 5080,
      "train_runtime": 65284.274,
      "train_tokens_per_second": 4670.843
    },
    {
      "epoch": 0.6485969387755102,
      "grad_norm": 0.5943785416732551,
      "learning_rate": 7.6227351917330835e-06,
      "loss": 0.32790634632110593,
      "num_input_tokens_seen": 305218960,
      "step": 5085,
      "train_runtime": 65353.0934,
      "train_tokens_per_second": 4670.306
    },
    {
      "epoch": 0.6492346938775511,
      "grad_norm": 0.5843599530027952,
      "learning_rate": 7.618469377407911e-06,
      "loss": 0.31712956428527833,
      "num_input_tokens_seen": 305520712,
      "step": 5090,
      "train_runtime": 65426.3681,
      "train_tokens_per_second": 4669.688
    },
    {
      "epoch": 0.6498724489795918,
      "grad_norm": 0.5914629103925383,
      "learning_rate": 7.614200935265845e-06,
      "loss": 0.30661187171936033,
      "num_input_tokens_seen": 305823824,
      "step": 5095,
      "train_runtime": 65494.4667,
      "train_tokens_per_second": 4669.46
    },
    {
      "epoch": 0.6505102040816326,
      "grad_norm": 0.6477041798359954,
      "learning_rate": 7.609929869590566e-06,
      "loss": 0.30426583290100095,
      "num_input_tokens_seen": 306127224,
      "step": 5100,
      "train_runtime": 65566.039,
      "train_tokens_per_second": 4668.991
    },
    {
      "epoch": 0.6511479591836735,
      "grad_norm": 0.5960900470317049,
      "learning_rate": 7.605656184668385e-06,
      "loss": 0.3433689117431641,
      "num_input_tokens_seen": 306433400,
      "step": 5105,
      "train_runtime": 65621.983,
      "train_tokens_per_second": 4669.676
    },
    {
      "epoch": 0.6517857142857143,
      "grad_norm": 0.6287577618544843,
      "learning_rate": 7.6013798847882455e-06,
      "loss": 0.30711252689361573,
      "num_input_tokens_seen": 306726808,
      "step": 5110,
      "train_runtime": 65691.7669,
      "train_tokens_per_second": 4669.182
    },
    {
      "epoch": 0.6524234693877551,
      "grad_norm": 0.5725835282668112,
      "learning_rate": 7.597100974241711e-06,
      "loss": 0.3282453536987305,
      "num_input_tokens_seen": 307036920,
      "step": 5115,
      "train_runtime": 65749.2341,
      "train_tokens_per_second": 4669.817
    },
    {
      "epoch": 0.6530612244897959,
      "grad_norm": 0.5735835153695586,
      "learning_rate": 7.592819457322967e-06,
      "loss": 0.31277282238006593,
      "num_input_tokens_seen": 307348912,
      "step": 5120,
      "train_runtime": 65805.5313,
      "train_tokens_per_second": 4670.563
    },
    {
      "epoch": 0.6536989795918368,
      "grad_norm": 0.6478752506247863,
      "learning_rate": 7.588535338328816e-06,
      "loss": 0.33504557609558105,
      "num_input_tokens_seen": 307662240,
      "step": 5125,
      "train_runtime": 65871.7396,
      "train_tokens_per_second": 4670.626
    },
    {
      "epoch": 0.6543367346938775,
      "grad_norm": 0.5360790048049417,
      "learning_rate": 7.584248621558669e-06,
      "loss": 0.33577775955200195,
      "num_input_tokens_seen": 307973576,
      "step": 5130,
      "train_runtime": 65945.9886,
      "train_tokens_per_second": 4670.088
    },
    {
      "epoch": 0.6549744897959183,
      "grad_norm": 0.630992103318103,
      "learning_rate": 7.579959311314549e-06,
      "loss": 0.32630224227905275,
      "num_input_tokens_seen": 308269776,
      "step": 5135,
      "train_runtime": 66000.5134,
      "train_tokens_per_second": 4670.718
    },
    {
      "epoch": 0.6556122448979592,
      "grad_norm": 0.585089749702886,
      "learning_rate": 7.575667411901074e-06,
      "loss": 0.33813388347625734,
      "num_input_tokens_seen": 308570536,
      "step": 5140,
      "train_runtime": 66059.6134,
      "train_tokens_per_second": 4671.092
    },
    {
      "epoch": 0.65625,
      "grad_norm": 0.6180172337705084,
      "learning_rate": 7.571372927625469e-06,
      "loss": 0.2856073141098022,
      "num_input_tokens_seen": 308858760,
      "step": 5145,
      "train_runtime": 66125.8292,
      "train_tokens_per_second": 4670.773
    },
    {
      "epoch": 0.6568877551020408,
      "grad_norm": 0.6205414200755942,
      "learning_rate": 7.567075862797546e-06,
      "loss": 0.3189757585525513,
      "num_input_tokens_seen": 309151432,
      "step": 5150,
      "train_runtime": 66195.2196,
      "train_tokens_per_second": 4670.298
    },
    {
      "epoch": 0.6575255102040817,
      "grad_norm": 0.45497411279431826,
      "learning_rate": 7.562776221729709e-06,
      "loss": 0.2730249404907227,
      "num_input_tokens_seen": 309448000,
      "step": 5155,
      "train_runtime": 66273.5363,
      "train_tokens_per_second": 4669.254
    },
    {
      "epoch": 0.6581632653061225,
      "grad_norm": 0.6431193726473291,
      "learning_rate": 7.558474008736951e-06,
      "loss": 0.33919711112976075,
      "num_input_tokens_seen": 309755312,
      "step": 5160,
      "train_runtime": 66337.0644,
      "train_tokens_per_second": 4669.415
    },
    {
      "epoch": 0.6588010204081632,
      "grad_norm": 0.6345790174538617,
      "learning_rate": 7.554169228136842e-06,
      "loss": 0.32866082191467283,
      "num_input_tokens_seen": 310046040,
      "step": 5165,
      "train_runtime": 66398.9286,
      "train_tokens_per_second": 4669.443
    },
    {
      "epoch": 0.6594387755102041,
      "grad_norm": 0.6466110440669572,
      "learning_rate": 7.549861884249529e-06,
      "loss": 0.31755714416503905,
      "num_input_tokens_seen": 310348952,
      "step": 5170,
      "train_runtime": 66464.8666,
      "train_tokens_per_second": 4669.368
    },
    {
      "epoch": 0.6600765306122449,
      "grad_norm": 0.6753687852421429,
      "learning_rate": 7.545551981397733e-06,
      "loss": 0.33164520263671876,
      "num_input_tokens_seen": 310648160,
      "step": 5175,
      "train_runtime": 66521.503,
      "train_tokens_per_second": 4669.891
    },
    {
      "epoch": 0.6607142857142857,
      "grad_norm": 0.658451408497734,
      "learning_rate": 7.541239523906742e-06,
      "loss": 0.30910749435424806,
      "num_input_tokens_seen": 310943712,
      "step": 5180,
      "train_runtime": 66574.9573,
      "train_tokens_per_second": 4670.581
    },
    {
      "epoch": 0.6613520408163265,
      "grad_norm": 0.6091814398062841,
      "learning_rate": 7.536924516104411e-06,
      "loss": 0.3098081111907959,
      "num_input_tokens_seen": 311228776,
      "step": 5185,
      "train_runtime": 66629.5605,
      "train_tokens_per_second": 4671.032
    },
    {
      "epoch": 0.6619897959183674,
      "grad_norm": 0.6174964320202159,
      "learning_rate": 7.532606962321149e-06,
      "loss": 0.31559340953826903,
      "num_input_tokens_seen": 311525824,
      "step": 5190,
      "train_runtime": 66687.8162,
      "train_tokens_per_second": 4671.405
    },
    {
      "epoch": 0.6626275510204082,
      "grad_norm": 0.5745193401631046,
      "learning_rate": 7.528286866889924e-06,
      "loss": 0.3417334079742432,
      "num_input_tokens_seen": 311825560,
      "step": 5195,
      "train_runtime": 66762.5988,
      "train_tokens_per_second": 4670.662
    },
    {
      "epoch": 0.6632653061224489,
      "grad_norm": 0.5378342298277153,
      "learning_rate": 7.523964234146251e-06,
      "loss": 0.3097379207611084,
      "num_input_tokens_seen": 312120744,
      "step": 5200,
      "train_runtime": 66833.4645,
      "train_tokens_per_second": 4670.127
    },
    {
      "epoch": 0.6639030612244898,
      "grad_norm": 0.5923373947572459,
      "learning_rate": 7.519639068428199e-06,
      "loss": 0.3182514667510986,
      "num_input_tokens_seen": 312409120,
      "step": 5205,
      "train_runtime": 66892.4783,
      "train_tokens_per_second": 4670.318
    },
    {
      "epoch": 0.6645408163265306,
      "grad_norm": 0.6071032131368624,
      "learning_rate": 7.515311374076368e-06,
      "loss": 0.3007327079772949,
      "num_input_tokens_seen": 312709064,
      "step": 5210,
      "train_runtime": 66954.6939,
      "train_tokens_per_second": 4670.458
    },
    {
      "epoch": 0.6651785714285714,
      "grad_norm": 0.6560051603946989,
      "learning_rate": 7.5109811554339054e-06,
      "loss": 0.33915157318115235,
      "num_input_tokens_seen": 313010520,
      "step": 5215,
      "train_runtime": 67023.5518,
      "train_tokens_per_second": 4670.157
    },
    {
      "epoch": 0.6658163265306123,
      "grad_norm": 0.5871591032812562,
      "learning_rate": 7.506648416846485e-06,
      "loss": 0.2778173446655273,
      "num_input_tokens_seen": 313304880,
      "step": 5220,
      "train_runtime": 67091.1107,
      "train_tokens_per_second": 4669.842
    },
    {
      "epoch": 0.6664540816326531,
      "grad_norm": 0.6423214904932472,
      "learning_rate": 7.502313162662316e-06,
      "loss": 0.3125695943832397,
      "num_input_tokens_seen": 313603488,
      "step": 5225,
      "train_runtime": 67152.8292,
      "train_tokens_per_second": 4669.997
    },
    {
      "epoch": 0.6670918367346939,
      "grad_norm": 0.6714441450406599,
      "learning_rate": 7.4979753972321265e-06,
      "loss": 0.32171783447265623,
      "num_input_tokens_seen": 313895880,
      "step": 5230,
      "train_runtime": 67209.7897,
      "train_tokens_per_second": 4670.389
    },
    {
      "epoch": 0.6677295918367347,
      "grad_norm": 0.6280272825117984,
      "learning_rate": 7.493635124909167e-06,
      "loss": 0.30132863521575926,
      "num_input_tokens_seen": 314181560,
      "step": 5235,
      "train_runtime": 67282.2628,
      "train_tokens_per_second": 4669.605
    },
    {
      "epoch": 0.6683673469387755,
      "grad_norm": 0.6046789286619518,
      "learning_rate": 7.489292350049205e-06,
      "loss": 0.34362359046936036,
      "num_input_tokens_seen": 314480704,
      "step": 5240,
      "train_runtime": 67344.5283,
      "train_tokens_per_second": 4669.729
    },
    {
      "epoch": 0.6690051020408163,
      "grad_norm": 0.6942861193479892,
      "learning_rate": 7.484947077010519e-06,
      "loss": 0.2994524955749512,
      "num_input_tokens_seen": 314785344,
      "step": 5245,
      "train_runtime": 67415.0252,
      "train_tokens_per_second": 4669.365
    },
    {
      "epoch": 0.6696428571428571,
      "grad_norm": 0.6531309656478049,
      "learning_rate": 7.480599310153891e-06,
      "loss": 0.3404392719268799,
      "num_input_tokens_seen": 315073712,
      "step": 5250,
      "train_runtime": 67475.7065,
      "train_tokens_per_second": 4669.439
    },
    {
      "epoch": 0.670280612244898,
      "grad_norm": 0.6287139881930139,
      "learning_rate": 7.476249053842613e-06,
      "loss": 0.31229596138000487,
      "num_input_tokens_seen": 315369368,
      "step": 5255,
      "train_runtime": 67542.6896,
      "train_tokens_per_second": 4669.186
    },
    {
      "epoch": 0.6709183673469388,
      "grad_norm": 0.6008136713743412,
      "learning_rate": 7.471896312442467e-06,
      "loss": 0.30401148796081545,
      "num_input_tokens_seen": 315671744,
      "step": 5260,
      "train_runtime": 67608.0524,
      "train_tokens_per_second": 4669.144
    },
    {
      "epoch": 0.6715561224489796,
      "grad_norm": 0.640502984649124,
      "learning_rate": 7.467541090321735e-06,
      "loss": 0.312398099899292,
      "num_input_tokens_seen": 315956648,
      "step": 5265,
      "train_runtime": 67674.1447,
      "train_tokens_per_second": 4668.794
    },
    {
      "epoch": 0.6721938775510204,
      "grad_norm": 0.6857237219956531,
      "learning_rate": 7.463183391851186e-06,
      "loss": 0.3361454963684082,
      "num_input_tokens_seen": 316269784,
      "step": 5270,
      "train_runtime": 67726.7386,
      "train_tokens_per_second": 4669.792
    },
    {
      "epoch": 0.6728316326530612,
      "grad_norm": 0.560048697643107,
      "learning_rate": 7.4588232214040744e-06,
      "loss": 0.3127558469772339,
      "num_input_tokens_seen": 316561816,
      "step": 5275,
      "train_runtime": 67802.9223,
      "train_tokens_per_second": 4668.852
    },
    {
      "epoch": 0.673469387755102,
      "grad_norm": 0.6315437137503608,
      "learning_rate": 7.454460583356138e-06,
      "loss": 0.33094162940979005,
      "num_input_tokens_seen": 316859640,
      "step": 5280,
      "train_runtime": 67868.741,
      "train_tokens_per_second": 4668.713
    },
    {
      "epoch": 0.6741071428571429,
      "grad_norm": 0.6589156035175846,
      "learning_rate": 7.4500954820855854e-06,
      "loss": 0.31834456920623777,
      "num_input_tokens_seen": 317163760,
      "step": 5285,
      "train_runtime": 67930.3554,
      "train_tokens_per_second": 4668.955
    },
    {
      "epoch": 0.6747448979591837,
      "grad_norm": 0.5594566819219117,
      "learning_rate": 7.445727921973104e-06,
      "loss": 0.31772470474243164,
      "num_input_tokens_seen": 317456784,
      "step": 5290,
      "train_runtime": 67995.0793,
      "train_tokens_per_second": 4668.82
    },
    {
      "epoch": 0.6753826530612245,
      "grad_norm": 0.5753294134784083,
      "learning_rate": 7.441357907401841e-06,
      "loss": 0.33693690299987794,
      "num_input_tokens_seen": 317774432,
      "step": 5295,
      "train_runtime": 68066.3482,
      "train_tokens_per_second": 4668.598
    },
    {
      "epoch": 0.6760204081632653,
      "grad_norm": 0.6508193992785211,
      "learning_rate": 7.436985442757415e-06,
      "loss": 0.33437199592590333,
      "num_input_tokens_seen": 318074152,
      "step": 5300,
      "train_runtime": 68138.8467,
      "train_tokens_per_second": 4668.03
    },
    {
      "epoch": 0.6766581632653061,
      "grad_norm": 0.6658023048885741,
      "learning_rate": 7.432610532427899e-06,
      "loss": 0.290529727935791,
      "num_input_tokens_seen": 318366544,
      "step": 5305,
      "train_runtime": 68204.2277,
      "train_tokens_per_second": 4667.842
    },
    {
      "epoch": 0.6772959183673469,
      "grad_norm": 0.6078737467243067,
      "learning_rate": 7.428233180803821e-06,
      "loss": 0.3208777904510498,
      "num_input_tokens_seen": 318669464,
      "step": 5310,
      "train_runtime": 68276.0065,
      "train_tokens_per_second": 4667.371
    },
    {
      "epoch": 0.6779336734693877,
      "grad_norm": 0.6370812769416774,
      "learning_rate": 7.423853392278158e-06,
      "loss": 0.3412617206573486,
      "num_input_tokens_seen": 318964520,
      "step": 5315,
      "train_runtime": 68341.6345,
      "train_tokens_per_second": 4667.206
    },
    {
      "epoch": 0.6785714285714286,
      "grad_norm": 0.6791444764089478,
      "learning_rate": 7.419471171246334e-06,
      "loss": 0.31648027896881104,
      "num_input_tokens_seen": 319266552,
      "step": 5320,
      "train_runtime": 68403.3871,
      "train_tokens_per_second": 4667.409
    },
    {
      "epoch": 0.6792091836734694,
      "grad_norm": 0.6209476744426583,
      "learning_rate": 7.415086522106215e-06,
      "loss": 0.3592404842376709,
      "num_input_tokens_seen": 319559784,
      "step": 5325,
      "train_runtime": 68469.0896,
      "train_tokens_per_second": 4667.212
    },
    {
      "epoch": 0.6798469387755102,
      "grad_norm": 0.6687067839958473,
      "learning_rate": 7.410699449258102e-06,
      "loss": 0.30386877059936523,
      "num_input_tokens_seen": 319855416,
      "step": 5330,
      "train_runtime": 68544.7873,
      "train_tokens_per_second": 4666.371
    },
    {
      "epoch": 0.6804846938775511,
      "grad_norm": 0.5567310646869349,
      "learning_rate": 7.406309957104727e-06,
      "loss": 0.3273475646972656,
      "num_input_tokens_seen": 320159136,
      "step": 5335,
      "train_runtime": 68607.8178,
      "train_tokens_per_second": 4666.511
    },
    {
      "epoch": 0.6811224489795918,
      "grad_norm": 0.6195204654488123,
      "learning_rate": 7.401918050051256e-06,
      "loss": 0.33095426559448243,
      "num_input_tokens_seen": 320459376,
      "step": 5340,
      "train_runtime": 68669.0539,
      "train_tokens_per_second": 4666.722
    },
    {
      "epoch": 0.6817602040816326,
      "grad_norm": 0.5936377968108272,
      "learning_rate": 7.39752373250527e-06,
      "loss": 0.31267762184143066,
      "num_input_tokens_seen": 320761440,
      "step": 5345,
      "train_runtime": 68729.2521,
      "train_tokens_per_second": 4667.029
    },
    {
      "epoch": 0.6823979591836735,
      "grad_norm": 0.5934863280670649,
      "learning_rate": 7.393127008876777e-06,
      "loss": 0.31867103576660155,
      "num_input_tokens_seen": 321048384,
      "step": 5350,
      "train_runtime": 68798.5837,
      "train_tokens_per_second": 4666.497
    },
    {
      "epoch": 0.6830357142857143,
      "grad_norm": 0.5841510400498795,
      "learning_rate": 7.388727883578195e-06,
      "loss": 0.30554609298706054,
      "num_input_tokens_seen": 321342320,
      "step": 5355,
      "train_runtime": 68871.2286,
      "train_tokens_per_second": 4665.843
    },
    {
      "epoch": 0.6836734693877551,
      "grad_norm": 0.6307707627641727,
      "learning_rate": 7.384326361024352e-06,
      "loss": 0.300091028213501,
      "num_input_tokens_seen": 321641256,
      "step": 5360,
      "train_runtime": 68936.4124,
      "train_tokens_per_second": 4665.767
    },
    {
      "epoch": 0.6843112244897959,
      "grad_norm": 0.6306372146818784,
      "learning_rate": 7.3799224456324865e-06,
      "loss": 0.30863926410675047,
      "num_input_tokens_seen": 321937816,
      "step": 5365,
      "train_runtime": 68994.569,
      "train_tokens_per_second": 4666.133
    },
    {
      "epoch": 0.6849489795918368,
      "grad_norm": 0.6008248350813888,
      "learning_rate": 7.375516141822232e-06,
      "loss": 0.3344412088394165,
      "num_input_tokens_seen": 322242640,
      "step": 5370,
      "train_runtime": 69054.7659,
      "train_tokens_per_second": 4666.479
    },
    {
      "epoch": 0.6855867346938775,
      "grad_norm": 0.8279346981897546,
      "learning_rate": 7.371107454015626e-06,
      "loss": 0.3194762945175171,
      "num_input_tokens_seen": 322524848,
      "step": 5375,
      "train_runtime": 69122.6831,
      "train_tokens_per_second": 4665.977
    },
    {
      "epoch": 0.6862244897959183,
      "grad_norm": 0.6209797654488943,
      "learning_rate": 7.3666963866370886e-06,
      "loss": 0.31194396018981935,
      "num_input_tokens_seen": 322829736,
      "step": 5380,
      "train_runtime": 69181.8802,
      "train_tokens_per_second": 4666.391
    },
    {
      "epoch": 0.6868622448979592,
      "grad_norm": 0.5691481622117558,
      "learning_rate": 7.3622829441134395e-06,
      "loss": 0.2930125713348389,
      "num_input_tokens_seen": 323123808,
      "step": 5385,
      "train_runtime": 69235.0295,
      "train_tokens_per_second": 4667.057
    },
    {
      "epoch": 0.6875,
      "grad_norm": 0.6473616644941033,
      "learning_rate": 7.357867130873875e-06,
      "loss": 0.3505560874938965,
      "num_input_tokens_seen": 323429168,
      "step": 5390,
      "train_runtime": 69292.5759,
      "train_tokens_per_second": 4667.588
    },
    {
      "epoch": 0.6881377551020408,
      "grad_norm": 0.6467814560569287,
      "learning_rate": 7.353448951349971e-06,
      "loss": 0.3212274074554443,
      "num_input_tokens_seen": 323721952,
      "step": 5395,
      "train_runtime": 69359.1103,
      "train_tokens_per_second": 4667.331
    },
    {
      "epoch": 0.6887755102040817,
      "grad_norm": 0.600696886308647,
      "learning_rate": 7.349028409975681e-06,
      "loss": 0.30536613464355467,
      "num_input_tokens_seen": 324008792,
      "step": 5400,
      "train_runtime": 69432.2952,
      "train_tokens_per_second": 4666.543
    },
    {
      "epoch": 0.6894132653061225,
      "grad_norm": 0.5884289747346985,
      "learning_rate": 7.344605511187322e-06,
      "loss": 0.32222843170166016,
      "num_input_tokens_seen": 324302264,
      "step": 5405,
      "train_runtime": 69497.2928,
      "train_tokens_per_second": 4666.401
    },
    {
      "epoch": 0.6900510204081632,
      "grad_norm": 0.6526508714808632,
      "learning_rate": 7.3401802594235894e-06,
      "loss": 0.32078051567077637,
      "num_input_tokens_seen": 324603400,
      "step": 5410,
      "train_runtime": 69564.2313,
      "train_tokens_per_second": 4666.24
    },
    {
      "epoch": 0.6906887755102041,
      "grad_norm": 0.587802745663465,
      "learning_rate": 7.3357526591255265e-06,
      "loss": 0.3128876924514771,
      "num_input_tokens_seen": 324905992,
      "step": 5415,
      "train_runtime": 69628.4336,
      "train_tokens_per_second": 4666.283
    },
    {
      "epoch": 0.6913265306122449,
      "grad_norm": 0.6294607766993028,
      "learning_rate": 7.331322714736543e-06,
      "loss": 0.310003662109375,
      "num_input_tokens_seen": 325216272,
      "step": 5420,
      "train_runtime": 69688.5627,
      "train_tokens_per_second": 4666.709
    },
    {
      "epoch": 0.6919642857142857,
      "grad_norm": 0.6494295041813904,
      "learning_rate": 7.326890430702396e-06,
      "loss": 0.34342279434204104,
      "num_input_tokens_seen": 325510864,
      "step": 5425,
      "train_runtime": 69761.3268,
      "train_tokens_per_second": 4666.065
    },
    {
      "epoch": 0.6926020408163265,
      "grad_norm": 0.5551213723940004,
      "learning_rate": 7.3224558114711905e-06,
      "loss": 0.31579127311706545,
      "num_input_tokens_seen": 325814872,
      "step": 5430,
      "train_runtime": 69835.6678,
      "train_tokens_per_second": 4665.451
    },
    {
      "epoch": 0.6932397959183674,
      "grad_norm": 0.570192939463917,
      "learning_rate": 7.31801886149338e-06,
      "loss": 0.3156977653503418,
      "num_input_tokens_seen": 326102152,
      "step": 5435,
      "train_runtime": 69901.8161,
      "train_tokens_per_second": 4665.146
    },
    {
      "epoch": 0.6938775510204082,
      "grad_norm": 0.6075826448906008,
      "learning_rate": 7.313579585221752e-06,
      "loss": 0.31879711151123047,
      "num_input_tokens_seen": 326405384,
      "step": 5440,
      "train_runtime": 69971.8967,
      "train_tokens_per_second": 4664.807
    },
    {
      "epoch": 0.6945153061224489,
      "grad_norm": 0.6530047373930488,
      "learning_rate": 7.309137987111428e-06,
      "loss": 0.3283576965332031,
      "num_input_tokens_seen": 326701896,
      "step": 5445,
      "train_runtime": 70026.3146,
      "train_tokens_per_second": 4665.416
    },
    {
      "epoch": 0.6951530612244898,
      "grad_norm": 0.5200186914870559,
      "learning_rate": 7.304694071619866e-06,
      "loss": 0.3029489040374756,
      "num_input_tokens_seen": 327008480,
      "step": 5450,
      "train_runtime": 70086.9124,
      "train_tokens_per_second": 4665.757
    },
    {
      "epoch": 0.6957908163265306,
      "grad_norm": 0.6400323639371005,
      "learning_rate": 7.300247843206844e-06,
      "loss": 0.3218910455703735,
      "num_input_tokens_seen": 327312080,
      "step": 5455,
      "train_runtime": 70151.711,
      "train_tokens_per_second": 4665.775
    },
    {
      "epoch": 0.6964285714285714,
      "grad_norm": 0.745126229894632,
      "learning_rate": 7.2957993063344615e-06,
      "loss": 0.3268340826034546,
      "num_input_tokens_seen": 327610720,
      "step": 5460,
      "train_runtime": 70211.1615,
      "train_tokens_per_second": 4666.077
    },
    {
      "epoch": 0.6970663265306123,
      "grad_norm": 0.6267138034083698,
      "learning_rate": 7.291348465467136e-06,
      "loss": 0.3033930778503418,
      "num_input_tokens_seen": 327908664,
      "step": 5465,
      "train_runtime": 70281.0144,
      "train_tokens_per_second": 4665.679
    },
    {
      "epoch": 0.6977040816326531,
      "grad_norm": 0.593440548594412,
      "learning_rate": 7.286895325071599e-06,
      "loss": 0.31491804122924805,
      "num_input_tokens_seen": 328213216,
      "step": 5470,
      "train_runtime": 70348.5324,
      "train_tokens_per_second": 4665.53
    },
    {
      "epoch": 0.6983418367346939,
      "grad_norm": 0.5532470434303797,
      "learning_rate": 7.282439889616887e-06,
      "loss": 0.31497039794921877,
      "num_input_tokens_seen": 328494408,
      "step": 5475,
      "train_runtime": 70417.7486,
      "train_tokens_per_second": 4664.938
    },
    {
      "epoch": 0.6989795918367347,
      "grad_norm": 0.6900719796874891,
      "learning_rate": 7.277982163574339e-06,
      "loss": 0.3080450534820557,
      "num_input_tokens_seen": 328776992,
      "step": 5480,
      "train_runtime": 70478.2147,
      "train_tokens_per_second": 4664.945
    },
    {
      "epoch": 0.6996173469387755,
      "grad_norm": 0.7299462946103753,
      "learning_rate": 7.273522151417598e-06,
      "loss": 0.3473284482955933,
      "num_input_tokens_seen": 329077928,
      "step": 5485,
      "train_runtime": 70545.7672,
      "train_tokens_per_second": 4664.744
    },
    {
      "epoch": 0.7002551020408163,
      "grad_norm": 0.5425858036772111,
      "learning_rate": 7.2690598576225954e-06,
      "loss": 0.3336508750915527,
      "num_input_tokens_seen": 329378848,
      "step": 5490,
      "train_runtime": 70608.5906,
      "train_tokens_per_second": 4664.855
    },
    {
      "epoch": 0.7008928571428571,
      "grad_norm": 0.5647942767407259,
      "learning_rate": 7.264595286667554e-06,
      "loss": 0.310113525390625,
      "num_input_tokens_seen": 329661384,
      "step": 5495,
      "train_runtime": 70661.2711,
      "train_tokens_per_second": 4665.376
    },
    {
      "epoch": 0.701530612244898,
      "grad_norm": 0.512781121429987,
      "learning_rate": 7.2601284430329836e-06,
      "loss": 0.3203456163406372,
      "num_input_tokens_seen": 329971352,
      "step": 5500,
      "train_runtime": 70737.8087,
      "train_tokens_per_second": 4664.71
    },
    {
      "epoch": 0.7021683673469388,
      "grad_norm": 0.5688066588778269,
      "learning_rate": 7.255659331201673e-06,
      "loss": 0.3099203109741211,
      "num_input_tokens_seen": 330291552,
      "step": 5505,
      "train_runtime": 70798.6698,
      "train_tokens_per_second": 4665.223
    },
    {
      "epoch": 0.7028061224489796,
      "grad_norm": 0.5891400739535447,
      "learning_rate": 7.251187955658691e-06,
      "loss": 0.3002140760421753,
      "num_input_tokens_seen": 330586912,
      "step": 5510,
      "train_runtime": 70862.5548,
      "train_tokens_per_second": 4665.185
    },
    {
      "epoch": 0.7034438775510204,
      "grad_norm": 0.5634930859443,
      "learning_rate": 7.246714320891371e-06,
      "loss": 0.303344202041626,
      "num_input_tokens_seen": 330881376,
      "step": 5515,
      "train_runtime": 70936.0328,
      "train_tokens_per_second": 4664.504
    },
    {
      "epoch": 0.7040816326530612,
      "grad_norm": 0.7030866651444997,
      "learning_rate": 7.24223843138932e-06,
      "loss": 0.34733502864837645,
      "num_input_tokens_seen": 331184952,
      "step": 5520,
      "train_runtime": 70995.1486,
      "train_tokens_per_second": 4664.896
    },
    {
      "epoch": 0.704719387755102,
      "grad_norm": 0.6239650926525834,
      "learning_rate": 7.237760291644405e-06,
      "loss": 0.3356089353561401,
      "num_input_tokens_seen": 331476552,
      "step": 5525,
      "train_runtime": 71056.1078,
      "train_tokens_per_second": 4664.997
    },
    {
      "epoch": 0.7053571428571429,
      "grad_norm": 0.5444120060158437,
      "learning_rate": 7.233279906150752e-06,
      "loss": 0.2855654239654541,
      "num_input_tokens_seen": 331770192,
      "step": 5530,
      "train_runtime": 71128.7262,
      "train_tokens_per_second": 4664.363
    },
    {
      "epoch": 0.7059948979591837,
      "grad_norm": 0.5728118319140764,
      "learning_rate": 7.228797279404744e-06,
      "loss": 0.3068604230880737,
      "num_input_tokens_seen": 332079320,
      "step": 5535,
      "train_runtime": 71196.7875,
      "train_tokens_per_second": 4664.246
    },
    {
      "epoch": 0.7066326530612245,
      "grad_norm": 0.6239924959947185,
      "learning_rate": 7.224312415905005e-06,
      "loss": 0.3272867202758789,
      "num_input_tokens_seen": 332367456,
      "step": 5540,
      "train_runtime": 71266.177,
      "train_tokens_per_second": 4663.748
    },
    {
      "epoch": 0.7072704081632653,
      "grad_norm": 0.6408351287033327,
      "learning_rate": 7.219825320152411e-06,
      "loss": 0.31500868797302245,
      "num_input_tokens_seen": 332646216,
      "step": 5545,
      "train_runtime": 71329.6649,
      "train_tokens_per_second": 4663.505
    },
    {
      "epoch": 0.7079081632653061,
      "grad_norm": 0.608909772396368,
      "learning_rate": 7.2153359966500765e-06,
      "loss": 0.342067551612854,
      "num_input_tokens_seen": 332943400,
      "step": 5550,
      "train_runtime": 71393.3297,
      "train_tokens_per_second": 4663.508
    },
    {
      "epoch": 0.7085459183673469,
      "grad_norm": 0.5893268458814994,
      "learning_rate": 7.210844449903352e-06,
      "loss": 0.2956761598587036,
      "num_input_tokens_seen": 333248504,
      "step": 5555,
      "train_runtime": 71456.0605,
      "train_tokens_per_second": 4663.684
    },
    {
      "epoch": 0.7091836734693877,
      "grad_norm": 0.5988037762181453,
      "learning_rate": 7.206350684419814e-06,
      "loss": 0.3130906105041504,
      "num_input_tokens_seen": 333548136,
      "step": 5560,
      "train_runtime": 71529.7065,
      "train_tokens_per_second": 4663.072
    },
    {
      "epoch": 0.7098214285714286,
      "grad_norm": 0.5755608587354144,
      "learning_rate": 7.2018547047092745e-06,
      "loss": 0.3113144874572754,
      "num_input_tokens_seen": 333859296,
      "step": 5565,
      "train_runtime": 71594.5593,
      "train_tokens_per_second": 4663.194
    },
    {
      "epoch": 0.7104591836734694,
      "grad_norm": 0.6933761028281489,
      "learning_rate": 7.1973565152837635e-06,
      "loss": 0.2980863094329834,
      "num_input_tokens_seen": 334133912,
      "step": 5570,
      "train_runtime": 71660.1254,
      "train_tokens_per_second": 4662.759
    },
    {
      "epoch": 0.7110969387755102,
      "grad_norm": 0.5744246362594235,
      "learning_rate": 7.192856120657524e-06,
      "loss": 0.3452368021011353,
      "num_input_tokens_seen": 334436520,
      "step": 5575,
      "train_runtime": 71727.3936,
      "train_tokens_per_second": 4662.605
    },
    {
      "epoch": 0.7117346938775511,
      "grad_norm": 0.5562666466329808,
      "learning_rate": 7.18835352534702e-06,
      "loss": 0.3156794786453247,
      "num_input_tokens_seen": 334724960,
      "step": 5580,
      "train_runtime": 71792.7717,
      "train_tokens_per_second": 4662.377
    },
    {
      "epoch": 0.7123724489795918,
      "grad_norm": 0.5810126803543275,
      "learning_rate": 7.183848733870917e-06,
      "loss": 0.31806468963623047,
      "num_input_tokens_seen": 335027432,
      "step": 5585,
      "train_runtime": 71853.7856,
      "train_tokens_per_second": 4662.627
    },
    {
      "epoch": 0.7130102040816326,
      "grad_norm": 0.5614850789313679,
      "learning_rate": 7.179341750750092e-06,
      "loss": 0.3097982883453369,
      "num_input_tokens_seen": 335315032,
      "step": 5590,
      "train_runtime": 71920.507,
      "train_tokens_per_second": 4662.301
    },
    {
      "epoch": 0.7136479591836735,
      "grad_norm": 0.7027806136416876,
      "learning_rate": 7.174832580507614e-06,
      "loss": 0.3387152671813965,
      "num_input_tokens_seen": 335615736,
      "step": 5595,
      "train_runtime": 71990.4425,
      "train_tokens_per_second": 4661.949
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.5677823402656198,
      "learning_rate": 7.170321227668752e-06,
      "loss": 0.3012820243835449,
      "num_input_tokens_seen": 335911536,
      "step": 5600,
      "train_runtime": 72042.9342,
      "train_tokens_per_second": 4662.658
    },
    {
      "epoch": 0.7149234693877551,
      "grad_norm": 0.6854391883576302,
      "learning_rate": 7.165807696760964e-06,
      "loss": 0.3246180534362793,
      "num_input_tokens_seen": 336216560,
      "step": 5605,
      "train_runtime": 72104.3182,
      "train_tokens_per_second": 4662.919
    },
    {
      "epoch": 0.7155612244897959,
      "grad_norm": 0.5519379710985469,
      "learning_rate": 7.16129199231389e-06,
      "loss": 0.3269120931625366,
      "num_input_tokens_seen": 336525464,
      "step": 5610,
      "train_runtime": 72167.3153,
      "train_tokens_per_second": 4663.128
    },
    {
      "epoch": 0.7161989795918368,
      "grad_norm": 0.513365672663749,
      "learning_rate": 7.15677411885936e-06,
      "loss": 0.3312812328338623,
      "num_input_tokens_seen": 336842232,
      "step": 5615,
      "train_runtime": 72212.8576,
      "train_tokens_per_second": 4664.574
    },
    {
      "epoch": 0.7168367346938775,
      "grad_norm": 0.6341148503548436,
      "learning_rate": 7.152254080931372e-06,
      "loss": 0.3205650806427002,
      "num_input_tokens_seen": 337150176,
      "step": 5620,
      "train_runtime": 72283.9459,
      "train_tokens_per_second": 4664.247
    },
    {
      "epoch": 0.7174744897959183,
      "grad_norm": 0.5548980557737851,
      "learning_rate": 7.1477318830661e-06,
      "loss": 0.3206507682800293,
      "num_input_tokens_seen": 337437632,
      "step": 5625,
      "train_runtime": 72345.8729,
      "train_tokens_per_second": 4664.228
    },
    {
      "epoch": 0.7181122448979592,
      "grad_norm": 0.5971686031945006,
      "learning_rate": 7.143207529801888e-06,
      "loss": 0.31696553230285646,
      "num_input_tokens_seen": 337741592,
      "step": 5630,
      "train_runtime": 72402.8212,
      "train_tokens_per_second": 4664.757
    },
    {
      "epoch": 0.71875,
      "grad_norm": 0.5987837254902296,
      "learning_rate": 7.1386810256792374e-06,
      "loss": 0.3265007495880127,
      "num_input_tokens_seen": 338054400,
      "step": 5635,
      "train_runtime": 72461.0792,
      "train_tokens_per_second": 4665.324
    },
    {
      "epoch": 0.7193877551020408,
      "grad_norm": 0.6011374940867165,
      "learning_rate": 7.134152375240812e-06,
      "loss": 0.3095307588577271,
      "num_input_tokens_seen": 338358256,
      "step": 5640,
      "train_runtime": 72513.7511,
      "train_tokens_per_second": 4666.125
    },
    {
      "epoch": 0.7200255102040817,
      "grad_norm": 0.5562678963814983,
      "learning_rate": 7.1296215830314295e-06,
      "loss": 0.3101684331893921,
      "num_input_tokens_seen": 338639064,
      "step": 5645,
      "train_runtime": 72581.1899,
      "train_tokens_per_second": 4665.659
    },
    {
      "epoch": 0.7206632653061225,
      "grad_norm": 0.6891181205804473,
      "learning_rate": 7.125088653598057e-06,
      "loss": 0.31908905506134033,
      "num_input_tokens_seen": 338936432,
      "step": 5650,
      "train_runtime": 72655.789,
      "train_tokens_per_second": 4664.961
    },
    {
      "epoch": 0.7213010204081632,
      "grad_norm": 0.599370342027607,
      "learning_rate": 7.1205535914898035e-06,
      "loss": 0.3092601537704468,
      "num_input_tokens_seen": 339243760,
      "step": 5655,
      "train_runtime": 72729.6511,
      "train_tokens_per_second": 4664.449
    },
    {
      "epoch": 0.7219387755102041,
      "grad_norm": 0.5428513321361451,
      "learning_rate": 7.116016401257921e-06,
      "loss": 0.309174108505249,
      "num_input_tokens_seen": 339537904,
      "step": 5660,
      "train_runtime": 72800.7996,
      "train_tokens_per_second": 4663.931
    },
    {
      "epoch": 0.7225765306122449,
      "grad_norm": 0.676493803165307,
      "learning_rate": 7.1114770874558e-06,
      "loss": 0.32912750244140626,
      "num_input_tokens_seen": 339838352,
      "step": 5665,
      "train_runtime": 72872.0773,
      "train_tokens_per_second": 4663.492
    },
    {
      "epoch": 0.7232142857142857,
      "grad_norm": 0.6124251596984999,
      "learning_rate": 7.1069356546389555e-06,
      "loss": 0.3122929573059082,
      "num_input_tokens_seen": 340124272,
      "step": 5670,
      "train_runtime": 72934.4532,
      "train_tokens_per_second": 4663.424
    },
    {
      "epoch": 0.7238520408163265,
      "grad_norm": 0.6551044599285598,
      "learning_rate": 7.1023921073650325e-06,
      "loss": 0.3367321491241455,
      "num_input_tokens_seen": 340427880,
      "step": 5675,
      "train_runtime": 73002.8802,
      "train_tokens_per_second": 4663.212
    },
    {
      "epoch": 0.7244897959183674,
      "grad_norm": 0.5838697926266972,
      "learning_rate": 7.0978464501938e-06,
      "loss": 0.3130692958831787,
      "num_input_tokens_seen": 340725064,
      "step": 5680,
      "train_runtime": 73061.914,
      "train_tokens_per_second": 4663.511
    },
    {
      "epoch": 0.7251275510204082,
      "grad_norm": 0.655893167777473,
      "learning_rate": 7.093298687687141e-06,
      "loss": 0.3281263828277588,
      "num_input_tokens_seen": 341012904,
      "step": 5685,
      "train_runtime": 73122.1396,
      "train_tokens_per_second": 4663.607
    },
    {
      "epoch": 0.7257653061224489,
      "grad_norm": 0.579608632469256,
      "learning_rate": 7.088748824409053e-06,
      "loss": 0.32417736053466795,
      "num_input_tokens_seen": 341309872,
      "step": 5690,
      "train_runtime": 73187.1886,
      "train_tokens_per_second": 4663.519
    },
    {
      "epoch": 0.7264030612244898,
      "grad_norm": 0.6737364289551899,
      "learning_rate": 7.084196864925643e-06,
      "loss": 0.3085744619369507,
      "num_input_tokens_seen": 341604184,
      "step": 5695,
      "train_runtime": 73250.1928,
      "train_tokens_per_second": 4663.526
    },
    {
      "epoch": 0.7270408163265306,
      "grad_norm": 0.5394877761343759,
      "learning_rate": 7.079642813805118e-06,
      "loss": 0.2835739850997925,
      "num_input_tokens_seen": 341909216,
      "step": 5700,
      "train_runtime": 73309.7341,
      "train_tokens_per_second": 4663.899
    },
    {
      "epoch": 0.7276785714285714,
      "grad_norm": 0.5693239010320096,
      "learning_rate": 7.075086675617788e-06,
      "loss": 0.31586780548095705,
      "num_input_tokens_seen": 342216536,
      "step": 5705,
      "train_runtime": 73368.6682,
      "train_tokens_per_second": 4664.342
    },
    {
      "epoch": 0.7283163265306123,
      "grad_norm": 0.6293092230529511,
      "learning_rate": 7.070528454936054e-06,
      "loss": 0.3154031753540039,
      "num_input_tokens_seen": 342510720,
      "step": 5710,
      "train_runtime": 73432.9702,
      "train_tokens_per_second": 4664.263
    },
    {
      "epoch": 0.7289540816326531,
      "grad_norm": 0.6247204617519868,
      "learning_rate": 7.065968156334413e-06,
      "loss": 0.3413267135620117,
      "num_input_tokens_seen": 342819008,
      "step": 5715,
      "train_runtime": 73505.2332,
      "train_tokens_per_second": 4663.872
    },
    {
      "epoch": 0.7295918367346939,
      "grad_norm": 0.5607540231944393,
      "learning_rate": 7.061405784389438e-06,
      "loss": 0.2918368101119995,
      "num_input_tokens_seen": 343129288,
      "step": 5720,
      "train_runtime": 73556.2738,
      "train_tokens_per_second": 4664.854
    },
    {
      "epoch": 0.7302295918367347,
      "grad_norm": 0.5737415637644253,
      "learning_rate": 7.05684134367979e-06,
      "loss": 0.32287158966064455,
      "num_input_tokens_seen": 343442576,
      "step": 5725,
      "train_runtime": 73619.9869,
      "train_tokens_per_second": 4665.072
    },
    {
      "epoch": 0.7308673469387755,
      "grad_norm": 0.6516254960633631,
      "learning_rate": 7.052274838786204e-06,
      "loss": 0.31599087715148927,
      "num_input_tokens_seen": 343729528,
      "step": 5730,
      "train_runtime": 73689.7979,
      "train_tokens_per_second": 4664.547
    },
    {
      "epoch": 0.7315051020408163,
      "grad_norm": 0.6419116302607135,
      "learning_rate": 7.047706274291488e-06,
      "loss": 0.33379082679748534,
      "num_input_tokens_seen": 344020672,
      "step": 5735,
      "train_runtime": 73747.1483,
      "train_tokens_per_second": 4664.867
    },
    {
      "epoch": 0.7321428571428571,
      "grad_norm": 0.6429067893548648,
      "learning_rate": 7.043135654780513e-06,
      "loss": 0.31350982189178467,
      "num_input_tokens_seen": 344330088,
      "step": 5740,
      "train_runtime": 73808.7612,
      "train_tokens_per_second": 4665.166
    },
    {
      "epoch": 0.732780612244898,
      "grad_norm": 0.6412908482581101,
      "learning_rate": 7.038562984840216e-06,
      "loss": 0.31070795059204104,
      "num_input_tokens_seen": 344616200,
      "step": 5745,
      "train_runtime": 73882.4156,
      "train_tokens_per_second": 4664.387
    },
    {
      "epoch": 0.7334183673469388,
      "grad_norm": 0.623040144230442,
      "learning_rate": 7.033988269059593e-06,
      "loss": 0.31840083599090574,
      "num_input_tokens_seen": 344917440,
      "step": 5750,
      "train_runtime": 73953.8959,
      "train_tokens_per_second": 4663.952
    },
    {
      "epoch": 0.7340561224489796,
      "grad_norm": 0.628578861280673,
      "learning_rate": 7.029411512029687e-06,
      "loss": 0.3089389085769653,
      "num_input_tokens_seen": 345217608,
      "step": 5755,
      "train_runtime": 74024.1975,
      "train_tokens_per_second": 4663.578
    },
    {
      "epoch": 0.7346938775510204,
      "grad_norm": 0.5606091803259069,
      "learning_rate": 7.024832718343594e-06,
      "loss": 0.29696319103240965,
      "num_input_tokens_seen": 345494232,
      "step": 5760,
      "train_runtime": 74084.842,
      "train_tokens_per_second": 4663.494
    },
    {
      "epoch": 0.7353316326530612,
      "grad_norm": 0.6110013208452163,
      "learning_rate": 7.020251892596456e-06,
      "loss": 0.33933837413787843,
      "num_input_tokens_seen": 345773888,
      "step": 5765,
      "train_runtime": 74147.9634,
      "train_tokens_per_second": 4663.296
    },
    {
      "epoch": 0.735969387755102,
      "grad_norm": 0.5937391578055509,
      "learning_rate": 7.015669039385449e-06,
      "loss": 0.32836318016052246,
      "num_input_tokens_seen": 346081664,
      "step": 5770,
      "train_runtime": 74214.095,
      "train_tokens_per_second": 4663.288
    },
    {
      "epoch": 0.7366071428571429,
      "grad_norm": 0.5767933191935242,
      "learning_rate": 7.011084163309786e-06,
      "loss": 0.3163835287094116,
      "num_input_tokens_seen": 346388624,
      "step": 5775,
      "train_runtime": 74295.7532,
      "train_tokens_per_second": 4662.294
    },
    {
      "epoch": 0.7372448979591837,
      "grad_norm": 0.6200013210658539,
      "learning_rate": 7.006497268970712e-06,
      "loss": 0.3465720176696777,
      "num_input_tokens_seen": 346690048,
      "step": 5780,
      "train_runtime": 74362.9628,
      "train_tokens_per_second": 4662.133
    },
    {
      "epoch": 0.7378826530612245,
      "grad_norm": 0.5860895337522811,
      "learning_rate": 7.001908360971495e-06,
      "loss": 0.30460400581359864,
      "num_input_tokens_seen": 346979944,
      "step": 5785,
      "train_runtime": 74427.8912,
      "train_tokens_per_second": 4661.961
    },
    {
      "epoch": 0.7385204081632653,
      "grad_norm": 0.5893342038131894,
      "learning_rate": 6.997317443917424e-06,
      "loss": 0.330953311920166,
      "num_input_tokens_seen": 347275248,
      "step": 5790,
      "train_runtime": 74495.4865,
      "train_tokens_per_second": 4661.695
    },
    {
      "epoch": 0.7391581632653061,
      "grad_norm": 0.6181135044247561,
      "learning_rate": 6.9927245224158066e-06,
      "loss": 0.29505696296691897,
      "num_input_tokens_seen": 347580432,
      "step": 5795,
      "train_runtime": 74559.5101,
      "train_tokens_per_second": 4661.785
    },
    {
      "epoch": 0.7397959183673469,
      "grad_norm": 0.5616472690246123,
      "learning_rate": 6.988129601075961e-06,
      "loss": 0.32417964935302734,
      "num_input_tokens_seen": 347877008,
      "step": 5800,
      "train_runtime": 74612.3328,
      "train_tokens_per_second": 4662.46
    },
    {
      "epoch": 0.7404336734693877,
      "grad_norm": 0.5771117709451439,
      "learning_rate": 6.983532684509208e-06,
      "loss": 0.31626770496368406,
      "num_input_tokens_seen": 348157464,
      "step": 5805,
      "train_runtime": 74682.9093,
      "train_tokens_per_second": 4661.809
    },
    {
      "epoch": 0.7410714285714286,
      "grad_norm": 0.6524021867150832,
      "learning_rate": 6.978933777328878e-06,
      "loss": 0.3219144344329834,
      "num_input_tokens_seen": 348450544,
      "step": 5810,
      "train_runtime": 74752.5374,
      "train_tokens_per_second": 4661.388
    },
    {
      "epoch": 0.7417091836734694,
      "grad_norm": 0.8802892201540943,
      "learning_rate": 6.974332884150292e-06,
      "loss": 0.3003734588623047,
      "num_input_tokens_seen": 348748816,
      "step": 5815,
      "train_runtime": 74812.6672,
      "train_tokens_per_second": 4661.628
    },
    {
      "epoch": 0.7423469387755102,
      "grad_norm": 0.5492570085094902,
      "learning_rate": 6.96973000959077e-06,
      "loss": 0.32568812370300293,
      "num_input_tokens_seen": 349044736,
      "step": 5820,
      "train_runtime": 74875.5254,
      "train_tokens_per_second": 4661.667
    },
    {
      "epoch": 0.7429846938775511,
      "grad_norm": 0.6516981459950403,
      "learning_rate": 6.965125158269619e-06,
      "loss": 0.33084864616394044,
      "num_input_tokens_seen": 349337304,
      "step": 5825,
      "train_runtime": 74948.0739,
      "train_tokens_per_second": 4661.058
    },
    {
      "epoch": 0.7436224489795918,
      "grad_norm": 0.5294860523386414,
      "learning_rate": 6.9605183348081265e-06,
      "loss": 0.31689841747283937,
      "num_input_tokens_seen": 349647744,
      "step": 5830,
      "train_runtime": 75008.9694,
      "train_tokens_per_second": 4661.412
    },
    {
      "epoch": 0.7442602040816326,
      "grad_norm": 0.6739558272302397,
      "learning_rate": 6.955909543829564e-06,
      "loss": 0.32897903919219973,
      "num_input_tokens_seen": 349938208,
      "step": 5835,
      "train_runtime": 75066.8762,
      "train_tokens_per_second": 4661.686
    },
    {
      "epoch": 0.7448979591836735,
      "grad_norm": 0.6373827784654852,
      "learning_rate": 6.951298789959172e-06,
      "loss": 0.31219475269317626,
      "num_input_tokens_seen": 350225080,
      "step": 5840,
      "train_runtime": 75124.6662,
      "train_tokens_per_second": 4661.919
    },
    {
      "epoch": 0.7455357142857143,
      "grad_norm": 0.5835360397196525,
      "learning_rate": 6.946686077824167e-06,
      "loss": 0.3332829475402832,
      "num_input_tokens_seen": 350524048,
      "step": 5845,
      "train_runtime": 75188.8705,
      "train_tokens_per_second": 4661.914
    },
    {
      "epoch": 0.7461734693877551,
      "grad_norm": 0.6520631081854955,
      "learning_rate": 6.942071412053728e-06,
      "loss": 0.2965895175933838,
      "num_input_tokens_seen": 350824584,
      "step": 5850,
      "train_runtime": 75248.6658,
      "train_tokens_per_second": 4662.203
    },
    {
      "epoch": 0.7468112244897959,
      "grad_norm": 0.6599584764606783,
      "learning_rate": 6.937454797278991e-06,
      "loss": 0.32917354106903074,
      "num_input_tokens_seen": 351137552,
      "step": 5855,
      "train_runtime": 75312.9623,
      "train_tokens_per_second": 4662.379
    },
    {
      "epoch": 0.7474489795918368,
      "grad_norm": 0.5688668088307691,
      "learning_rate": 6.932836238133054e-06,
      "loss": 0.3156865358352661,
      "num_input_tokens_seen": 351434200,
      "step": 5860,
      "train_runtime": 75378.7756,
      "train_tokens_per_second": 4662.243
    },
    {
      "epoch": 0.7480867346938775,
      "grad_norm": 0.691188150805711,
      "learning_rate": 6.928215739250963e-06,
      "loss": 0.29285922050476076,
      "num_input_tokens_seen": 351721312,
      "step": 5865,
      "train_runtime": 75442.2192,
      "train_tokens_per_second": 4662.128
    },
    {
      "epoch": 0.7487244897959183,
      "grad_norm": 0.6367175792798876,
      "learning_rate": 6.923593305269711e-06,
      "loss": 0.3241368293762207,
      "num_input_tokens_seen": 352017416,
      "step": 5870,
      "train_runtime": 75505.6431,
      "train_tokens_per_second": 4662.134
    },
    {
      "epoch": 0.7493622448979592,
      "grad_norm": 0.5737297755630368,
      "learning_rate": 6.9189689408282345e-06,
      "loss": 0.3378509521484375,
      "num_input_tokens_seen": 352320464,
      "step": 5875,
      "train_runtime": 75570.3406,
      "train_tokens_per_second": 4662.153
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6365716099382133,
      "learning_rate": 6.914342650567405e-06,
      "loss": 0.3078855037689209,
      "num_input_tokens_seen": 352612088,
      "step": 5880,
      "train_runtime": 75630.9608,
      "train_tokens_per_second": 4662.272
    },
    {
      "epoch": 0.7506377551020408,
      "grad_norm": 0.585545534743707,
      "learning_rate": 6.9097144391300285e-06,
      "loss": 0.318976092338562,
      "num_input_tokens_seen": 352905800,
      "step": 5885,
      "train_runtime": 75697.1376,
      "train_tokens_per_second": 4662.076
    },
    {
      "epoch": 0.7512755102040817,
      "grad_norm": 0.5863621066790211,
      "learning_rate": 6.905084311160837e-06,
      "loss": 0.32762937545776366,
      "num_input_tokens_seen": 353216736,
      "step": 5890,
      "train_runtime": 75764.3689,
      "train_tokens_per_second": 4662.043
    },
    {
      "epoch": 0.7519132653061225,
      "grad_norm": 0.6353532987193916,
      "learning_rate": 6.90045227130649e-06,
      "loss": 0.32003273963928225,
      "num_input_tokens_seen": 353530984,
      "step": 5895,
      "train_runtime": 75832.3062,
      "train_tokens_per_second": 4662.01
    },
    {
      "epoch": 0.7525510204081632,
      "grad_norm": 0.6017119626901829,
      "learning_rate": 6.8958183242155615e-06,
      "loss": 0.29575016498565676,
      "num_input_tokens_seen": 353840848,
      "step": 5900,
      "train_runtime": 75892.0809,
      "train_tokens_per_second": 4662.421
    },
    {
      "epoch": 0.7531887755102041,
      "grad_norm": 0.6034869442022718,
      "learning_rate": 6.891182474538539e-06,
      "loss": 0.3226417064666748,
      "num_input_tokens_seen": 354129328,
      "step": 5905,
      "train_runtime": 75967.9806,
      "train_tokens_per_second": 4661.56
    },
    {
      "epoch": 0.7538265306122449,
      "grad_norm": 0.6269025894057103,
      "learning_rate": 6.886544726927826e-06,
      "loss": 0.32311596870422366,
      "num_input_tokens_seen": 354431368,
      "step": 5910,
      "train_runtime": 76043.1807,
      "train_tokens_per_second": 4660.922
    },
    {
      "epoch": 0.7544642857142857,
      "grad_norm": 0.5905587693464358,
      "learning_rate": 6.881905086037721e-06,
      "loss": 0.28950068950653074,
      "num_input_tokens_seen": 354746904,
      "step": 5915,
      "train_runtime": 76101.1077,
      "train_tokens_per_second": 4661.521
    },
    {
      "epoch": 0.7551020408163265,
      "grad_norm": 0.6341417919505897,
      "learning_rate": 6.877263556524432e-06,
      "loss": 0.333311939239502,
      "num_input_tokens_seen": 355047584,
      "step": 5920,
      "train_runtime": 76168.9611,
      "train_tokens_per_second": 4661.316
    },
    {
      "epoch": 0.7557397959183674,
      "grad_norm": 0.5848267786266075,
      "learning_rate": 6.872620143046056e-06,
      "loss": 0.30979280471801757,
      "num_input_tokens_seen": 355342328,
      "step": 5925,
      "train_runtime": 76233.81,
      "train_tokens_per_second": 4661.217
    },
    {
      "epoch": 0.7563775510204082,
      "grad_norm": 0.5941793485823651,
      "learning_rate": 6.867974850262582e-06,
      "loss": 0.3388402462005615,
      "num_input_tokens_seen": 355650912,
      "step": 5930,
      "train_runtime": 76289.5218,
      "train_tokens_per_second": 4661.858
    },
    {
      "epoch": 0.7570153061224489,
      "grad_norm": 0.5416742865479929,
      "learning_rate": 6.863327682835887e-06,
      "loss": 0.31301352977752683,
      "num_input_tokens_seen": 355972024,
      "step": 5935,
      "train_runtime": 76343.9285,
      "train_tokens_per_second": 4662.742
    },
    {
      "epoch": 0.7576530612244898,
      "grad_norm": 0.5944668862745454,
      "learning_rate": 6.858678645429728e-06,
      "loss": 0.3231534957885742,
      "num_input_tokens_seen": 356273760,
      "step": 5940,
      "train_runtime": 76416.5962,
      "train_tokens_per_second": 4662.256
    },
    {
      "epoch": 0.7582908163265306,
      "grad_norm": 0.60876421575269,
      "learning_rate": 6.85402774270974e-06,
      "loss": 0.3035454034805298,
      "num_input_tokens_seen": 356561896,
      "step": 5945,
      "train_runtime": 76476.3086,
      "train_tokens_per_second": 4662.384
    },
    {
      "epoch": 0.7589285714285714,
      "grad_norm": 0.584356004877175,
      "learning_rate": 6.849374979343426e-06,
      "loss": 0.33470563888549804,
      "num_input_tokens_seen": 356883248,
      "step": 5950,
      "train_runtime": 76543.6473,
      "train_tokens_per_second": 4662.48
    },
    {
      "epoch": 0.7595663265306123,
      "grad_norm": 0.6735457874399653,
      "learning_rate": 6.84472036000016e-06,
      "loss": 0.3113273620605469,
      "num_input_tokens_seen": 357186760,
      "step": 5955,
      "train_runtime": 76604.0499,
      "train_tokens_per_second": 4662.766
    },
    {
      "epoch": 0.7602040816326531,
      "grad_norm": 0.5398270797593189,
      "learning_rate": 6.840063889351177e-06,
      "loss": 0.29850049018859864,
      "num_input_tokens_seen": 357493416,
      "step": 5960,
      "train_runtime": 76666.5356,
      "train_tokens_per_second": 4662.966
    },
    {
      "epoch": 0.7608418367346939,
      "grad_norm": 0.637683964437538,
      "learning_rate": 6.835405572069572e-06,
      "loss": 0.3150575876235962,
      "num_input_tokens_seen": 357801504,
      "step": 5965,
      "train_runtime": 76730.531,
      "train_tokens_per_second": 4663.092
    },
    {
      "epoch": 0.7614795918367347,
      "grad_norm": 0.5882936203541248,
      "learning_rate": 6.830745412830291e-06,
      "loss": 0.3131128787994385,
      "num_input_tokens_seen": 358090144,
      "step": 5970,
      "train_runtime": 76793.9281,
      "train_tokens_per_second": 4663.001
    },
    {
      "epoch": 0.7621173469387755,
      "grad_norm": 0.5407032572429494,
      "learning_rate": 6.826083416310129e-06,
      "loss": 0.3144979000091553,
      "num_input_tokens_seen": 358382488,
      "step": 5975,
      "train_runtime": 76847.6766,
      "train_tokens_per_second": 4663.544
    },
    {
      "epoch": 0.7627551020408163,
      "grad_norm": 0.6315366716254545,
      "learning_rate": 6.821419587187727e-06,
      "loss": 0.3227547645568848,
      "num_input_tokens_seen": 358660592,
      "step": 5980,
      "train_runtime": 76912.7466,
      "train_tokens_per_second": 4663.214
    },
    {
      "epoch": 0.7633928571428571,
      "grad_norm": 0.6037809244004879,
      "learning_rate": 6.816753930143558e-06,
      "loss": 0.31183757781982424,
      "num_input_tokens_seen": 358960128,
      "step": 5985,
      "train_runtime": 76979.2434,
      "train_tokens_per_second": 4663.077
    },
    {
      "epoch": 0.764030612244898,
      "grad_norm": 0.6946191668857217,
      "learning_rate": 6.812086449859941e-06,
      "loss": 0.311761212348938,
      "num_input_tokens_seen": 359267296,
      "step": 5990,
      "train_runtime": 77031.7524,
      "train_tokens_per_second": 4663.886
    },
    {
      "epoch": 0.7646683673469388,
      "grad_norm": 0.565472235354576,
      "learning_rate": 6.807417151021015e-06,
      "loss": 0.3410171985626221,
      "num_input_tokens_seen": 359558824,
      "step": 5995,
      "train_runtime": 77092.7315,
      "train_tokens_per_second": 4663.978
    },
    {
      "epoch": 0.7653061224489796,
      "grad_norm": 0.5555826911790634,
      "learning_rate": 6.802746038312749e-06,
      "loss": 0.3195962905883789,
      "num_input_tokens_seen": 359868936,
      "step": 6000,
      "train_runtime": 77155.0352,
      "train_tokens_per_second": 4664.231
    },
    {
      "epoch": 0.7659438775510204,
      "grad_norm": 0.5342150038853042,
      "learning_rate": 6.79807311642293e-06,
      "loss": 0.31259951591491697,
      "num_input_tokens_seen": 360164520,
      "step": 6005,
      "train_runtime": 77222.711,
      "train_tokens_per_second": 4663.971
    },
    {
      "epoch": 0.7665816326530612,
      "grad_norm": 0.6055030347230119,
      "learning_rate": 6.793398390041159e-06,
      "loss": 0.3051009654998779,
      "num_input_tokens_seen": 360474088,
      "step": 6010,
      "train_runtime": 77269.2801,
      "train_tokens_per_second": 4665.167
    },
    {
      "epoch": 0.767219387755102,
      "grad_norm": 0.6039069580622654,
      "learning_rate": 6.788721863858856e-06,
      "loss": 0.3037103176116943,
      "num_input_tokens_seen": 360754400,
      "step": 6015,
      "train_runtime": 77331.8812,
      "train_tokens_per_second": 4665.015
    },
    {
      "epoch": 0.7678571428571429,
      "grad_norm": 0.5920840464539026,
      "learning_rate": 6.784043542569236e-06,
      "loss": 0.3146142721176147,
      "num_input_tokens_seen": 361050456,
      "step": 6020,
      "train_runtime": 77403.7861,
      "train_tokens_per_second": 4664.506
    },
    {
      "epoch": 0.7684948979591837,
      "grad_norm": 0.6063580119658678,
      "learning_rate": 6.7793634308673265e-06,
      "loss": 0.3116027593612671,
      "num_input_tokens_seen": 361360328,
      "step": 6025,
      "train_runtime": 77473.4802,
      "train_tokens_per_second": 4664.31
    },
    {
      "epoch": 0.7691326530612245,
      "grad_norm": 0.6070484223576093,
      "learning_rate": 6.7746815334499426e-06,
      "loss": 0.30540153980255125,
      "num_input_tokens_seen": 361662224,
      "step": 6030,
      "train_runtime": 77545.4846,
      "train_tokens_per_second": 4663.872
    },
    {
      "epoch": 0.7697704081632653,
      "grad_norm": 0.5792044891291048,
      "learning_rate": 6.7699978550156954e-06,
      "loss": 0.31980509757995607,
      "num_input_tokens_seen": 361964112,
      "step": 6035,
      "train_runtime": 77606.0584,
      "train_tokens_per_second": 4664.122
    },
    {
      "epoch": 0.7704081632653061,
      "grad_norm": 0.5202278457370635,
      "learning_rate": 6.765312400264985e-06,
      "loss": 0.3199894428253174,
      "num_input_tokens_seen": 362279936,
      "step": 6040,
      "train_runtime": 77659.8235,
      "train_tokens_per_second": 4664.96
    },
    {
      "epoch": 0.7710459183673469,
      "grad_norm": 0.5867446845865297,
      "learning_rate": 6.760625173899992e-06,
      "loss": 0.31152822971343996,
      "num_input_tokens_seen": 362570056,
      "step": 6045,
      "train_runtime": 77726.1988,
      "train_tokens_per_second": 4664.708
    },
    {
      "epoch": 0.7716836734693877,
      "grad_norm": 0.606783979605095,
      "learning_rate": 6.755936180624674e-06,
      "loss": 0.29929795265197756,
      "num_input_tokens_seen": 362853320,
      "step": 6050,
      "train_runtime": 77783.7031,
      "train_tokens_per_second": 4664.902
    },
    {
      "epoch": 0.7723214285714286,
      "grad_norm": 0.5882279974676495,
      "learning_rate": 6.751245425144765e-06,
      "loss": 0.3252225875854492,
      "num_input_tokens_seen": 363154000,
      "step": 6055,
      "train_runtime": 77854.9987,
      "train_tokens_per_second": 4664.492
    },
    {
      "epoch": 0.7729591836734694,
      "grad_norm": 0.6204455476082986,
      "learning_rate": 6.746552912167766e-06,
      "loss": 0.2846733570098877,
      "num_input_tokens_seen": 363454440,
      "step": 6060,
      "train_runtime": 77915.8005,
      "train_tokens_per_second": 4664.708
    },
    {
      "epoch": 0.7735969387755102,
      "grad_norm": 0.6324651027545711,
      "learning_rate": 6.741858646402941e-06,
      "loss": 0.3407684326171875,
      "num_input_tokens_seen": 363751104,
      "step": 6065,
      "train_runtime": 77977.5438,
      "train_tokens_per_second": 4664.819
    },
    {
      "epoch": 0.7742346938775511,
      "grad_norm": 0.5352416376404033,
      "learning_rate": 6.737162632561311e-06,
      "loss": 0.29468593597412107,
      "num_input_tokens_seen": 364061960,
      "step": 6070,
      "train_runtime": 78040.1638,
      "train_tokens_per_second": 4665.059
    },
    {
      "epoch": 0.7748724489795918,
      "grad_norm": 0.6185847427444751,
      "learning_rate": 6.732464875355657e-06,
      "loss": 0.3351437568664551,
      "num_input_tokens_seen": 364357224,
      "step": 6075,
      "train_runtime": 78101.9157,
      "train_tokens_per_second": 4665.151
    },
    {
      "epoch": 0.7755102040816326,
      "grad_norm": 0.6091453075723209,
      "learning_rate": 6.7277653795005085e-06,
      "loss": 0.32475030422210693,
      "num_input_tokens_seen": 364642304,
      "step": 6080,
      "train_runtime": 78171.4356,
      "train_tokens_per_second": 4664.649
    },
    {
      "epoch": 0.7761479591836735,
      "grad_norm": 0.6654318166872217,
      "learning_rate": 6.723064149712131e-06,
      "loss": 0.3222649097442627,
      "num_input_tokens_seen": 364942760,
      "step": 6085,
      "train_runtime": 78236.8926,
      "train_tokens_per_second": 4664.587
    },
    {
      "epoch": 0.7767857142857143,
      "grad_norm": 0.6193247613367058,
      "learning_rate": 6.718361190708542e-06,
      "loss": 0.3156604290008545,
      "num_input_tokens_seen": 365259032,
      "step": 6090,
      "train_runtime": 78293.4974,
      "train_tokens_per_second": 4665.254
    },
    {
      "epoch": 0.7774234693877551,
      "grad_norm": 0.665726352101644,
      "learning_rate": 6.713656507209489e-06,
      "loss": 0.33889386653900144,
      "num_input_tokens_seen": 365556656,
      "step": 6095,
      "train_runtime": 78342.3526,
      "train_tokens_per_second": 4666.143
    },
    {
      "epoch": 0.7780612244897959,
      "grad_norm": 0.6369863038873577,
      "learning_rate": 6.708950103936448e-06,
      "loss": 0.3098180294036865,
      "num_input_tokens_seen": 365845600,
      "step": 6100,
      "train_runtime": 78402.5516,
      "train_tokens_per_second": 4666.246
    },
    {
      "epoch": 0.7786989795918368,
      "grad_norm": 0.5770540204745025,
      "learning_rate": 6.704241985612625e-06,
      "loss": 0.32828497886657715,
      "num_input_tokens_seen": 366150704,
      "step": 6105,
      "train_runtime": 78470.2801,
      "train_tokens_per_second": 4666.107
    },
    {
      "epoch": 0.7793367346938775,
      "grad_norm": 0.5658637392761506,
      "learning_rate": 6.699532156962945e-06,
      "loss": 0.31436471939086913,
      "num_input_tokens_seen": 366448944,
      "step": 6110,
      "train_runtime": 78530.699,
      "train_tokens_per_second": 4666.315
    },
    {
      "epoch": 0.7799744897959183,
      "grad_norm": 0.6531724313969975,
      "learning_rate": 6.6948206227140525e-06,
      "loss": 0.33806796073913575,
      "num_input_tokens_seen": 366749016,
      "step": 6115,
      "train_runtime": 78591.4583,
      "train_tokens_per_second": 4666.525
    },
    {
      "epoch": 0.7806122448979592,
      "grad_norm": 0.705172393284724,
      "learning_rate": 6.690107387594298e-06,
      "loss": 0.3181670427322388,
      "num_input_tokens_seen": 367047544,
      "step": 6120,
      "train_runtime": 78653.9582,
      "train_tokens_per_second": 4666.612
    },
    {
      "epoch": 0.78125,
      "grad_norm": 0.6320122617674263,
      "learning_rate": 6.685392456333744e-06,
      "loss": 0.29224820137023927,
      "num_input_tokens_seen": 367342096,
      "step": 6125,
      "train_runtime": 78721.4587,
      "train_tokens_per_second": 4666.353
    },
    {
      "epoch": 0.7818877551020408,
      "grad_norm": 0.5751095955705227,
      "learning_rate": 6.680675833664151e-06,
      "loss": 0.34242610931396483,
      "num_input_tokens_seen": 367646520,
      "step": 6130,
      "train_runtime": 78776.8337,
      "train_tokens_per_second": 4666.937
    },
    {
      "epoch": 0.7825255102040817,
      "grad_norm": 0.7247333049487453,
      "learning_rate": 6.67595752431898e-06,
      "loss": 0.3134866714477539,
      "num_input_tokens_seen": 367920896,
      "step": 6135,
      "train_runtime": 78844.1279,
      "train_tokens_per_second": 4666.434
    },
    {
      "epoch": 0.7831632653061225,
      "grad_norm": 0.6181022749426421,
      "learning_rate": 6.671237533033388e-06,
      "loss": 0.3164665699005127,
      "num_input_tokens_seen": 368226224,
      "step": 6140,
      "train_runtime": 78894.1503,
      "train_tokens_per_second": 4667.345
    },
    {
      "epoch": 0.7838010204081632,
      "grad_norm": 0.6088077988506396,
      "learning_rate": 6.66651586454421e-06,
      "loss": 0.3136889934539795,
      "num_input_tokens_seen": 368522792,
      "step": 6145,
      "train_runtime": 78959.7175,
      "train_tokens_per_second": 4667.225
    },
    {
      "epoch": 0.7844387755102041,
      "grad_norm": 0.5976310044878483,
      "learning_rate": 6.661792523589972e-06,
      "loss": 0.29033560752868653,
      "num_input_tokens_seen": 368821224,
      "step": 6150,
      "train_runtime": 79019.8184,
      "train_tokens_per_second": 4667.452
    },
    {
      "epoch": 0.7850765306122449,
      "grad_norm": 0.6629498888375844,
      "learning_rate": 6.65706751491088e-06,
      "loss": 0.3224802494049072,
      "num_input_tokens_seen": 369100760,
      "step": 6155,
      "train_runtime": 79083.1157,
      "train_tokens_per_second": 4667.251
    },
    {
      "epoch": 0.7857142857142857,
      "grad_norm": 0.6142346816031169,
      "learning_rate": 6.652340843248806e-06,
      "loss": 0.3112659454345703,
      "num_input_tokens_seen": 369393584,
      "step": 6160,
      "train_runtime": 79141.4529,
      "train_tokens_per_second": 4667.511
    },
    {
      "epoch": 0.7863520408163265,
      "grad_norm": 0.5517826119783504,
      "learning_rate": 6.647612513347297e-06,
      "loss": 0.2997623682022095,
      "num_input_tokens_seen": 369672280,
      "step": 6165,
      "train_runtime": 79196.9514,
      "train_tokens_per_second": 4667.759
    },
    {
      "epoch": 0.7869897959183674,
      "grad_norm": 0.604744948814889,
      "learning_rate": 6.642882529951561e-06,
      "loss": 0.30200629234313964,
      "num_input_tokens_seen": 369972056,
      "step": 6170,
      "train_runtime": 79268.6732,
      "train_tokens_per_second": 4667.317
    },
    {
      "epoch": 0.7876275510204082,
      "grad_norm": 0.5928238184254466,
      "learning_rate": 6.638150897808469e-06,
      "loss": 0.3180851936340332,
      "num_input_tokens_seen": 370285480,
      "step": 6175,
      "train_runtime": 79337.6191,
      "train_tokens_per_second": 4667.212
    },
    {
      "epoch": 0.7882653061224489,
      "grad_norm": 0.5578302357166072,
      "learning_rate": 6.6334176216665406e-06,
      "loss": 0.3170527458190918,
      "num_input_tokens_seen": 370593456,
      "step": 6180,
      "train_runtime": 79399.5846,
      "train_tokens_per_second": 4667.448
    },
    {
      "epoch": 0.7889030612244898,
      "grad_norm": 0.8621658286713906,
      "learning_rate": 6.628682706275953e-06,
      "loss": 0.33642194271087644,
      "num_input_tokens_seen": 370893072,
      "step": 6185,
      "train_runtime": 79467.2199,
      "train_tokens_per_second": 4667.246
    },
    {
      "epoch": 0.7895408163265306,
      "grad_norm": 0.5393751716222243,
      "learning_rate": 6.623946156388523e-06,
      "loss": 0.3066478490829468,
      "num_input_tokens_seen": 371208088,
      "step": 6190,
      "train_runtime": 79540.416,
      "train_tokens_per_second": 4666.912
    },
    {
      "epoch": 0.7901785714285714,
      "grad_norm": 0.5671943980487131,
      "learning_rate": 6.619207976757708e-06,
      "loss": 0.2984332799911499,
      "num_input_tokens_seen": 371503888,
      "step": 6195,
      "train_runtime": 79599.2555,
      "train_tokens_per_second": 4667.178
    },
    {
      "epoch": 0.7908163265306123,
      "grad_norm": 0.6903010968290758,
      "learning_rate": 6.614468172138603e-06,
      "loss": 0.3375420093536377,
      "num_input_tokens_seen": 371784424,
      "step": 6200,
      "train_runtime": 79659.4687,
      "train_tokens_per_second": 4667.172
    },
    {
      "epoch": 0.7914540816326531,
      "grad_norm": 0.5678990709503928,
      "learning_rate": 6.609726747287934e-06,
      "loss": 0.3009253978729248,
      "num_input_tokens_seen": 372074512,
      "step": 6205,
      "train_runtime": 79731.8077,
      "train_tokens_per_second": 4666.576
    },
    {
      "epoch": 0.7920918367346939,
      "grad_norm": 0.5884104225346212,
      "learning_rate": 6.60498370696405e-06,
      "loss": 0.3048467874526978,
      "num_input_tokens_seen": 372391472,
      "step": 6210,
      "train_runtime": 79782.987,
      "train_tokens_per_second": 4667.555
    },
    {
      "epoch": 0.7927295918367347,
      "grad_norm": 0.6219725624461221,
      "learning_rate": 6.600239055926923e-06,
      "loss": 0.342468786239624,
      "num_input_tokens_seen": 372690712,
      "step": 6215,
      "train_runtime": 79850.5693,
      "train_tokens_per_second": 4667.352
    },
    {
      "epoch": 0.7933673469387755,
      "grad_norm": 0.5759103556707833,
      "learning_rate": 6.595492798938146e-06,
      "loss": 0.2964707612991333,
      "num_input_tokens_seen": 372980160,
      "step": 6220,
      "train_runtime": 79907.2822,
      "train_tokens_per_second": 4667.662
    },
    {
      "epoch": 0.7940051020408163,
      "grad_norm": 0.6560616953026334,
      "learning_rate": 6.5907449407609145e-06,
      "loss": 0.3294534206390381,
      "num_input_tokens_seen": 373276592,
      "step": 6225,
      "train_runtime": 79971.6459,
      "train_tokens_per_second": 4667.612
    },
    {
      "epoch": 0.7946428571428571,
      "grad_norm": 0.6075933232196977,
      "learning_rate": 6.585995486160038e-06,
      "loss": 0.32330219745635985,
      "num_input_tokens_seen": 373580264,
      "step": 6230,
      "train_runtime": 80031.9246,
      "train_tokens_per_second": 4667.891
    },
    {
      "epoch": 0.795280612244898,
      "grad_norm": 0.6117048961658761,
      "learning_rate": 6.581244439901926e-06,
      "loss": 0.31589932441711427,
      "num_input_tokens_seen": 373881384,
      "step": 6235,
      "train_runtime": 80106.9612,
      "train_tokens_per_second": 4667.277
    },
    {
      "epoch": 0.7959183673469388,
      "grad_norm": 0.6364040012997616,
      "learning_rate": 6.576491806754583e-06,
      "loss": 0.3358692407608032,
      "num_input_tokens_seen": 374181560,
      "step": 6240,
      "train_runtime": 80175.2874,
      "train_tokens_per_second": 4667.044
    },
    {
      "epoch": 0.7965561224489796,
      "grad_norm": 0.6078783651176742,
      "learning_rate": 6.571737591487611e-06,
      "loss": 0.2995163440704346,
      "num_input_tokens_seen": 374481944,
      "step": 6245,
      "train_runtime": 80231.6106,
      "train_tokens_per_second": 4667.511
    },
    {
      "epoch": 0.7971938775510204,
      "grad_norm": 0.5631314811210728,
      "learning_rate": 6.566981798872196e-06,
      "loss": 0.2826210975646973,
      "num_input_tokens_seen": 374787232,
      "step": 6250,
      "train_runtime": 80302.0862,
      "train_tokens_per_second": 4667.217
    },
    {
      "epoch": 0.7978316326530612,
      "grad_norm": 0.6573391038039677,
      "learning_rate": 6.562224433681108e-06,
      "loss": 0.3200786828994751,
      "num_input_tokens_seen": 375072512,
      "step": 6255,
      "train_runtime": 80365.0699,
      "train_tokens_per_second": 4667.109
    },
    {
      "epoch": 0.798469387755102,
      "grad_norm": 0.6524553859053104,
      "learning_rate": 6.557465500688696e-06,
      "loss": 0.3107785940170288,
      "num_input_tokens_seen": 375375488,
      "step": 6260,
      "train_runtime": 80436.89,
      "train_tokens_per_second": 4666.708
    },
    {
      "epoch": 0.7991071428571429,
      "grad_norm": 0.5721551379940883,
      "learning_rate": 6.55270500467088e-06,
      "loss": 0.3176266193389893,
      "num_input_tokens_seen": 375665088,
      "step": 6265,
      "train_runtime": 80497.857,
      "train_tokens_per_second": 4666.771
    },
    {
      "epoch": 0.7997448979591837,
      "grad_norm": 0.570386334621138,
      "learning_rate": 6.5479429504051526e-06,
      "loss": 0.30910840034484866,
      "num_input_tokens_seen": 375975144,
      "step": 6270,
      "train_runtime": 80558.5344,
      "train_tokens_per_second": 4667.105
    },
    {
      "epoch": 0.8003826530612245,
      "grad_norm": 0.6624500993470845,
      "learning_rate": 6.543179342670565e-06,
      "loss": 0.3345738410949707,
      "num_input_tokens_seen": 376273768,
      "step": 6275,
      "train_runtime": 80613.4859,
      "train_tokens_per_second": 4667.628
    },
    {
      "epoch": 0.8010204081632653,
      "grad_norm": 0.5514177978725938,
      "learning_rate": 6.53841418624773e-06,
      "loss": 0.27700390815734866,
      "num_input_tokens_seen": 376570160,
      "step": 6280,
      "train_runtime": 80680.357,
      "train_tokens_per_second": 4667.433
    },
    {
      "epoch": 0.8016581632653061,
      "grad_norm": 0.6103525141332224,
      "learning_rate": 6.53364748591882e-06,
      "loss": 0.30763068199157717,
      "num_input_tokens_seen": 376871272,
      "step": 6285,
      "train_runtime": 80742.579,
      "train_tokens_per_second": 4667.565
    },
    {
      "epoch": 0.8022959183673469,
      "grad_norm": 0.5973259444864897,
      "learning_rate": 6.528879246467546e-06,
      "loss": 0.32227473258972167,
      "num_input_tokens_seen": 377162112,
      "step": 6290,
      "train_runtime": 80812.9176,
      "train_tokens_per_second": 4667.102
    },
    {
      "epoch": 0.8029336734693877,
      "grad_norm": 0.5886180237196361,
      "learning_rate": 6.524109472679172e-06,
      "loss": 0.3222978115081787,
      "num_input_tokens_seen": 377468504,
      "step": 6295,
      "train_runtime": 80882.3703,
      "train_tokens_per_second": 4666.882
    },
    {
      "epoch": 0.8035714285714286,
      "grad_norm": 0.6296556623300213,
      "learning_rate": 6.519338169340499e-06,
      "loss": 0.3378443241119385,
      "num_input_tokens_seen": 377766856,
      "step": 6300,
      "train_runtime": 80946.6364,
      "train_tokens_per_second": 4666.863
    },
    {
      "epoch": 0.8042091836734694,
      "grad_norm": 0.6521390135069829,
      "learning_rate": 6.514565341239861e-06,
      "loss": 0.3549522638320923,
      "num_input_tokens_seen": 378056736,
      "step": 6305,
      "train_runtime": 81012.3663,
      "train_tokens_per_second": 4666.655
    },
    {
      "epoch": 0.8048469387755102,
      "grad_norm": 0.5831745656351125,
      "learning_rate": 6.509790993167129e-06,
      "loss": 0.3268167018890381,
      "num_input_tokens_seen": 378359720,
      "step": 6310,
      "train_runtime": 81079.6852,
      "train_tokens_per_second": 4666.517
    },
    {
      "epoch": 0.8054846938775511,
      "grad_norm": 0.6768644315495205,
      "learning_rate": 6.505015129913689e-06,
      "loss": 0.31839518547058104,
      "num_input_tokens_seen": 378660256,
      "step": 6315,
      "train_runtime": 81154.1649,
      "train_tokens_per_second": 4665.937
    },
    {
      "epoch": 0.8061224489795918,
      "grad_norm": 0.5333202428201802,
      "learning_rate": 6.5002377562724585e-06,
      "loss": 0.3217827320098877,
      "num_input_tokens_seen": 378966872,
      "step": 6320,
      "train_runtime": 81218.2191,
      "train_tokens_per_second": 4666.033
    },
    {
      "epoch": 0.8067602040816326,
      "grad_norm": 0.5724622174452084,
      "learning_rate": 6.495458877037861e-06,
      "loss": 0.30530538558959963,
      "num_input_tokens_seen": 379259880,
      "step": 6325,
      "train_runtime": 81283.7854,
      "train_tokens_per_second": 4665.874
    },
    {
      "epoch": 0.8073979591836735,
      "grad_norm": 0.5754135264649226,
      "learning_rate": 6.490678497005839e-06,
      "loss": 0.3361009359359741,
      "num_input_tokens_seen": 379574600,
      "step": 6330,
      "train_runtime": 81343.1769,
      "train_tokens_per_second": 4666.336
    },
    {
      "epoch": 0.8080357142857143,
      "grad_norm": 0.6991322085025484,
      "learning_rate": 6.485896620973836e-06,
      "loss": 0.3305660724639893,
      "num_input_tokens_seen": 379873224,
      "step": 6335,
      "train_runtime": 81409.2661,
      "train_tokens_per_second": 4666.216
    },
    {
      "epoch": 0.8086734693877551,
      "grad_norm": 0.5648402154997423,
      "learning_rate": 6.481113253740799e-06,
      "loss": 0.31418297290802,
      "num_input_tokens_seen": 380154392,
      "step": 6340,
      "train_runtime": 81473.2564,
      "train_tokens_per_second": 4666.002
    },
    {
      "epoch": 0.8093112244897959,
      "grad_norm": 0.5991247444358954,
      "learning_rate": 6.4763284001071715e-06,
      "loss": 0.32512798309326174,
      "num_input_tokens_seen": 380458984,
      "step": 6345,
      "train_runtime": 81525.0358,
      "train_tokens_per_second": 4666.775
    },
    {
      "epoch": 0.8099489795918368,
      "grad_norm": 0.5988449159868657,
      "learning_rate": 6.4715420648748875e-06,
      "loss": 0.3281686305999756,
      "num_input_tokens_seen": 380748208,
      "step": 6350,
      "train_runtime": 81587.4934,
      "train_tokens_per_second": 4666.747
    },
    {
      "epoch": 0.8105867346938775,
      "grad_norm": 0.5806090235931327,
      "learning_rate": 6.46675425284737e-06,
      "loss": 0.3032946825027466,
      "num_input_tokens_seen": 381038696,
      "step": 6355,
      "train_runtime": 81654.954,
      "train_tokens_per_second": 4666.449
    },
    {
      "epoch": 0.8112244897959183,
      "grad_norm": 0.6360053599256708,
      "learning_rate": 6.461964968829521e-06,
      "loss": 0.3584984064102173,
      "num_input_tokens_seen": 381344440,
      "step": 6360,
      "train_runtime": 81729.3812,
      "train_tokens_per_second": 4665.941
    },
    {
      "epoch": 0.8118622448979592,
      "grad_norm": 0.53782709800568,
      "learning_rate": 6.457174217627722e-06,
      "loss": 0.31561667919158937,
      "num_input_tokens_seen": 381634800,
      "step": 6365,
      "train_runtime": 81783.7636,
      "train_tokens_per_second": 4666.388
    },
    {
      "epoch": 0.8125,
      "grad_norm": 0.585889169753346,
      "learning_rate": 6.452382004049829e-06,
      "loss": 0.3004056453704834,
      "num_input_tokens_seen": 381921968,
      "step": 6370,
      "train_runtime": 81844.6119,
      "train_tokens_per_second": 4666.428
    },
    {
      "epoch": 0.8131377551020408,
      "grad_norm": 0.5274551192749435,
      "learning_rate": 6.447588332905159e-06,
      "loss": 0.3129526376724243,
      "num_input_tokens_seen": 382232480,
      "step": 6375,
      "train_runtime": 81904.6225,
      "train_tokens_per_second": 4666.8
    },
    {
      "epoch": 0.8137755102040817,
      "grad_norm": 0.5857955769609019,
      "learning_rate": 6.4427932090044975e-06,
      "loss": 0.3357415676116943,
      "num_input_tokens_seen": 382532080,
      "step": 6380,
      "train_runtime": 81965.6802,
      "train_tokens_per_second": 4666.979
    },
    {
      "epoch": 0.8144132653061225,
      "grad_norm": 0.6156441051791702,
      "learning_rate": 6.437996637160086e-06,
      "loss": 0.30234622955322266,
      "num_input_tokens_seen": 382833360,
      "step": 6385,
      "train_runtime": 82033.0697,
      "train_tokens_per_second": 4666.817
    },
    {
      "epoch": 0.8150510204081632,
      "grad_norm": 0.6447229491703402,
      "learning_rate": 6.433198622185621e-06,
      "loss": 0.3164085388183594,
      "num_input_tokens_seen": 383129024,
      "step": 6390,
      "train_runtime": 82102.0275,
      "train_tokens_per_second": 4666.499
    },
    {
      "epoch": 0.8156887755102041,
      "grad_norm": 0.5711706385685005,
      "learning_rate": 6.428399168896244e-06,
      "loss": 0.32012438774108887,
      "num_input_tokens_seen": 383427432,
      "step": 6395,
      "train_runtime": 82169.0532,
      "train_tokens_per_second": 4666.324
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 0.5529100640187739,
      "learning_rate": 6.423598282108541e-06,
      "loss": 0.3004781246185303,
      "num_input_tokens_seen": 383734304,
      "step": 6400,
      "train_runtime": 82230.7124,
      "train_tokens_per_second": 4666.557
    },
    {
      "epoch": 0.8169642857142857,
      "grad_norm": 0.5651625948307254,
      "learning_rate": 6.41879596664054e-06,
      "loss": 0.31356689929962156,
      "num_input_tokens_seen": 384050208,
      "step": 6405,
      "train_runtime": 82297.3081,
      "train_tokens_per_second": 4666.619
    },
    {
      "epoch": 0.8176020408163265,
      "grad_norm": 0.6052901987965118,
      "learning_rate": 6.413992227311695e-06,
      "loss": 0.3242011070251465,
      "num_input_tokens_seen": 384352520,
      "step": 6410,
      "train_runtime": 82354.7932,
      "train_tokens_per_second": 4667.033
    },
    {
      "epoch": 0.8182397959183674,
      "grad_norm": 0.5824791938820854,
      "learning_rate": 6.409187068942899e-06,
      "loss": 0.3379192352294922,
      "num_input_tokens_seen": 384661472,
      "step": 6415,
      "train_runtime": 82410.6773,
      "train_tokens_per_second": 4667.617
    },
    {
      "epoch": 0.8188775510204082,
      "grad_norm": 0.597160115833879,
      "learning_rate": 6.4043804963564616e-06,
      "loss": 0.30432326793670655,
      "num_input_tokens_seen": 384952496,
      "step": 6420,
      "train_runtime": 82477.6853,
      "train_tokens_per_second": 4667.353
    },
    {
      "epoch": 0.8195153061224489,
      "grad_norm": 0.5829624892103409,
      "learning_rate": 6.399572514376114e-06,
      "loss": 0.3168663263320923,
      "num_input_tokens_seen": 385260312,
      "step": 6425,
      "train_runtime": 82554.1325,
      "train_tokens_per_second": 4666.76
    },
    {
      "epoch": 0.8201530612244898,
      "grad_norm": 0.5881912162459694,
      "learning_rate": 6.394763127827001e-06,
      "loss": 0.31240570545196533,
      "num_input_tokens_seen": 385552880,
      "step": 6430,
      "train_runtime": 82620.6023,
      "train_tokens_per_second": 4666.546
    },
    {
      "epoch": 0.8207908163265306,
      "grad_norm": 0.6143150978869795,
      "learning_rate": 6.389952341535681e-06,
      "loss": 0.32782998085021975,
      "num_input_tokens_seen": 385839272,
      "step": 6435,
      "train_runtime": 82685.204,
      "train_tokens_per_second": 4666.364
    },
    {
      "epoch": 0.8214285714285714,
      "grad_norm": 0.604183169197261,
      "learning_rate": 6.385140160330113e-06,
      "loss": 0.3201569080352783,
      "num_input_tokens_seen": 386133216,
      "step": 6440,
      "train_runtime": 82753.7661,
      "train_tokens_per_second": 4666.05
    },
    {
      "epoch": 0.8220663265306123,
      "grad_norm": 0.641592597660603,
      "learning_rate": 6.380326589039653e-06,
      "loss": 0.32984390258789065,
      "num_input_tokens_seen": 386436536,
      "step": 6445,
      "train_runtime": 82820.0168,
      "train_tokens_per_second": 4665.98
    },
    {
      "epoch": 0.8227040816326531,
      "grad_norm": 0.4990669626030833,
      "learning_rate": 6.375511632495062e-06,
      "loss": 0.31126041412353517,
      "num_input_tokens_seen": 386728840,
      "step": 6450,
      "train_runtime": 82890.8014,
      "train_tokens_per_second": 4665.522
    },
    {
      "epoch": 0.8233418367346939,
      "grad_norm": 0.5958384582118913,
      "learning_rate": 6.370695295528482e-06,
      "loss": 0.3371361494064331,
      "num_input_tokens_seen": 387049320,
      "step": 6455,
      "train_runtime": 82949.2764,
      "train_tokens_per_second": 4666.096
    },
    {
      "epoch": 0.8239795918367347,
      "grad_norm": 0.582417258146294,
      "learning_rate": 6.365877582973442e-06,
      "loss": 0.3114039897918701,
      "num_input_tokens_seen": 387353424,
      "step": 6460,
      "train_runtime": 83022.9156,
      "train_tokens_per_second": 4665.621
    },
    {
      "epoch": 0.8246173469387755,
      "grad_norm": 0.6247624844944313,
      "learning_rate": 6.361058499664856e-06,
      "loss": 0.29075353145599364,
      "num_input_tokens_seen": 387662184,
      "step": 6465,
      "train_runtime": 83080.1501,
      "train_tokens_per_second": 4666.123
    },
    {
      "epoch": 0.8252551020408163,
      "grad_norm": 0.6715022866722622,
      "learning_rate": 6.3562380504390095e-06,
      "loss": 0.3124758005142212,
      "num_input_tokens_seen": 387962320,
      "step": 6470,
      "train_runtime": 83143.6752,
      "train_tokens_per_second": 4666.168
    },
    {
      "epoch": 0.8258928571428571,
      "grad_norm": 0.5886998087369898,
      "learning_rate": 6.35141624013356e-06,
      "loss": 0.3298856258392334,
      "num_input_tokens_seen": 388268512,
      "step": 6475,
      "train_runtime": 83207.6486,
      "train_tokens_per_second": 4666.26
    },
    {
      "epoch": 0.826530612244898,
      "grad_norm": 0.682117483213479,
      "learning_rate": 6.34659307358753e-06,
      "loss": 0.32467265129089357,
      "num_input_tokens_seen": 388584360,
      "step": 6480,
      "train_runtime": 83276.0966,
      "train_tokens_per_second": 4666.217
    },
    {
      "epoch": 0.8271683673469388,
      "grad_norm": 0.673165921740877,
      "learning_rate": 6.341768555641306e-06,
      "loss": 0.30635595321655273,
      "num_input_tokens_seen": 388895224,
      "step": 6485,
      "train_runtime": 83331.6552,
      "train_tokens_per_second": 4666.837
    },
    {
      "epoch": 0.8278061224489796,
      "grad_norm": 0.6245067172447053,
      "learning_rate": 6.3369426911366275e-06,
      "loss": 0.2976788759231567,
      "num_input_tokens_seen": 389181952,
      "step": 6490,
      "train_runtime": 83396.7508,
      "train_tokens_per_second": 4666.632
    },
    {
      "epoch": 0.8284438775510204,
      "grad_norm": 0.6731416193267825,
      "learning_rate": 6.332115484916586e-06,
      "loss": 0.32232754230499266,
      "num_input_tokens_seen": 389468936,
      "step": 6495,
      "train_runtime": 83449.3886,
      "train_tokens_per_second": 4667.127
    },
    {
      "epoch": 0.8290816326530612,
      "grad_norm": 0.5547321497121352,
      "learning_rate": 6.327286941825621e-06,
      "loss": 0.30284967422485354,
      "num_input_tokens_seen": 389772488,
      "step": 6500,
      "train_runtime": 83507.3897,
      "train_tokens_per_second": 4667.521
    },
    {
      "epoch": 0.829719387755102,
      "grad_norm": 0.5731879400528207,
      "learning_rate": 6.322457066709511e-06,
      "loss": 0.3258717775344849,
      "num_input_tokens_seen": 390076200,
      "step": 6505,
      "train_runtime": 83564.1325,
      "train_tokens_per_second": 4667.986
    },
    {
      "epoch": 0.8303571428571429,
      "grad_norm": 0.5851361866151272,
      "learning_rate": 6.317625864415374e-06,
      "loss": 0.34453649520874025,
      "num_input_tokens_seen": 390373984,
      "step": 6510,
      "train_runtime": 83627.1732,
      "train_tokens_per_second": 4668.028
    },
    {
      "epoch": 0.8309948979591837,
      "grad_norm": 0.5933232999638829,
      "learning_rate": 6.312793339791662e-06,
      "loss": 0.3070641040802002,
      "num_input_tokens_seen": 390670072,
      "step": 6515,
      "train_runtime": 83690.651,
      "train_tokens_per_second": 4668.025
    },
    {
      "epoch": 0.8316326530612245,
      "grad_norm": 0.5669500400298843,
      "learning_rate": 6.307959497688146e-06,
      "loss": 0.30180792808532714,
      "num_input_tokens_seen": 390961488,
      "step": 6520,
      "train_runtime": 83750.8039,
      "train_tokens_per_second": 4668.152
    },
    {
      "epoch": 0.8322704081632653,
      "grad_norm": 0.5348433107701372,
      "learning_rate": 6.303124342955928e-06,
      "loss": 0.28929288387298585,
      "num_input_tokens_seen": 391264560,
      "step": 6525,
      "train_runtime": 83806.0417,
      "train_tokens_per_second": 4668.692
    },
    {
      "epoch": 0.8329081632653061,
      "grad_norm": 0.5821412719588734,
      "learning_rate": 6.29828788044742e-06,
      "loss": 0.30586791038513184,
      "num_input_tokens_seen": 391565080,
      "step": 6530,
      "train_runtime": 83873.2438,
      "train_tokens_per_second": 4668.534
    },
    {
      "epoch": 0.8335459183673469,
      "grad_norm": 0.6481037498029766,
      "learning_rate": 6.293450115016353e-06,
      "loss": 0.32516870498657224,
      "num_input_tokens_seen": 391871952,
      "step": 6535,
      "train_runtime": 83938.3482,
      "train_tokens_per_second": 4668.569
    },
    {
      "epoch": 0.8341836734693877,
      "grad_norm": 0.5962720550715901,
      "learning_rate": 6.288611051517761e-06,
      "loss": 0.32063119411468505,
      "num_input_tokens_seen": 392158336,
      "step": 6540,
      "train_runtime": 84015.0472,
      "train_tokens_per_second": 4667.715
    },
    {
      "epoch": 0.8348214285714286,
      "grad_norm": 0.5814538878413128,
      "learning_rate": 6.283770694807983e-06,
      "loss": 0.31683061122894285,
      "num_input_tokens_seen": 392467928,
      "step": 6545,
      "train_runtime": 84084.7952,
      "train_tokens_per_second": 4667.526
    },
    {
      "epoch": 0.8354591836734694,
      "grad_norm": 0.609185462134041,
      "learning_rate": 6.2789290497446546e-06,
      "loss": 0.2990833044052124,
      "num_input_tokens_seen": 392754144,
      "step": 6550,
      "train_runtime": 84146.9144,
      "train_tokens_per_second": 4667.481
    },
    {
      "epoch": 0.8360969387755102,
      "grad_norm": 0.5970171483645796,
      "learning_rate": 6.274086121186704e-06,
      "loss": 0.30519733428955076,
      "num_input_tokens_seen": 393035544,
      "step": 6555,
      "train_runtime": 84208.4249,
      "train_tokens_per_second": 4667.414
    },
    {
      "epoch": 0.8367346938775511,
      "grad_norm": 0.6206139250303385,
      "learning_rate": 6.269241913994348e-06,
      "loss": 0.34519333839416505,
      "num_input_tokens_seen": 393349192,
      "step": 6560,
      "train_runtime": 84265.1631,
      "train_tokens_per_second": 4667.993
    },
    {
      "epoch": 0.8373724489795918,
      "grad_norm": 0.5771468950095635,
      "learning_rate": 6.264396433029089e-06,
      "loss": 0.31926770210266114,
      "num_input_tokens_seen": 393662968,
      "step": 6565,
      "train_runtime": 84333.6413,
      "train_tokens_per_second": 4667.923
    },
    {
      "epoch": 0.8380102040816326,
      "grad_norm": 0.5315154151047687,
      "learning_rate": 6.2595496831537025e-06,
      "loss": 0.31521120071411135,
      "num_input_tokens_seen": 393960976,
      "step": 6570,
      "train_runtime": 84386.6592,
      "train_tokens_per_second": 4668.522
    },
    {
      "epoch": 0.8386479591836735,
      "grad_norm": 0.5414787300923072,
      "learning_rate": 6.254701669232243e-06,
      "loss": 0.30462558269500734,
      "num_input_tokens_seen": 394258296,
      "step": 6575,
      "train_runtime": 84455.7533,
      "train_tokens_per_second": 4668.223
    },
    {
      "epoch": 0.8392857142857143,
      "grad_norm": 0.6603055840351818,
      "learning_rate": 6.24985239613003e-06,
      "loss": 0.3152480602264404,
      "num_input_tokens_seen": 394552432,
      "step": 6580,
      "train_runtime": 84514.3729,
      "train_tokens_per_second": 4668.465
    },
    {
      "epoch": 0.8399234693877551,
      "grad_norm": 0.584825007451768,
      "learning_rate": 6.24500186871365e-06,
      "loss": 0.30623745918273926,
      "num_input_tokens_seen": 394853408,
      "step": 6585,
      "train_runtime": 84581.1855,
      "train_tokens_per_second": 4668.336
    },
    {
      "epoch": 0.8405612244897959,
      "grad_norm": 0.4930003653474112,
      "learning_rate": 6.240150091850941e-06,
      "loss": 0.3308518886566162,
      "num_input_tokens_seen": 395160664,
      "step": 6590,
      "train_runtime": 84641.9631,
      "train_tokens_per_second": 4668.614
    },
    {
      "epoch": 0.8411989795918368,
      "grad_norm": 0.6160894940287035,
      "learning_rate": 6.235297070411004e-06,
      "loss": 0.33232975006103516,
      "num_input_tokens_seen": 395461608,
      "step": 6595,
      "train_runtime": 84700.4224,
      "train_tokens_per_second": 4668.945
    },
    {
      "epoch": 0.8418367346938775,
      "grad_norm": 0.5630953384824021,
      "learning_rate": 6.230442809264186e-06,
      "loss": 0.31195032596588135,
      "num_input_tokens_seen": 395759016,
      "step": 6600,
      "train_runtime": 84774.0173,
      "train_tokens_per_second": 4668.4
    },
    {
      "epoch": 0.8424744897959183,
      "grad_norm": 0.6195254986287914,
      "learning_rate": 6.225587313282071e-06,
      "loss": 0.32104930877685545,
      "num_input_tokens_seen": 396054736,
      "step": 6605,
      "train_runtime": 84835.3441,
      "train_tokens_per_second": 4668.511
    },
    {
      "epoch": 0.8431122448979592,
      "grad_norm": 0.5809396114170757,
      "learning_rate": 6.220730587337496e-06,
      "loss": 0.3140249490737915,
      "num_input_tokens_seen": 396361608,
      "step": 6610,
      "train_runtime": 84907.1987,
      "train_tokens_per_second": 4668.174
    },
    {
      "epoch": 0.84375,
      "grad_norm": 0.6362805032477586,
      "learning_rate": 6.21587263630452e-06,
      "loss": 0.2962120771408081,
      "num_input_tokens_seen": 396646088,
      "step": 6615,
      "train_runtime": 84980.0625,
      "train_tokens_per_second": 4667.519
    },
    {
      "epoch": 0.8443877551020408,
      "grad_norm": 0.5937557493652235,
      "learning_rate": 6.2110134650584355e-06,
      "loss": 0.3020913124084473,
      "num_input_tokens_seen": 396945720,
      "step": 6620,
      "train_runtime": 85053.2493,
      "train_tokens_per_second": 4667.026
    },
    {
      "epoch": 0.8450255102040817,
      "grad_norm": 0.6226118867048471,
      "learning_rate": 6.2061530784757625e-06,
      "loss": 0.3444786787033081,
      "num_input_tokens_seen": 397248272,
      "step": 6625,
      "train_runtime": 85118.0255,
      "train_tokens_per_second": 4667.029
    },
    {
      "epoch": 0.8456632653061225,
      "grad_norm": 0.5892118579679192,
      "learning_rate": 6.201291481434238e-06,
      "loss": 0.3050083637237549,
      "num_input_tokens_seen": 397539016,
      "step": 6630,
      "train_runtime": 85182.4486,
      "train_tokens_per_second": 4666.912
    },
    {
      "epoch": 0.8463010204081632,
      "grad_norm": 0.5508066480329329,
      "learning_rate": 6.196428678812816e-06,
      "loss": 0.2922708034515381,
      "num_input_tokens_seen": 397836288,
      "step": 6635,
      "train_runtime": 85254.9993,
      "train_tokens_per_second": 4666.428
    },
    {
      "epoch": 0.8469387755102041,
      "grad_norm": 0.613685181300794,
      "learning_rate": 6.191564675491653e-06,
      "loss": 0.30857760906219484,
      "num_input_tokens_seen": 398137728,
      "step": 6640,
      "train_runtime": 85324.112,
      "train_tokens_per_second": 4666.181
    },
    {
      "epoch": 0.8475765306122449,
      "grad_norm": 0.5585401274307471,
      "learning_rate": 6.1866994763521215e-06,
      "loss": 0.29730515480041503,
      "num_input_tokens_seen": 398424976,
      "step": 6645,
      "train_runtime": 85390.0874,
      "train_tokens_per_second": 4665.939
    },
    {
      "epoch": 0.8482142857142857,
      "grad_norm": 0.5759389942635973,
      "learning_rate": 6.1818330862767864e-06,
      "loss": 0.30812749862670896,
      "num_input_tokens_seen": 398730408,
      "step": 6650,
      "train_runtime": 85464.1197,
      "train_tokens_per_second": 4665.471
    },
    {
      "epoch": 0.8488520408163265,
      "grad_norm": 0.6197027624764138,
      "learning_rate": 6.176965510149409e-06,
      "loss": 0.3033617973327637,
      "num_input_tokens_seen": 399037552,
      "step": 6655,
      "train_runtime": 85530.1083,
      "train_tokens_per_second": 4665.463
    },
    {
      "epoch": 0.8494897959183674,
      "grad_norm": 0.622364215380817,
      "learning_rate": 6.172096752854943e-06,
      "loss": 0.3203045606613159,
      "num_input_tokens_seen": 399333192,
      "step": 6660,
      "train_runtime": 85584.7089,
      "train_tokens_per_second": 4665.941
    },
    {
      "epoch": 0.8501275510204082,
      "grad_norm": 0.6401552047276462,
      "learning_rate": 6.1672268192795285e-06,
      "loss": 0.29885144233703614,
      "num_input_tokens_seen": 399628280,
      "step": 6665,
      "train_runtime": 85646.3763,
      "train_tokens_per_second": 4666.027
    },
    {
      "epoch": 0.8507653061224489,
      "grad_norm": 0.7250927430486945,
      "learning_rate": 6.16235571431048e-06,
      "loss": 0.3294283390045166,
      "num_input_tokens_seen": 399924800,
      "step": 6670,
      "train_runtime": 85720.061,
      "train_tokens_per_second": 4665.475
    },
    {
      "epoch": 0.8514030612244898,
      "grad_norm": 0.6634142320835769,
      "learning_rate": 6.157483442836294e-06,
      "loss": 0.33495144844055175,
      "num_input_tokens_seen": 400216624,
      "step": 6675,
      "train_runtime": 85784.2314,
      "train_tokens_per_second": 4665.387
    },
    {
      "epoch": 0.8520408163265306,
      "grad_norm": 0.5997126035442919,
      "learning_rate": 6.1526100097466345e-06,
      "loss": 0.31827507019042967,
      "num_input_tokens_seen": 400523824,
      "step": 6680,
      "train_runtime": 85846.9255,
      "train_tokens_per_second": 4665.558
    },
    {
      "epoch": 0.8526785714285714,
      "grad_norm": 0.64754068179524,
      "learning_rate": 6.147735419932334e-06,
      "loss": 0.3034789562225342,
      "num_input_tokens_seen": 400827320,
      "step": 6685,
      "train_runtime": 85912.644,
      "train_tokens_per_second": 4665.522
    },
    {
      "epoch": 0.8533163265306123,
      "grad_norm": 0.799892030264851,
      "learning_rate": 6.142859678285378e-06,
      "loss": 0.33355953693389895,
      "num_input_tokens_seen": 401125816,
      "step": 6690,
      "train_runtime": 85982.7443,
      "train_tokens_per_second": 4665.19
    },
    {
      "epoch": 0.8539540816326531,
      "grad_norm": 0.6392905494428465,
      "learning_rate": 6.13798278969892e-06,
      "loss": 0.3342573642730713,
      "num_input_tokens_seen": 401410360,
      "step": 6695,
      "train_runtime": 86048.236,
      "train_tokens_per_second": 4664.946
    },
    {
      "epoch": 0.8545918367346939,
      "grad_norm": 0.5203894297352434,
      "learning_rate": 6.133104759067257e-06,
      "loss": 0.30130274295806886,
      "num_input_tokens_seen": 401717752,
      "step": 6700,
      "train_runtime": 86109.7386,
      "train_tokens_per_second": 4665.184
    },
    {
      "epoch": 0.8552295918367347,
      "grad_norm": 0.5605527505484186,
      "learning_rate": 6.1282255912858315e-06,
      "loss": 0.29971344470977784,
      "num_input_tokens_seen": 402013016,
      "step": 6705,
      "train_runtime": 86170.1548,
      "train_tokens_per_second": 4665.339
    },
    {
      "epoch": 0.8558673469387755,
      "grad_norm": 0.604448906313011,
      "learning_rate": 6.1233452912512295e-06,
      "loss": 0.3489644765853882,
      "num_input_tokens_seen": 402319832,
      "step": 6710,
      "train_runtime": 86235.6136,
      "train_tokens_per_second": 4665.356
    },
    {
      "epoch": 0.8565051020408163,
      "grad_norm": 0.5895474848157423,
      "learning_rate": 6.118463863861174e-06,
      "loss": 0.2796917915344238,
      "num_input_tokens_seen": 402625488,
      "step": 6715,
      "train_runtime": 86298.1269,
      "train_tokens_per_second": 4665.518
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.6249355630850112,
      "learning_rate": 6.113581314014517e-06,
      "loss": 0.31249542236328126,
      "num_input_tokens_seen": 402925992,
      "step": 6720,
      "train_runtime": 86355.461,
      "train_tokens_per_second": 4665.901
    },
    {
      "epoch": 0.857780612244898,
      "grad_norm": 0.6304542140852867,
      "learning_rate": 6.10869764661124e-06,
      "loss": 0.3219332456588745,
      "num_input_tokens_seen": 403207984,
      "step": 6725,
      "train_runtime": 86413.3914,
      "train_tokens_per_second": 4666.036
    },
    {
      "epoch": 0.8584183673469388,
      "grad_norm": 0.5918915350014844,
      "learning_rate": 6.103812866552444e-06,
      "loss": 0.294742488861084,
      "num_input_tokens_seen": 403494608,
      "step": 6730,
      "train_runtime": 86472.1025,
      "train_tokens_per_second": 4666.182
    },
    {
      "epoch": 0.8590561224489796,
      "grad_norm": 0.6322048157286655,
      "learning_rate": 6.0989269787403445e-06,
      "loss": 0.3376746416091919,
      "num_input_tokens_seen": 403784440,
      "step": 6735,
      "train_runtime": 86533.3987,
      "train_tokens_per_second": 4666.227
    },
    {
      "epoch": 0.8596938775510204,
      "grad_norm": 0.5858078669370196,
      "learning_rate": 6.094039988078273e-06,
      "loss": 0.31620571613311765,
      "num_input_tokens_seen": 404085600,
      "step": 6740,
      "train_runtime": 86598.9299,
      "train_tokens_per_second": 4666.173
    },
    {
      "epoch": 0.8603316326530612,
      "grad_norm": 0.5598138022206285,
      "learning_rate": 6.089151899470668e-06,
      "loss": 0.2858736038208008,
      "num_input_tokens_seen": 404382184,
      "step": 6745,
      "train_runtime": 86672.1992,
      "train_tokens_per_second": 4665.65
    },
    {
      "epoch": 0.860969387755102,
      "grad_norm": 0.6743694114584343,
      "learning_rate": 6.084262717823064e-06,
      "loss": 0.2900573253631592,
      "num_input_tokens_seen": 404686752,
      "step": 6750,
      "train_runtime": 86725.718,
      "train_tokens_per_second": 4666.283
    },
    {
      "epoch": 0.8616071428571429,
      "grad_norm": 0.5880382421521227,
      "learning_rate": 6.079372448042098e-06,
      "loss": 0.2982300043106079,
      "num_input_tokens_seen": 404985704,
      "step": 6755,
      "train_runtime": 86801.7811,
      "train_tokens_per_second": 4665.638
    },
    {
      "epoch": 0.8622448979591837,
      "grad_norm": 0.556309105213148,
      "learning_rate": 6.074481095035498e-06,
      "loss": 0.32536613941192627,
      "num_input_tokens_seen": 405288072,
      "step": 6760,
      "train_runtime": 86862.2901,
      "train_tokens_per_second": 4665.869
    },
    {
      "epoch": 0.8628826530612245,
      "grad_norm": 0.612299823377922,
      "learning_rate": 6.069588663712077e-06,
      "loss": 0.3190335273742676,
      "num_input_tokens_seen": 405576032,
      "step": 6765,
      "train_runtime": 86933.1464,
      "train_tokens_per_second": 4665.378
    },
    {
      "epoch": 0.8635204081632653,
      "grad_norm": 0.6356943946491983,
      "learning_rate": 6.064695158981732e-06,
      "loss": 0.3045177459716797,
      "num_input_tokens_seen": 405871224,
      "step": 6770,
      "train_runtime": 87006.0629,
      "train_tokens_per_second": 4664.861
    },
    {
      "epoch": 0.8641581632653061,
      "grad_norm": 0.596680131580734,
      "learning_rate": 6.059800585755436e-06,
      "loss": 0.333588695526123,
      "num_input_tokens_seen": 406175840,
      "step": 6775,
      "train_runtime": 87075.4011,
      "train_tokens_per_second": 4664.645
    },
    {
      "epoch": 0.8647959183673469,
      "grad_norm": 0.6638468096818336,
      "learning_rate": 6.054904948945238e-06,
      "loss": 0.2864166259765625,
      "num_input_tokens_seen": 406470040,
      "step": 6780,
      "train_runtime": 87135.9405,
      "train_tokens_per_second": 4664.781
    },
    {
      "epoch": 0.8654336734693877,
      "grad_norm": 0.6896758525257892,
      "learning_rate": 6.050008253464247e-06,
      "loss": 0.3302961587905884,
      "num_input_tokens_seen": 406767296,
      "step": 6785,
      "train_runtime": 87200.7229,
      "train_tokens_per_second": 4664.724
    },
    {
      "epoch": 0.8660714285714286,
      "grad_norm": 0.627576937677198,
      "learning_rate": 6.045110504226643e-06,
      "loss": 0.2932382583618164,
      "num_input_tokens_seen": 407074112,
      "step": 6790,
      "train_runtime": 87260.0977,
      "train_tokens_per_second": 4665.066
    },
    {
      "epoch": 0.8667091836734694,
      "grad_norm": 0.5683611368657269,
      "learning_rate": 6.0402117061476574e-06,
      "loss": 0.3078974485397339,
      "num_input_tokens_seen": 407361312,
      "step": 6795,
      "train_runtime": 87320.2148,
      "train_tokens_per_second": 4665.143
    },
    {
      "epoch": 0.8673469387755102,
      "grad_norm": 0.5748921175063846,
      "learning_rate": 6.035311864143576e-06,
      "loss": 0.3090816974639893,
      "num_input_tokens_seen": 407661936,
      "step": 6800,
      "train_runtime": 87384.4751,
      "train_tokens_per_second": 4665.153
    },
    {
      "epoch": 0.8679846938775511,
      "grad_norm": 0.5863373467183365,
      "learning_rate": 6.030410983131733e-06,
      "loss": 0.3100122928619385,
      "num_input_tokens_seen": 407956216,
      "step": 6805,
      "train_runtime": 87453.438,
      "train_tokens_per_second": 4664.839
    },
    {
      "epoch": 0.8686224489795918,
      "grad_norm": 0.5968343807116784,
      "learning_rate": 6.025509068030504e-06,
      "loss": 0.3232810735702515,
      "num_input_tokens_seen": 408256432,
      "step": 6810,
      "train_runtime": 87532.1653,
      "train_tokens_per_second": 4664.073
    },
    {
      "epoch": 0.8692602040816326,
      "grad_norm": 0.5521085973090777,
      "learning_rate": 6.020606123759305e-06,
      "loss": 0.3054074287414551,
      "num_input_tokens_seen": 408540944,
      "step": 6815,
      "train_runtime": 87598.7119,
      "train_tokens_per_second": 4663.778
    },
    {
      "epoch": 0.8698979591836735,
      "grad_norm": 0.6592982922710323,
      "learning_rate": 6.0157021552385805e-06,
      "loss": 0.35090322494506837,
      "num_input_tokens_seen": 408838824,
      "step": 6820,
      "train_runtime": 87658.1677,
      "train_tokens_per_second": 4664.013
    },
    {
      "epoch": 0.8705357142857143,
      "grad_norm": 0.6056940862517836,
      "learning_rate": 6.010797167389808e-06,
      "loss": 0.325609827041626,
      "num_input_tokens_seen": 409138552,
      "step": 6825,
      "train_runtime": 87726.4786,
      "train_tokens_per_second": 4663.798
    },
    {
      "epoch": 0.8711734693877551,
      "grad_norm": 0.5546337459085608,
      "learning_rate": 6.005891165135484e-06,
      "loss": 0.298302149772644,
      "num_input_tokens_seen": 409427272,
      "step": 6830,
      "train_runtime": 87792.2703,
      "train_tokens_per_second": 4663.591
    },
    {
      "epoch": 0.8718112244897959,
      "grad_norm": 0.5880820071208903,
      "learning_rate": 6.000984153399123e-06,
      "loss": 0.31217527389526367,
      "num_input_tokens_seen": 409724232,
      "step": 6835,
      "train_runtime": 87859.937,
      "train_tokens_per_second": 4663.38
    },
    {
      "epoch": 0.8724489795918368,
      "grad_norm": 0.701466624618536,
      "learning_rate": 5.996076137105258e-06,
      "loss": 0.3225777387619019,
      "num_input_tokens_seen": 410015192,
      "step": 6840,
      "train_runtime": 87919.7174,
      "train_tokens_per_second": 4663.518
    },
    {
      "epoch": 0.8730867346938775,
      "grad_norm": 0.6201285590127129,
      "learning_rate": 5.991167121179422e-06,
      "loss": 0.3138850212097168,
      "num_input_tokens_seen": 410303616,
      "step": 6845,
      "train_runtime": 87982.5899,
      "train_tokens_per_second": 4663.464
    },
    {
      "epoch": 0.8737244897959183,
      "grad_norm": 0.5214246251077226,
      "learning_rate": 5.986257110548157e-06,
      "loss": 0.31941967010498046,
      "num_input_tokens_seen": 410602016,
      "step": 6850,
      "train_runtime": 88040.7969,
      "train_tokens_per_second": 4663.77
    },
    {
      "epoch": 0.8743622448979592,
      "grad_norm": 0.6001896043025774,
      "learning_rate": 5.981346110139001e-06,
      "loss": 0.3214601516723633,
      "num_input_tokens_seen": 410906336,
      "step": 6855,
      "train_runtime": 88105.2609,
      "train_tokens_per_second": 4663.812
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.6660455026594599,
      "learning_rate": 5.976434124880487e-06,
      "loss": 0.3431683540344238,
      "num_input_tokens_seen": 411215096,
      "step": 6860,
      "train_runtime": 88171.4904,
      "train_tokens_per_second": 4663.81
    },
    {
      "epoch": 0.8756377551020408,
      "grad_norm": 0.6252393868053129,
      "learning_rate": 5.971521159702136e-06,
      "loss": 0.3205599784851074,
      "num_input_tokens_seen": 411510640,
      "step": 6865,
      "train_runtime": 88239.2355,
      "train_tokens_per_second": 4663.579
    },
    {
      "epoch": 0.8762755102040817,
      "grad_norm": 0.5547679986866066,
      "learning_rate": 5.966607219534448e-06,
      "loss": 0.32694807052612307,
      "num_input_tokens_seen": 411818544,
      "step": 6870,
      "train_runtime": 88302.8578,
      "train_tokens_per_second": 4663.706
    },
    {
      "epoch": 0.8769132653061225,
      "grad_norm": 0.632475699148163,
      "learning_rate": 5.9616923093089095e-06,
      "loss": 0.33402485847473146,
      "num_input_tokens_seen": 412117168,
      "step": 6875,
      "train_runtime": 88363.6338,
      "train_tokens_per_second": 4663.878
    },
    {
      "epoch": 0.8775510204081632,
      "grad_norm": 0.6107142526376932,
      "learning_rate": 5.956776433957974e-06,
      "loss": 0.30840229988098145,
      "num_input_tokens_seen": 412418192,
      "step": 6880,
      "train_runtime": 88419.5643,
      "train_tokens_per_second": 4664.332
    },
    {
      "epoch": 0.8781887755102041,
      "grad_norm": 0.9774654780676465,
      "learning_rate": 5.951859598415063e-06,
      "loss": 0.30924062728881835,
      "num_input_tokens_seen": 412735408,
      "step": 6885,
      "train_runtime": 88470.7151,
      "train_tokens_per_second": 4665.221
    },
    {
      "epoch": 0.8788265306122449,
      "grad_norm": 1.0116220455961549,
      "learning_rate": 5.94694180761457e-06,
      "loss": 0.30491957664489744,
      "num_input_tokens_seen": 413040720,
      "step": 6890,
      "train_runtime": 88543.2452,
      "train_tokens_per_second": 4664.847
    },
    {
      "epoch": 0.8794642857142857,
      "grad_norm": 0.6616404324530736,
      "learning_rate": 5.942023066491839e-06,
      "loss": 0.3186971187591553,
      "num_input_tokens_seen": 413349792,
      "step": 6895,
      "train_runtime": 88603.0006,
      "train_tokens_per_second": 4665.19
    },
    {
      "epoch": 0.8801020408163265,
      "grad_norm": 0.5868765548473264,
      "learning_rate": 5.937103379983169e-06,
      "loss": 0.3613459587097168,
      "num_input_tokens_seen": 413649416,
      "step": 6900,
      "train_runtime": 88669.8198,
      "train_tokens_per_second": 4665.053
    },
    {
      "epoch": 0.8807397959183674,
      "grad_norm": 0.6703907898729683,
      "learning_rate": 5.93218275302581e-06,
      "loss": 0.31890115737915037,
      "num_input_tokens_seen": 413946488,
      "step": 6905,
      "train_runtime": 88735.7122,
      "train_tokens_per_second": 4664.937
    },
    {
      "epoch": 0.8813775510204082,
      "grad_norm": 0.5597040222849586,
      "learning_rate": 5.927261190557955e-06,
      "loss": 0.29169273376464844,
      "num_input_tokens_seen": 414246552,
      "step": 6910,
      "train_runtime": 88800.3166,
      "train_tokens_per_second": 4664.922
    },
    {
      "epoch": 0.8820153061224489,
      "grad_norm": 0.541876198766452,
      "learning_rate": 5.922338697518736e-06,
      "loss": 0.31300668716430663,
      "num_input_tokens_seen": 414536744,
      "step": 6915,
      "train_runtime": 88870.94,
      "train_tokens_per_second": 4664.48
    },
    {
      "epoch": 0.8826530612244898,
      "grad_norm": 0.5846556553233282,
      "learning_rate": 5.917415278848217e-06,
      "loss": 0.28698785305023194,
      "num_input_tokens_seen": 414818808,
      "step": 6920,
      "train_runtime": 88935.0086,
      "train_tokens_per_second": 4664.292
    },
    {
      "epoch": 0.8832908163265306,
      "grad_norm": 0.650362363722194,
      "learning_rate": 5.9124909394873955e-06,
      "loss": 0.31494836807250975,
      "num_input_tokens_seen": 415118352,
      "step": 6925,
      "train_runtime": 88997.3932,
      "train_tokens_per_second": 4664.388
    },
    {
      "epoch": 0.8839285714285714,
      "grad_norm": 0.6267998906671569,
      "learning_rate": 5.907565684378187e-06,
      "loss": 0.3305811882019043,
      "num_input_tokens_seen": 415427312,
      "step": 6930,
      "train_runtime": 89062.0108,
      "train_tokens_per_second": 4664.473
    },
    {
      "epoch": 0.8845663265306123,
      "grad_norm": 0.563557148742557,
      "learning_rate": 5.902639518463429e-06,
      "loss": 0.32275214195251467,
      "num_input_tokens_seen": 415736648,
      "step": 6935,
      "train_runtime": 89134.0945,
      "train_tokens_per_second": 4664.171
    },
    {
      "epoch": 0.8852040816326531,
      "grad_norm": 0.7094212717207782,
      "learning_rate": 5.897712446686875e-06,
      "loss": 0.3357053756713867,
      "num_input_tokens_seen": 416036176,
      "step": 6940,
      "train_runtime": 89190.4817,
      "train_tokens_per_second": 4664.58
    },
    {
      "epoch": 0.8858418367346939,
      "grad_norm": 0.6162545250950107,
      "learning_rate": 5.892784473993184e-06,
      "loss": 0.3229557275772095,
      "num_input_tokens_seen": 416352976,
      "step": 6945,
      "train_runtime": 89253.7085,
      "train_tokens_per_second": 4664.826
    },
    {
      "epoch": 0.8864795918367347,
      "grad_norm": 0.6215196508406043,
      "learning_rate": 5.887855605327921e-06,
      "loss": 0.3064050436019897,
      "num_input_tokens_seen": 416634768,
      "step": 6950,
      "train_runtime": 89323.6516,
      "train_tokens_per_second": 4664.328
    },
    {
      "epoch": 0.8871173469387755,
      "grad_norm": 0.5379298754808535,
      "learning_rate": 5.882925845637551e-06,
      "loss": 0.28948819637298584,
      "num_input_tokens_seen": 416930736,
      "step": 6955,
      "train_runtime": 89393.0204,
      "train_tokens_per_second": 4664.019
    },
    {
      "epoch": 0.8877551020408163,
      "grad_norm": 0.5644010264450229,
      "learning_rate": 5.877995199869433e-06,
      "loss": 0.3060819387435913,
      "num_input_tokens_seen": 417231360,
      "step": 6960,
      "train_runtime": 89448.7321,
      "train_tokens_per_second": 4664.475
    },
    {
      "epoch": 0.8883928571428571,
      "grad_norm": 0.6798951714405044,
      "learning_rate": 5.87306367297181e-06,
      "loss": 0.30504908561706545,
      "num_input_tokens_seen": 417533072,
      "step": 6965,
      "train_runtime": 89508.645,
      "train_tokens_per_second": 4664.723
    },
    {
      "epoch": 0.889030612244898,
      "grad_norm": 0.5536973330920619,
      "learning_rate": 5.868131269893818e-06,
      "loss": 0.3097344398498535,
      "num_input_tokens_seen": 417836296,
      "step": 6970,
      "train_runtime": 89576.4258,
      "train_tokens_per_second": 4664.579
    },
    {
      "epoch": 0.8896683673469388,
      "grad_norm": 0.6021738763175897,
      "learning_rate": 5.863197995585468e-06,
      "loss": 0.3265312910079956,
      "num_input_tokens_seen": 418148064,
      "step": 6975,
      "train_runtime": 89638.8695,
      "train_tokens_per_second": 4664.807
    },
    {
      "epoch": 0.8903061224489796,
      "grad_norm": 0.618131289530354,
      "learning_rate": 5.858263854997642e-06,
      "loss": 0.2940220832824707,
      "num_input_tokens_seen": 418428672,
      "step": 6980,
      "train_runtime": 89704.8581,
      "train_tokens_per_second": 4664.504
    },
    {
      "epoch": 0.8909438775510204,
      "grad_norm": 0.5704197760787771,
      "learning_rate": 5.853328853082098e-06,
      "loss": 0.2685279846191406,
      "num_input_tokens_seen": 418726304,
      "step": 6985,
      "train_runtime": 89773.3817,
      "train_tokens_per_second": 4664.259
    },
    {
      "epoch": 0.8915816326530612,
      "grad_norm": 0.5365262731253033,
      "learning_rate": 5.848392994791452e-06,
      "loss": 0.3194777488708496,
      "num_input_tokens_seen": 419029440,
      "step": 6990,
      "train_runtime": 89846.132,
      "train_tokens_per_second": 4663.856
    },
    {
      "epoch": 0.892219387755102,
      "grad_norm": 0.5611762037072114,
      "learning_rate": 5.843456285079185e-06,
      "loss": 0.3002311229705811,
      "num_input_tokens_seen": 419339984,
      "step": 6995,
      "train_runtime": 89910.4977,
      "train_tokens_per_second": 4663.971
    },
    {
      "epoch": 0.8928571428571429,
      "grad_norm": 0.5660753227530252,
      "learning_rate": 5.838518728899627e-06,
      "loss": 0.32773146629333494,
      "num_input_tokens_seen": 419652256,
      "step": 7000,
      "train_runtime": 89975.8773,
      "train_tokens_per_second": 4664.053
    },
    {
      "epoch": 0.8934948979591837,
      "grad_norm": 0.5787960280049785,
      "learning_rate": 5.833580331207964e-06,
      "loss": 0.3284013748168945,
      "num_input_tokens_seen": 419949944,
      "step": 7005,
      "train_runtime": 90041.6254,
      "train_tokens_per_second": 4663.953
    },
    {
      "epoch": 0.8941326530612245,
      "grad_norm": 0.7800960751316296,
      "learning_rate": 5.828641096960223e-06,
      "loss": 0.3397322654724121,
      "num_input_tokens_seen": 420245304,
      "step": 7010,
      "train_runtime": 90104.637,
      "train_tokens_per_second": 4663.97
    },
    {
      "epoch": 0.8947704081632653,
      "grad_norm": 0.5767757632144042,
      "learning_rate": 5.823701031113267e-06,
      "loss": 0.314349889755249,
      "num_input_tokens_seen": 420564848,
      "step": 7015,
      "train_runtime": 90164.0479,
      "train_tokens_per_second": 4664.441
    },
    {
      "epoch": 0.8954081632653061,
      "grad_norm": 0.7347328141444277,
      "learning_rate": 5.8187601386248004e-06,
      "loss": 0.3227765798568726,
      "num_input_tokens_seen": 420840552,
      "step": 7020,
      "train_runtime": 90229.1894,
      "train_tokens_per_second": 4664.129
    },
    {
      "epoch": 0.8960459183673469,
      "grad_norm": 0.6435925456275953,
      "learning_rate": 5.8138184244533516e-06,
      "loss": 0.3211270570755005,
      "num_input_tokens_seen": 421141232,
      "step": 7025,
      "train_runtime": 90286.2359,
      "train_tokens_per_second": 4664.512
    },
    {
      "epoch": 0.8966836734693877,
      "grad_norm": 0.5845604759406505,
      "learning_rate": 5.808875893558277e-06,
      "loss": 0.3297579288482666,
      "num_input_tokens_seen": 421447712,
      "step": 7030,
      "train_runtime": 90363.4074,
      "train_tokens_per_second": 4663.92
    },
    {
      "epoch": 0.8973214285714286,
      "grad_norm": 0.6969413341437088,
      "learning_rate": 5.8039325508997524e-06,
      "loss": 0.3073333740234375,
      "num_input_tokens_seen": 421743216,
      "step": 7035,
      "train_runtime": 90437.1693,
      "train_tokens_per_second": 4663.384
    },
    {
      "epoch": 0.8979591836734694,
      "grad_norm": 0.6433918640370359,
      "learning_rate": 5.798988401438765e-06,
      "loss": 0.3218099117279053,
      "num_input_tokens_seen": 422058392,
      "step": 7040,
      "train_runtime": 90491.6651,
      "train_tokens_per_second": 4664.058
    },
    {
      "epoch": 0.8985969387755102,
      "grad_norm": 0.5405343373056593,
      "learning_rate": 5.7940434501371175e-06,
      "loss": 0.3057726860046387,
      "num_input_tokens_seen": 422363208,
      "step": 7045,
      "train_runtime": 90550.6574,
      "train_tokens_per_second": 4664.386
    },
    {
      "epoch": 0.8992346938775511,
      "grad_norm": 0.5352537262482511,
      "learning_rate": 5.78909770195741e-06,
      "loss": 0.32413716316223146,
      "num_input_tokens_seen": 422670128,
      "step": 7050,
      "train_runtime": 90623.295,
      "train_tokens_per_second": 4664.034
    },
    {
      "epoch": 0.8998724489795918,
      "grad_norm": 0.506117529730911,
      "learning_rate": 5.7841511618630484e-06,
      "loss": 0.27179660797119143,
      "num_input_tokens_seen": 422979688,
      "step": 7055,
      "train_runtime": 90686.3358,
      "train_tokens_per_second": 4664.205
    },
    {
      "epoch": 0.9005102040816326,
      "grad_norm": 0.5578594055386115,
      "learning_rate": 5.779203834818232e-06,
      "loss": 0.30615389347076416,
      "num_input_tokens_seen": 423259432,
      "step": 7060,
      "train_runtime": 90748.6292,
      "train_tokens_per_second": 4664.086
    },
    {
      "epoch": 0.9011479591836735,
      "grad_norm": 0.5554250561137382,
      "learning_rate": 5.774255725787946e-06,
      "loss": 0.3139845371246338,
      "num_input_tokens_seen": 423564312,
      "step": 7065,
      "train_runtime": 90816.878,
      "train_tokens_per_second": 4663.938
    },
    {
      "epoch": 0.9017857142857143,
      "grad_norm": 0.6190073776215691,
      "learning_rate": 5.769306839737968e-06,
      "loss": 0.33006107807159424,
      "num_input_tokens_seen": 423874328,
      "step": 7070,
      "train_runtime": 90880.9723,
      "train_tokens_per_second": 4664.06
    },
    {
      "epoch": 0.9024234693877551,
      "grad_norm": 0.5699923662864511,
      "learning_rate": 5.764357181634847e-06,
      "loss": 0.31683037281036375,
      "num_input_tokens_seen": 424187552,
      "step": 7075,
      "train_runtime": 90946.2654,
      "train_tokens_per_second": 4664.156
    },
    {
      "epoch": 0.9030612244897959,
      "grad_norm": 0.6382617981880363,
      "learning_rate": 5.759406756445911e-06,
      "loss": 0.2922383785247803,
      "num_input_tokens_seen": 424485240,
      "step": 7080,
      "train_runtime": 91006.2202,
      "train_tokens_per_second": 4664.354
    },
    {
      "epoch": 0.9036989795918368,
      "grad_norm": 0.5703120241122546,
      "learning_rate": 5.754455569139258e-06,
      "loss": 0.28797054290771484,
      "num_input_tokens_seen": 424786256,
      "step": 7085,
      "train_runtime": 91068.2514,
      "train_tokens_per_second": 4664.482
    },
    {
      "epoch": 0.9043367346938775,
      "grad_norm": 0.5873448565170643,
      "learning_rate": 5.74950362468375e-06,
      "loss": 0.31067476272583006,
      "num_input_tokens_seen": 425076784,
      "step": 7090,
      "train_runtime": 91123.2751,
      "train_tokens_per_second": 4664.854
    },
    {
      "epoch": 0.9049744897959183,
      "grad_norm": 0.5987005729754719,
      "learning_rate": 5.74455092804901e-06,
      "loss": 0.3357563018798828,
      "num_input_tokens_seen": 425390160,
      "step": 7095,
      "train_runtime": 91184.8039,
      "train_tokens_per_second": 4665.143
    },
    {
      "epoch": 0.9056122448979592,
      "grad_norm": 0.5980979454653461,
      "learning_rate": 5.739597484205415e-06,
      "loss": 0.3039411544799805,
      "num_input_tokens_seen": 425694392,
      "step": 7100,
      "train_runtime": 91240.766,
      "train_tokens_per_second": 4665.616
    },
    {
      "epoch": 0.90625,
      "grad_norm": 0.5747493898487499,
      "learning_rate": 5.734643298124091e-06,
      "loss": 0.328928279876709,
      "num_input_tokens_seen": 425988048,
      "step": 7105,
      "train_runtime": 91313.1806,
      "train_tokens_per_second": 4665.132
    },
    {
      "epoch": 0.9068877551020408,
      "grad_norm": 0.5756988861054836,
      "learning_rate": 5.7296883747769096e-06,
      "loss": 0.3134473323822021,
      "num_input_tokens_seen": 426290152,
      "step": 7110,
      "train_runtime": 91377.9608,
      "train_tokens_per_second": 4665.131
    },
    {
      "epoch": 0.9075255102040817,
      "grad_norm": 0.6001531054678257,
      "learning_rate": 5.724732719136481e-06,
      "loss": 0.30744113922119143,
      "num_input_tokens_seen": 426598472,
      "step": 7115,
      "train_runtime": 91438.2323,
      "train_tokens_per_second": 4665.428
    },
    {
      "epoch": 0.9081632653061225,
      "grad_norm": 0.6068903422869749,
      "learning_rate": 5.719776336176156e-06,
      "loss": 0.3021916151046753,
      "num_input_tokens_seen": 426889416,
      "step": 7120,
      "train_runtime": 91503.7633,
      "train_tokens_per_second": 4665.266
    },
    {
      "epoch": 0.9088010204081632,
      "grad_norm": 0.5827151496950657,
      "learning_rate": 5.714819230870008e-06,
      "loss": 0.3168118715286255,
      "num_input_tokens_seen": 427194840,
      "step": 7125,
      "train_runtime": 91578.983,
      "train_tokens_per_second": 4664.769
    },
    {
      "epoch": 0.9094387755102041,
      "grad_norm": 0.5584850520855441,
      "learning_rate": 5.709861408192838e-06,
      "loss": 0.3096959590911865,
      "num_input_tokens_seen": 427492600,
      "step": 7130,
      "train_runtime": 91641.1969,
      "train_tokens_per_second": 4664.852
    },
    {
      "epoch": 0.9100765306122449,
      "grad_norm": 0.6197603815914464,
      "learning_rate": 5.704902873120169e-06,
      "loss": 0.29640045166015627,
      "num_input_tokens_seen": 427795448,
      "step": 7135,
      "train_runtime": 91692.3431,
      "train_tokens_per_second": 4665.553
    },
    {
      "epoch": 0.9107142857142857,
      "grad_norm": 0.6434202792469139,
      "learning_rate": 5.699943630628237e-06,
      "loss": 0.33102595806121826,
      "num_input_tokens_seen": 428104504,
      "step": 7140,
      "train_runtime": 91751.1201,
      "train_tokens_per_second": 4665.932
    },
    {
      "epoch": 0.9113520408163265,
      "grad_norm": 0.6292040750369453,
      "learning_rate": 5.694983685693987e-06,
      "loss": 0.3137689113616943,
      "num_input_tokens_seen": 428404848,
      "step": 7145,
      "train_runtime": 91799.9214,
      "train_tokens_per_second": 4666.723
    },
    {
      "epoch": 0.9119897959183674,
      "grad_norm": 0.6662152030153948,
      "learning_rate": 5.690023043295071e-06,
      "loss": 0.3325874090194702,
      "num_input_tokens_seen": 428703112,
      "step": 7150,
      "train_runtime": 91871.4612,
      "train_tokens_per_second": 4666.336
    },
    {
      "epoch": 0.9126275510204082,
      "grad_norm": 0.6540754605598266,
      "learning_rate": 5.6850617084098416e-06,
      "loss": 0.3043696641921997,
      "num_input_tokens_seen": 429000456,
      "step": 7155,
      "train_runtime": 91925.1303,
      "train_tokens_per_second": 4666.846
    },
    {
      "epoch": 0.9132653061224489,
      "grad_norm": 0.6632619898495971,
      "learning_rate": 5.6800996860173404e-06,
      "loss": 0.3231004476547241,
      "num_input_tokens_seen": 429297088,
      "step": 7160,
      "train_runtime": 91982.0064,
      "train_tokens_per_second": 4667.186
    },
    {
      "epoch": 0.9139030612244898,
      "grad_norm": 0.5703567693286303,
      "learning_rate": 5.675136981097309e-06,
      "loss": 0.3142138719558716,
      "num_input_tokens_seen": 429579968,
      "step": 7165,
      "train_runtime": 92041.4932,
      "train_tokens_per_second": 4667.242
    },
    {
      "epoch": 0.9145408163265306,
      "grad_norm": 0.5687135244709596,
      "learning_rate": 5.670173598630163e-06,
      "loss": 0.3060192108154297,
      "num_input_tokens_seen": 429874288,
      "step": 7170,
      "train_runtime": 92110.0601,
      "train_tokens_per_second": 4666.963
    },
    {
      "epoch": 0.9151785714285714,
      "grad_norm": 0.589292917190867,
      "learning_rate": 5.6652095435970074e-06,
      "loss": 0.30223712921142576,
      "num_input_tokens_seen": 430160784,
      "step": 7175,
      "train_runtime": 92179.9907,
      "train_tokens_per_second": 4666.531
    },
    {
      "epoch": 0.9158163265306123,
      "grad_norm": 0.5288173131215935,
      "learning_rate": 5.6602448209796165e-06,
      "loss": 0.30651304721832273,
      "num_input_tokens_seen": 430467256,
      "step": 7180,
      "train_runtime": 92247.5774,
      "train_tokens_per_second": 4666.434
    },
    {
      "epoch": 0.9164540816326531,
      "grad_norm": 0.6095056093311735,
      "learning_rate": 5.655279435760436e-06,
      "loss": 0.32366907596588135,
      "num_input_tokens_seen": 430770832,
      "step": 7185,
      "train_runtime": 92308.2816,
      "train_tokens_per_second": 4666.654
    },
    {
      "epoch": 0.9170918367346939,
      "grad_norm": 0.5543637119691666,
      "learning_rate": 5.650313392922578e-06,
      "loss": 0.3008253812789917,
      "num_input_tokens_seen": 431072344,
      "step": 7190,
      "train_runtime": 92367.9908,
      "train_tokens_per_second": 4666.902
    },
    {
      "epoch": 0.9177295918367347,
      "grad_norm": 0.5866655140224115,
      "learning_rate": 5.64534669744981e-06,
      "loss": 0.31891732215881347,
      "num_input_tokens_seen": 431381272,
      "step": 7195,
      "train_runtime": 92440.1997,
      "train_tokens_per_second": 4666.598
    },
    {
      "epoch": 0.9183673469387755,
      "grad_norm": 0.6073417287934477,
      "learning_rate": 5.640379354326563e-06,
      "loss": 0.32229931354522706,
      "num_input_tokens_seen": 431680488,
      "step": 7200,
      "train_runtime": 92505.6784,
      "train_tokens_per_second": 4666.53
    },
    {
      "epoch": 0.9190051020408163,
      "grad_norm": 0.661024416370543,
      "learning_rate": 5.63541136853791e-06,
      "loss": 0.2939903736114502,
      "num_input_tokens_seen": 431969632,
      "step": 7205,
      "train_runtime": 92574.933,
      "train_tokens_per_second": 4666.162
    },
    {
      "epoch": 0.9196428571428571,
      "grad_norm": 0.6845318201418956,
      "learning_rate": 5.630442745069568e-06,
      "loss": 0.3262001037597656,
      "num_input_tokens_seen": 432267888,
      "step": 7210,
      "train_runtime": 92638.8251,
      "train_tokens_per_second": 4666.163
    },
    {
      "epoch": 0.920280612244898,
      "grad_norm": 0.5940156412457124,
      "learning_rate": 5.625473488907905e-06,
      "loss": 0.3024473190307617,
      "num_input_tokens_seen": 432570896,
      "step": 7215,
      "train_runtime": 92710.2873,
      "train_tokens_per_second": 4665.835
    },
    {
      "epoch": 0.9209183673469388,
      "grad_norm": 0.5800769800283246,
      "learning_rate": 5.620503605039912e-06,
      "loss": 0.3193344593048096,
      "num_input_tokens_seen": 432878976,
      "step": 7220,
      "train_runtime": 92777.5906,
      "train_tokens_per_second": 4665.771
    },
    {
      "epoch": 0.9215561224489796,
      "grad_norm": 0.5572565007548289,
      "learning_rate": 5.615533098453215e-06,
      "loss": 0.2801225185394287,
      "num_input_tokens_seen": 433186248,
      "step": 7225,
      "train_runtime": 92838.4875,
      "train_tokens_per_second": 4666.02
    },
    {
      "epoch": 0.9221938775510204,
      "grad_norm": 0.5186992946677108,
      "learning_rate": 5.610561974136065e-06,
      "loss": 0.29192466735839845,
      "num_input_tokens_seen": 433486960,
      "step": 7230,
      "train_runtime": 92886.9217,
      "train_tokens_per_second": 4666.824
    },
    {
      "epoch": 0.9228316326530612,
      "grad_norm": 0.7562562116699331,
      "learning_rate": 5.605590237077332e-06,
      "loss": 0.3300828218460083,
      "num_input_tokens_seen": 433780840,
      "step": 7235,
      "train_runtime": 92956.316,
      "train_tokens_per_second": 4666.502
    },
    {
      "epoch": 0.923469387755102,
      "grad_norm": 0.5801308207234838,
      "learning_rate": 5.600617892266503e-06,
      "loss": 0.29742567539215087,
      "num_input_tokens_seen": 434096064,
      "step": 7240,
      "train_runtime": 93028.6449,
      "train_tokens_per_second": 4666.262
    },
    {
      "epoch": 0.9241071428571429,
      "grad_norm": 0.6394480180012891,
      "learning_rate": 5.5956449446936715e-06,
      "loss": 0.30805795192718505,
      "num_input_tokens_seen": 434398496,
      "step": 7245,
      "train_runtime": 93103.9915,
      "train_tokens_per_second": 4665.734
    },
    {
      "epoch": 0.9247448979591837,
      "grad_norm": 0.7338940794726408,
      "learning_rate": 5.590671399349539e-06,
      "loss": 0.3318042516708374,
      "num_input_tokens_seen": 434695112,
      "step": 7250,
      "train_runtime": 93170.4705,
      "train_tokens_per_second": 4665.589
    },
    {
      "epoch": 0.9253826530612245,
      "grad_norm": 0.6256606219661659,
      "learning_rate": 5.585697261225404e-06,
      "loss": 0.31543722152709963,
      "num_input_tokens_seen": 434997880,
      "step": 7255,
      "train_runtime": 93234.4865,
      "train_tokens_per_second": 4665.633
    },
    {
      "epoch": 0.9260204081632653,
      "grad_norm": 0.6650623900070087,
      "learning_rate": 5.58072253531316e-06,
      "loss": 0.32381477355957033,
      "num_input_tokens_seen": 435307224,
      "step": 7260,
      "train_runtime": 93296.8306,
      "train_tokens_per_second": 4665.831
    },
    {
      "epoch": 0.9266581632653061,
      "grad_norm": 0.6038362676540124,
      "learning_rate": 5.575747226605298e-06,
      "loss": 0.31850357055664064,
      "num_input_tokens_seen": 435620200,
      "step": 7265,
      "train_runtime": 93358.8387,
      "train_tokens_per_second": 4666.084
    },
    {
      "epoch": 0.9272959183673469,
      "grad_norm": 0.5095918793011849,
      "learning_rate": 5.570771340094883e-06,
      "loss": 0.28309895992279055,
      "num_input_tokens_seen": 435941080,
      "step": 7270,
      "train_runtime": 93415.0525,
      "train_tokens_per_second": 4666.711
    },
    {
      "epoch": 0.9279336734693877,
      "grad_norm": 0.6440571472363364,
      "learning_rate": 5.565794880775565e-06,
      "loss": 0.30919201374053956,
      "num_input_tokens_seen": 436238216,
      "step": 7275,
      "train_runtime": 93482.4774,
      "train_tokens_per_second": 4666.524
    },
    {
      "epoch": 0.9285714285714286,
      "grad_norm": 0.6201475695430417,
      "learning_rate": 5.560817853641568e-06,
      "loss": 0.2955244779586792,
      "num_input_tokens_seen": 436542688,
      "step": 7280,
      "train_runtime": 93551.8807,
      "train_tokens_per_second": 4666.317
    },
    {
      "epoch": 0.9292091836734694,
      "grad_norm": 0.5196330467583057,
      "learning_rate": 5.5558402636876885e-06,
      "loss": 0.2989202976226807,
      "num_input_tokens_seen": 436838992,
      "step": 7285,
      "train_runtime": 93607.9933,
      "train_tokens_per_second": 4666.685
    },
    {
      "epoch": 0.9298469387755102,
      "grad_norm": 0.5986134892779362,
      "learning_rate": 5.550862115909287e-06,
      "loss": 0.34584856033325195,
      "num_input_tokens_seen": 437148752,
      "step": 7290,
      "train_runtime": 93664.1461,
      "train_tokens_per_second": 4667.194
    },
    {
      "epoch": 0.9304846938775511,
      "grad_norm": 0.6363052132408222,
      "learning_rate": 5.545883415302276e-06,
      "loss": 0.3146224498748779,
      "num_input_tokens_seen": 437450384,
      "step": 7295,
      "train_runtime": 93722.1008,
      "train_tokens_per_second": 4667.526
    },
    {
      "epoch": 0.9311224489795918,
      "grad_norm": 0.6579653171206922,
      "learning_rate": 5.540904166863136e-06,
      "loss": 0.29396662712097166,
      "num_input_tokens_seen": 437743192,
      "step": 7300,
      "train_runtime": 93791.3072,
      "train_tokens_per_second": 4667.204
    },
    {
      "epoch": 0.9317602040816326,
      "grad_norm": 0.5407312888731015,
      "learning_rate": 5.535924375588887e-06,
      "loss": 0.276171875,
      "num_input_tokens_seen": 438038192,
      "step": 7305,
      "train_runtime": 93843.6526,
      "train_tokens_per_second": 4667.744
    },
    {
      "epoch": 0.9323979591836735,
      "grad_norm": 0.6408789940643321,
      "learning_rate": 5.530944046477098e-06,
      "loss": 0.3132428407669067,
      "num_input_tokens_seen": 438329456,
      "step": 7310,
      "train_runtime": 93898.2718,
      "train_tokens_per_second": 4668.131
    },
    {
      "epoch": 0.9330357142857143,
      "grad_norm": 0.636867795471314,
      "learning_rate": 5.525963184525877e-06,
      "loss": 0.3216026067733765,
      "num_input_tokens_seen": 438628168,
      "step": 7315,
      "train_runtime": 93963.7142,
      "train_tokens_per_second": 4668.059
    },
    {
      "epoch": 0.9336734693877551,
      "grad_norm": 0.5997438277249362,
      "learning_rate": 5.5209817947338685e-06,
      "loss": 0.32601065635681153,
      "num_input_tokens_seen": 438913408,
      "step": 7320,
      "train_runtime": 94031.2371,
      "train_tokens_per_second": 4667.74
    },
    {
      "epoch": 0.9343112244897959,
      "grad_norm": 0.6171801976002438,
      "learning_rate": 5.5159998821002425e-06,
      "loss": 0.3318033218383789,
      "num_input_tokens_seen": 439212088,
      "step": 7325,
      "train_runtime": 94093.8948,
      "train_tokens_per_second": 4667.806
    },
    {
      "epoch": 0.9349489795918368,
      "grad_norm": 0.5738435965055434,
      "learning_rate": 5.511017451624698e-06,
      "loss": 0.2951535701751709,
      "num_input_tokens_seen": 439503280,
      "step": 7330,
      "train_runtime": 94164.4511,
      "train_tokens_per_second": 4667.401
    },
    {
      "epoch": 0.9355867346938775,
      "grad_norm": 0.6756323345305557,
      "learning_rate": 5.506034508307453e-06,
      "loss": 0.29136927127838136,
      "num_input_tokens_seen": 439792240,
      "step": 7335,
      "train_runtime": 94229.1144,
      "train_tokens_per_second": 4667.265
    },
    {
      "epoch": 0.9362244897959183,
      "grad_norm": 0.555093105086905,
      "learning_rate": 5.501051057149234e-06,
      "loss": 0.3056096315383911,
      "num_input_tokens_seen": 440094736,
      "step": 7340,
      "train_runtime": 94295.2096,
      "train_tokens_per_second": 4667.201
    },
    {
      "epoch": 0.9368622448979592,
      "grad_norm": 0.5501806298116312,
      "learning_rate": 5.496067103151288e-06,
      "loss": 0.32032227516174316,
      "num_input_tokens_seen": 440391912,
      "step": 7345,
      "train_runtime": 94357.626,
      "train_tokens_per_second": 4667.264
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.5850828313124876,
      "learning_rate": 5.491082651315359e-06,
      "loss": 0.3087699174880981,
      "num_input_tokens_seen": 440686592,
      "step": 7350,
      "train_runtime": 94431.6859,
      "train_tokens_per_second": 4666.724
    },
    {
      "epoch": 0.9381377551020408,
      "grad_norm": 0.5713305456694097,
      "learning_rate": 5.4860977066436915e-06,
      "loss": 0.33030338287353517,
      "num_input_tokens_seen": 440978816,
      "step": 7355,
      "train_runtime": 94493.1498,
      "train_tokens_per_second": 4666.781
    },
    {
      "epoch": 0.9387755102040817,
      "grad_norm": 0.6320981171774707,
      "learning_rate": 5.481112274139026e-06,
      "loss": 0.3123514652252197,
      "num_input_tokens_seen": 441283240,
      "step": 7360,
      "train_runtime": 94549.0858,
      "train_tokens_per_second": 4667.24
    },
    {
      "epoch": 0.9394132653061225,
      "grad_norm": 0.6413611906967972,
      "learning_rate": 5.476126358804594e-06,
      "loss": 0.30727410316467285,
      "num_input_tokens_seen": 441574960,
      "step": 7365,
      "train_runtime": 94612.4054,
      "train_tokens_per_second": 4667.199
    },
    {
      "epoch": 0.9400510204081632,
      "grad_norm": 0.5797745517308095,
      "learning_rate": 5.471139965644108e-06,
      "loss": 0.296018648147583,
      "num_input_tokens_seen": 441853960,
      "step": 7370,
      "train_runtime": 94673.828,
      "train_tokens_per_second": 4667.118
    },
    {
      "epoch": 0.9406887755102041,
      "grad_norm": 0.5731409662928846,
      "learning_rate": 5.46615309966176e-06,
      "loss": 0.3353337287902832,
      "num_input_tokens_seen": 442166752,
      "step": 7375,
      "train_runtime": 94732.356,
      "train_tokens_per_second": 4667.537
    },
    {
      "epoch": 0.9413265306122449,
      "grad_norm": 0.6622186231748212,
      "learning_rate": 5.461165765862223e-06,
      "loss": 0.3128879308700562,
      "num_input_tokens_seen": 442456888,
      "step": 7380,
      "train_runtime": 94794.4704,
      "train_tokens_per_second": 4667.539
    },
    {
      "epoch": 0.9419642857142857,
      "grad_norm": 0.6549417567270942,
      "learning_rate": 5.456177969250632e-06,
      "loss": 0.3014249324798584,
      "num_input_tokens_seen": 442747864,
      "step": 7385,
      "train_runtime": 94863.7335,
      "train_tokens_per_second": 4667.198
    },
    {
      "epoch": 0.9426020408163265,
      "grad_norm": 0.6086689270822957,
      "learning_rate": 5.451189714832588e-06,
      "loss": 0.3215111255645752,
      "num_input_tokens_seen": 443043328,
      "step": 7390,
      "train_runtime": 94929.0736,
      "train_tokens_per_second": 4667.098
    },
    {
      "epoch": 0.9432397959183674,
      "grad_norm": 0.5829099727877867,
      "learning_rate": 5.446201007614156e-06,
      "loss": 0.2938807487487793,
      "num_input_tokens_seen": 443333072,
      "step": 7395,
      "train_runtime": 94991.3903,
      "train_tokens_per_second": 4667.087
    },
    {
      "epoch": 0.9438775510204082,
      "grad_norm": 0.6091982000250068,
      "learning_rate": 5.441211852601849e-06,
      "loss": 0.32256112098693845,
      "num_input_tokens_seen": 443631760,
      "step": 7400,
      "train_runtime": 95066.4991,
      "train_tokens_per_second": 4666.541
    },
    {
      "epoch": 0.9445153061224489,
      "grad_norm": 0.6618698775504013,
      "learning_rate": 5.4362222548026325e-06,
      "loss": 0.33964810371398924,
      "num_input_tokens_seen": 443919000,
      "step": 7405,
      "train_runtime": 95133.364,
      "train_tokens_per_second": 4666.281
    },
    {
      "epoch": 0.9451530612244898,
      "grad_norm": 0.5938798444680511,
      "learning_rate": 5.43123221922392e-06,
      "loss": 0.31477525234222414,
      "num_input_tokens_seen": 444222072,
      "step": 7410,
      "train_runtime": 95198.4376,
      "train_tokens_per_second": 4666.275
    },
    {
      "epoch": 0.9457908163265306,
      "grad_norm": 0.596341767375758,
      "learning_rate": 5.426241750873557e-06,
      "loss": 0.31355721950531007,
      "num_input_tokens_seen": 444526456,
      "step": 7415,
      "train_runtime": 95262.728,
      "train_tokens_per_second": 4666.321
    },
    {
      "epoch": 0.9464285714285714,
      "grad_norm": 0.816500727563092,
      "learning_rate": 5.42125085475983e-06,
      "loss": 0.3167865753173828,
      "num_input_tokens_seen": 444840448,
      "step": 7420,
      "train_runtime": 95334.2699,
      "train_tokens_per_second": 4666.113
    },
    {
      "epoch": 0.9470663265306123,
      "grad_norm": 0.5760463089242822,
      "learning_rate": 5.4162595358914475e-06,
      "loss": 0.3132263422012329,
      "num_input_tokens_seen": 445140360,
      "step": 7425,
      "train_runtime": 95407.5386,
      "train_tokens_per_second": 4665.673
    },
    {
      "epoch": 0.9477040816326531,
      "grad_norm": 0.6454927109178517,
      "learning_rate": 5.41126779927755e-06,
      "loss": 0.2961220979690552,
      "num_input_tokens_seen": 445441328,
      "step": 7430,
      "train_runtime": 95477.4718,
      "train_tokens_per_second": 4665.408
    },
    {
      "epoch": 0.9483418367346939,
      "grad_norm": 0.6100564771314775,
      "learning_rate": 5.4062756499276945e-06,
      "loss": 0.3174859046936035,
      "num_input_tokens_seen": 445745480,
      "step": 7435,
      "train_runtime": 95541.4066,
      "train_tokens_per_second": 4665.469
    },
    {
      "epoch": 0.9489795918367347,
      "grad_norm": 0.670902823072716,
      "learning_rate": 5.401283092851848e-06,
      "loss": 0.3194777011871338,
      "num_input_tokens_seen": 446040072,
      "step": 7440,
      "train_runtime": 95602.6469,
      "train_tokens_per_second": 4665.562
    },
    {
      "epoch": 0.9496173469387755,
      "grad_norm": 0.5924655478586123,
      "learning_rate": 5.396290133060394e-06,
      "loss": 0.3156425952911377,
      "num_input_tokens_seen": 446342496,
      "step": 7445,
      "train_runtime": 95670.2199,
      "train_tokens_per_second": 4665.428
    },
    {
      "epoch": 0.9502551020408163,
      "grad_norm": 0.6293344566850531,
      "learning_rate": 5.391296775564114e-06,
      "loss": 0.35019216537475584,
      "num_input_tokens_seen": 446653656,
      "step": 7450,
      "train_runtime": 95736.8092,
      "train_tokens_per_second": 4665.433
    },
    {
      "epoch": 0.9508928571428571,
      "grad_norm": 0.5680684445110501,
      "learning_rate": 5.386303025374192e-06,
      "loss": 0.3140332460403442,
      "num_input_tokens_seen": 446957728,
      "step": 7455,
      "train_runtime": 95801.9799,
      "train_tokens_per_second": 4665.433
    },
    {
      "epoch": 0.951530612244898,
      "grad_norm": 0.6153391098064949,
      "learning_rate": 5.381308887502204e-06,
      "loss": 0.31685566902160645,
      "num_input_tokens_seen": 447262552,
      "step": 7460,
      "train_runtime": 95873.869,
      "train_tokens_per_second": 4665.114
    },
    {
      "epoch": 0.9521683673469388,
      "grad_norm": 0.5404774882365914,
      "learning_rate": 5.376314366960118e-06,
      "loss": 0.32549347877502444,
      "num_input_tokens_seen": 447571488,
      "step": 7465,
      "train_runtime": 95938.0777,
      "train_tokens_per_second": 4665.212
    },
    {
      "epoch": 0.9528061224489796,
      "grad_norm": 0.6105593376103999,
      "learning_rate": 5.371319468760283e-06,
      "loss": 0.3242738485336304,
      "num_input_tokens_seen": 447887584,
      "step": 7470,
      "train_runtime": 96006.8173,
      "train_tokens_per_second": 4665.164
    },
    {
      "epoch": 0.9534438775510204,
      "grad_norm": 0.5776911685050845,
      "learning_rate": 5.366324197915426e-06,
      "loss": 0.30603318214416503,
      "num_input_tokens_seen": 448191752,
      "step": 7475,
      "train_runtime": 96073.7649,
      "train_tokens_per_second": 4665.08
    },
    {
      "epoch": 0.9540816326530612,
      "grad_norm": 0.5698671243647103,
      "learning_rate": 5.361328559438655e-06,
      "loss": 0.3016000747680664,
      "num_input_tokens_seen": 448502592,
      "step": 7480,
      "train_runtime": 96149.5553,
      "train_tokens_per_second": 4664.635
    },
    {
      "epoch": 0.954719387755102,
      "grad_norm": 0.6147263058767498,
      "learning_rate": 5.3563325583434376e-06,
      "loss": 0.327193021774292,
      "num_input_tokens_seen": 448812280,
      "step": 7485,
      "train_runtime": 96212.5883,
      "train_tokens_per_second": 4664.798
    },
    {
      "epoch": 0.9553571428571429,
      "grad_norm": 0.591686025763404,
      "learning_rate": 5.351336199643612e-06,
      "loss": 0.3120138645172119,
      "num_input_tokens_seen": 449115768,
      "step": 7490,
      "train_runtime": 96273.7713,
      "train_tokens_per_second": 4664.986
    },
    {
      "epoch": 0.9559948979591837,
      "grad_norm": 0.5987198258971944,
      "learning_rate": 5.346339488353374e-06,
      "loss": 0.2821963310241699,
      "num_input_tokens_seen": 449426136,
      "step": 7495,
      "train_runtime": 96340.1704,
      "train_tokens_per_second": 4664.992
    },
    {
      "epoch": 0.9566326530612245,
      "grad_norm": 0.5593344935745522,
      "learning_rate": 5.341342429487271e-06,
      "loss": 0.3055206537246704,
      "num_input_tokens_seen": 449735360,
      "step": 7500,
      "train_runtime": 96395.2839,
      "train_tokens_per_second": 4665.533
    },
    {
      "epoch": 0.9572704081632653,
      "grad_norm": 0.5987260749649931,
      "learning_rate": 5.336345028060199e-06,
      "loss": 0.3117807388305664,
      "num_input_tokens_seen": 450037880,
      "step": 7505,
      "train_runtime": 96443.9631,
      "train_tokens_per_second": 4666.315
    },
    {
      "epoch": 0.9579081632653061,
      "grad_norm": 0.6234861581669837,
      "learning_rate": 5.331347289087402e-06,
      "loss": 0.30819096565246584,
      "num_input_tokens_seen": 450333008,
      "step": 7510,
      "train_runtime": 96499.3839,
      "train_tokens_per_second": 4666.693
    },
    {
      "epoch": 0.9585459183673469,
      "grad_norm": 0.6060985686918107,
      "learning_rate": 5.326349217584459e-06,
      "loss": 0.29360136985778806,
      "num_input_tokens_seen": 450608408,
      "step": 7515,
      "train_runtime": 96566.1646,
      "train_tokens_per_second": 4666.318
    },
    {
      "epoch": 0.9591836734693877,
      "grad_norm": 0.5757095572868233,
      "learning_rate": 5.321350818567285e-06,
      "loss": 0.2921141624450684,
      "num_input_tokens_seen": 450896456,
      "step": 7520,
      "train_runtime": 96622.3478,
      "train_tokens_per_second": 4666.586
    },
    {
      "epoch": 0.9598214285714286,
      "grad_norm": 0.6448208027024613,
      "learning_rate": 5.3163520970521194e-06,
      "loss": 0.3008954286575317,
      "num_input_tokens_seen": 451192672,
      "step": 7525,
      "train_runtime": 96678.0882,
      "train_tokens_per_second": 4666.959
    },
    {
      "epoch": 0.9604591836734694,
      "grad_norm": 0.6534088140144655,
      "learning_rate": 5.311353058055533e-06,
      "loss": 0.3075784921646118,
      "num_input_tokens_seen": 451490672,
      "step": 7530,
      "train_runtime": 96751.2105,
      "train_tokens_per_second": 4666.512
    },
    {
      "epoch": 0.9610969387755102,
      "grad_norm": 0.6482668306282363,
      "learning_rate": 5.306353706594407e-06,
      "loss": 0.3319397926330566,
      "num_input_tokens_seen": 451796936,
      "step": 7535,
      "train_runtime": 96815.6519,
      "train_tokens_per_second": 4666.569
    },
    {
      "epoch": 0.9617346938775511,
      "grad_norm": 0.6424720638157772,
      "learning_rate": 5.3013540476859404e-06,
      "loss": 0.3318361520767212,
      "num_input_tokens_seen": 452106944,
      "step": 7540,
      "train_runtime": 96884.3575,
      "train_tokens_per_second": 4666.46
    },
    {
      "epoch": 0.9623724489795918,
      "grad_norm": 0.6268719170355765,
      "learning_rate": 5.2963540863476404e-06,
      "loss": 0.33273561000823976,
      "num_input_tokens_seen": 452387040,
      "step": 7545,
      "train_runtime": 96946.3593,
      "train_tokens_per_second": 4666.364
    },
    {
      "epoch": 0.9630102040816326,
      "grad_norm": 0.6354858722146296,
      "learning_rate": 5.291353827597317e-06,
      "loss": 0.28787033557891845,
      "num_input_tokens_seen": 452697032,
      "step": 7550,
      "train_runtime": 97014.8052,
      "train_tokens_per_second": 4666.267
    },
    {
      "epoch": 0.9636479591836735,
      "grad_norm": 0.600150934837588,
      "learning_rate": 5.28635327645308e-06,
      "loss": 0.32324178218841554,
      "num_input_tokens_seen": 452992904,
      "step": 7555,
      "train_runtime": 97081.2505,
      "train_tokens_per_second": 4666.121
    },
    {
      "epoch": 0.9642857142857143,
      "grad_norm": 0.5994538828916294,
      "learning_rate": 5.281352437933331e-06,
      "loss": 0.33991477489471433,
      "num_input_tokens_seen": 453279512,
      "step": 7560,
      "train_runtime": 97141.0514,
      "train_tokens_per_second": 4666.199
    },
    {
      "epoch": 0.9649234693877551,
      "grad_norm": 0.6487132106250739,
      "learning_rate": 5.276351317056762e-06,
      "loss": 0.3196584463119507,
      "num_input_tokens_seen": 453575816,
      "step": 7565,
      "train_runtime": 97203.0516,
      "train_tokens_per_second": 4666.271
    },
    {
      "epoch": 0.9655612244897959,
      "grad_norm": 0.546755169161235,
      "learning_rate": 5.271349918842344e-06,
      "loss": 0.33159217834472654,
      "num_input_tokens_seen": 453874112,
      "step": 7570,
      "train_runtime": 97262.0756,
      "train_tokens_per_second": 4666.507
    },
    {
      "epoch": 0.9661989795918368,
      "grad_norm": 0.5919380980375913,
      "learning_rate": 5.266348248309332e-06,
      "loss": 0.3290063142776489,
      "num_input_tokens_seen": 454165920,
      "step": 7575,
      "train_runtime": 97322.3772,
      "train_tokens_per_second": 4666.614
    },
    {
      "epoch": 0.9668367346938775,
      "grad_norm": 0.5550551918094889,
      "learning_rate": 5.2613463104772535e-06,
      "loss": 0.2901280879974365,
      "num_input_tokens_seen": 454470512,
      "step": 7580,
      "train_runtime": 97390.3922,
      "train_tokens_per_second": 4666.482
    },
    {
      "epoch": 0.9674744897959183,
      "grad_norm": 0.589767468565175,
      "learning_rate": 5.256344110365896e-06,
      "loss": 0.3240648746490479,
      "num_input_tokens_seen": 454775944,
      "step": 7585,
      "train_runtime": 97454.1725,
      "train_tokens_per_second": 4666.562
    },
    {
      "epoch": 0.9681122448979592,
      "grad_norm": 0.6522874216313619,
      "learning_rate": 5.251341652995325e-06,
      "loss": 0.33693685531616213,
      "num_input_tokens_seen": 455077336,
      "step": 7590,
      "train_runtime": 97514.4635,
      "train_tokens_per_second": 4666.768
    },
    {
      "epoch": 0.96875,
      "grad_norm": 0.5625013432026852,
      "learning_rate": 5.246338943385849e-06,
      "loss": 0.34079480171203613,
      "num_input_tokens_seen": 455362376,
      "step": 7595,
      "train_runtime": 97575.7883,
      "train_tokens_per_second": 4666.756
    },
    {
      "epoch": 0.9693877551020408,
      "grad_norm": 0.6701653593534399,
      "learning_rate": 5.24133598655804e-06,
      "loss": 0.3260797023773193,
      "num_input_tokens_seen": 455661464,
      "step": 7600,
      "train_runtime": 97637.8168,
      "train_tokens_per_second": 4666.854
    },
    {
      "epoch": 0.9700255102040817,
      "grad_norm": 0.6336350263548726,
      "learning_rate": 5.2363327875327115e-06,
      "loss": 0.29935464859008787,
      "num_input_tokens_seen": 455954448,
      "step": 7605,
      "train_runtime": 97705.3349,
      "train_tokens_per_second": 4666.628
    },
    {
      "epoch": 0.9706632653061225,
      "grad_norm": 0.5434675291168316,
      "learning_rate": 5.231329351330927e-06,
      "loss": 0.29101612567901614,
      "num_input_tokens_seen": 456271840,
      "step": 7610,
      "train_runtime": 97766.1481,
      "train_tokens_per_second": 4666.972
    },
    {
      "epoch": 0.9713010204081632,
      "grad_norm": 0.5576678165573105,
      "learning_rate": 5.226325682973984e-06,
      "loss": 0.3211846351623535,
      "num_input_tokens_seen": 456578832,
      "step": 7615,
      "train_runtime": 97829.9184,
      "train_tokens_per_second": 4667.067
    },
    {
      "epoch": 0.9719387755102041,
      "grad_norm": 0.6763873645307562,
      "learning_rate": 5.221321787483409e-06,
      "loss": 0.3094940662384033,
      "num_input_tokens_seen": 456888984,
      "step": 7620,
      "train_runtime": 97893.9988,
      "train_tokens_per_second": 4667.181
    },
    {
      "epoch": 0.9725765306122449,
      "grad_norm": 0.6651163303626488,
      "learning_rate": 5.216317669880965e-06,
      "loss": 0.3145118236541748,
      "num_input_tokens_seen": 457173616,
      "step": 7625,
      "train_runtime": 97961.9243,
      "train_tokens_per_second": 4666.85
    },
    {
      "epoch": 0.9732142857142857,
      "grad_norm": 0.5701022534247328,
      "learning_rate": 5.21131333518863e-06,
      "loss": 0.3109163284301758,
      "num_input_tokens_seen": 457485832,
      "step": 7630,
      "train_runtime": 98034.3858,
      "train_tokens_per_second": 4666.585
    },
    {
      "epoch": 0.9738520408163265,
      "grad_norm": 0.6080132650043788,
      "learning_rate": 5.206308788428604e-06,
      "loss": 0.32394387722015383,
      "num_input_tokens_seen": 457789536,
      "step": 7635,
      "train_runtime": 98109.9725,
      "train_tokens_per_second": 4666.086
    },
    {
      "epoch": 0.9744897959183674,
      "grad_norm": 0.6046240464616601,
      "learning_rate": 5.201304034623303e-06,
      "loss": 0.3452263355255127,
      "num_input_tokens_seen": 458099200,
      "step": 7640,
      "train_runtime": 98172.8133,
      "train_tokens_per_second": 4666.253
    },
    {
      "epoch": 0.9751275510204082,
      "grad_norm": 0.6475743563371117,
      "learning_rate": 5.1962990787953436e-06,
      "loss": 0.32486186027526853,
      "num_input_tokens_seen": 458388136,
      "step": 7645,
      "train_runtime": 98242.5093,
      "train_tokens_per_second": 4665.884
    },
    {
      "epoch": 0.9757653061224489,
      "grad_norm": 0.5980063889313315,
      "learning_rate": 5.191293925967548e-06,
      "loss": 0.29793720245361327,
      "num_input_tokens_seen": 458675736,
      "step": 7650,
      "train_runtime": 98310.9341,
      "train_tokens_per_second": 4665.562
    },
    {
      "epoch": 0.9764030612244898,
      "grad_norm": 0.5292021897954804,
      "learning_rate": 5.186288581162939e-06,
      "loss": 0.3058775424957275,
      "num_input_tokens_seen": 458982360,
      "step": 7655,
      "train_runtime": 98377.2266,
      "train_tokens_per_second": 4665.535
    },
    {
      "epoch": 0.9770408163265306,
      "grad_norm": 0.5996618706074013,
      "learning_rate": 5.181283049404728e-06,
      "loss": 0.30587172508239746,
      "num_input_tokens_seen": 459279984,
      "step": 7660,
      "train_runtime": 98450.0886,
      "train_tokens_per_second": 4665.105
    },
    {
      "epoch": 0.9776785714285714,
      "grad_norm": 0.5248917801822073,
      "learning_rate": 5.1762773357163175e-06,
      "loss": 0.27962732315063477,
      "num_input_tokens_seen": 459570208,
      "step": 7665,
      "train_runtime": 98512.6984,
      "train_tokens_per_second": 4665.086
    },
    {
      "epoch": 0.9783163265306123,
      "grad_norm": 0.5926032130293605,
      "learning_rate": 5.171271445121286e-06,
      "loss": 0.2774610996246338,
      "num_input_tokens_seen": 459860232,
      "step": 7670,
      "train_runtime": 98570.0192,
      "train_tokens_per_second": 4665.315
    },
    {
      "epoch": 0.9789540816326531,
      "grad_norm": 0.6484820227148456,
      "learning_rate": 5.166265382643402e-06,
      "loss": 0.3253363370895386,
      "num_input_tokens_seen": 460147904,
      "step": 7675,
      "train_runtime": 98639.7667,
      "train_tokens_per_second": 4664.933
    },
    {
      "epoch": 0.9795918367346939,
      "grad_norm": 0.4907343854867156,
      "learning_rate": 5.161259153306592e-06,
      "loss": 0.27699477672576905,
      "num_input_tokens_seen": 460444704,
      "step": 7680,
      "train_runtime": 98707.1327,
      "train_tokens_per_second": 4664.756
    },
    {
      "epoch": 0.9802295918367347,
      "grad_norm": 0.6138292638952311,
      "learning_rate": 5.156252762134959e-06,
      "loss": 0.29723730087280276,
      "num_input_tokens_seen": 460735744,
      "step": 7685,
      "train_runtime": 98780.7921,
      "train_tokens_per_second": 4664.224
    },
    {
      "epoch": 0.9808673469387755,
      "grad_norm": 0.5790012334958022,
      "learning_rate": 5.1512462141527675e-06,
      "loss": 0.31710524559020997,
      "num_input_tokens_seen": 461045584,
      "step": 7690,
      "train_runtime": 98848.0018,
      "train_tokens_per_second": 4664.187
    },
    {
      "epoch": 0.9815051020408163,
      "grad_norm": 0.5729531652566596,
      "learning_rate": 5.146239514384437e-06,
      "loss": 0.32338848114013674,
      "num_input_tokens_seen": 461354272,
      "step": 7695,
      "train_runtime": 98905.6236,
      "train_tokens_per_second": 4664.591
    },
    {
      "epoch": 0.9821428571428571,
      "grad_norm": 0.605641575088013,
      "learning_rate": 5.14123266785454e-06,
      "loss": 0.3240955352783203,
      "num_input_tokens_seen": 461642712,
      "step": 7700,
      "train_runtime": 98973.9734,
      "train_tokens_per_second": 4664.284
    },
    {
      "epoch": 0.982780612244898,
      "grad_norm": 0.5543165704011824,
      "learning_rate": 5.136225679587797e-06,
      "loss": 0.29973626136779785,
      "num_input_tokens_seen": 461941752,
      "step": 7705,
      "train_runtime": 99031.9811,
      "train_tokens_per_second": 4664.571
    },
    {
      "epoch": 0.9834183673469388,
      "grad_norm": 0.5908493236810771,
      "learning_rate": 5.131218554609072e-06,
      "loss": 0.3147848129272461,
      "num_input_tokens_seen": 462253096,
      "step": 7710,
      "train_runtime": 99095.786,
      "train_tokens_per_second": 4664.71
    },
    {
      "epoch": 0.9840561224489796,
      "grad_norm": 0.5496162048277468,
      "learning_rate": 5.126211297943362e-06,
      "loss": 0.3021842002868652,
      "num_input_tokens_seen": 462550192,
      "step": 7715,
      "train_runtime": 99158.9418,
      "train_tokens_per_second": 4664.735
    },
    {
      "epoch": 0.9846938775510204,
      "grad_norm": 0.6397961246977143,
      "learning_rate": 5.121203914615799e-06,
      "loss": 0.3146205425262451,
      "num_input_tokens_seen": 462872160,
      "step": 7720,
      "train_runtime": 99217.5475,
      "train_tokens_per_second": 4665.225
    },
    {
      "epoch": 0.9853316326530612,
      "grad_norm": 0.5947195068505116,
      "learning_rate": 5.116196409651645e-06,
      "loss": 0.3136601448059082,
      "num_input_tokens_seen": 463170848,
      "step": 7725,
      "train_runtime": 99281.4491,
      "train_tokens_per_second": 4665.231
    },
    {
      "epoch": 0.985969387755102,
      "grad_norm": 0.583155497751477,
      "learning_rate": 5.111188788076278e-06,
      "loss": 0.33538773059844973,
      "num_input_tokens_seen": 463476088,
      "step": 7730,
      "train_runtime": 99361.6793,
      "train_tokens_per_second": 4664.536
    },
    {
      "epoch": 0.9866071428571429,
      "grad_norm": 0.6345423842564097,
      "learning_rate": 5.106181054915196e-06,
      "loss": 0.32474794387817385,
      "num_input_tokens_seen": 463795016,
      "step": 7735,
      "train_runtime": 99415.135,
      "train_tokens_per_second": 4665.235
    },
    {
      "epoch": 0.9872448979591837,
      "grad_norm": 0.6211131091765834,
      "learning_rate": 5.10117321519401e-06,
      "loss": 0.3254901885986328,
      "num_input_tokens_seen": 464095728,
      "step": 7740,
      "train_runtime": 99477.4373,
      "train_tokens_per_second": 4665.337
    },
    {
      "epoch": 0.9878826530612245,
      "grad_norm": 0.6283576703196156,
      "learning_rate": 5.0961652739384356e-06,
      "loss": 0.29070339202880857,
      "num_input_tokens_seen": 464401824,
      "step": 7745,
      "train_runtime": 99541.3676,
      "train_tokens_per_second": 4665.415
    },
    {
      "epoch": 0.9885204081632653,
      "grad_norm": 0.6765271402073013,
      "learning_rate": 5.091157236174295e-06,
      "loss": 0.3426629066467285,
      "num_input_tokens_seen": 464722208,
      "step": 7750,
      "train_runtime": 99592.6593,
      "train_tokens_per_second": 4666.23
    },
    {
      "epoch": 0.9891581632653061,
      "grad_norm": 0.540617019497874,
      "learning_rate": 5.086149106927499e-06,
      "loss": 0.2858463525772095,
      "num_input_tokens_seen": 465023720,
      "step": 7755,
      "train_runtime": 99667.3487,
      "train_tokens_per_second": 4665.758
    },
    {
      "epoch": 0.9897959183673469,
      "grad_norm": 0.6074886827015675,
      "learning_rate": 5.081140891224061e-06,
      "loss": 0.28305037021636964,
      "num_input_tokens_seen": 465316320,
      "step": 7760,
      "train_runtime": 99736.8677,
      "train_tokens_per_second": 4665.439
    },
    {
      "epoch": 0.9904336734693877,
      "grad_norm": 0.5861559786845644,
      "learning_rate": 5.076132594090068e-06,
      "loss": 0.30130348205566404,
      "num_input_tokens_seen": 465627800,
      "step": 7765,
      "train_runtime": 99786.5775,
      "train_tokens_per_second": 4666.237
    },
    {
      "epoch": 0.9910714285714286,
      "grad_norm": 0.603733945471739,
      "learning_rate": 5.071124220551703e-06,
      "loss": 0.30890641212463377,
      "num_input_tokens_seen": 465929352,
      "step": 7770,
      "train_runtime": 99859.418,
      "train_tokens_per_second": 4665.853
    },
    {
      "epoch": 0.9917091836734694,
      "grad_norm": 0.4571867681651973,
      "learning_rate": 5.066115775635213e-06,
      "loss": 0.30515527725219727,
      "num_input_tokens_seen": 466240856,
      "step": 7775,
      "train_runtime": 99924.8984,
      "train_tokens_per_second": 4665.913
    },
    {
      "epoch": 0.9923469387755102,
      "grad_norm": 0.6394776832577932,
      "learning_rate": 5.061107264366926e-06,
      "loss": 0.2977296829223633,
      "num_input_tokens_seen": 466530752,
      "step": 7780,
      "train_runtime": 99982.0987,
      "train_tokens_per_second": 4666.143
    },
    {
      "epoch": 0.9929846938775511,
      "grad_norm": 0.6321535056897735,
      "learning_rate": 5.05609869177323e-06,
      "loss": 0.30818977355957033,
      "num_input_tokens_seen": 466843744,
      "step": 7785,
      "train_runtime": 100056.2757,
      "train_tokens_per_second": 4665.812
    },
    {
      "epoch": 0.9936224489795918,
      "grad_norm": 0.5400668637736669,
      "learning_rate": 5.05109006288058e-06,
      "loss": 0.31366040706634524,
      "num_input_tokens_seen": 467149864,
      "step": 7790,
      "train_runtime": 100125.7303,
      "train_tokens_per_second": 4665.633
    },
    {
      "epoch": 0.9942602040816326,
      "grad_norm": 0.6452088140800006,
      "learning_rate": 5.046081382715484e-06,
      "loss": 0.29803402423858644,
      "num_input_tokens_seen": 467449768,
      "step": 7795,
      "train_runtime": 100179.7084,
      "train_tokens_per_second": 4666.112
    },
    {
      "epoch": 0.9948979591836735,
      "grad_norm": 0.5701539179741821,
      "learning_rate": 5.0410726563045005e-06,
      "loss": 0.3019112586975098,
      "num_input_tokens_seen": 467753208,
      "step": 7800,
      "train_runtime": 100246.42,
      "train_tokens_per_second": 4666.034
    },
    {
      "epoch": 0.9955357142857143,
      "grad_norm": 0.5601865942144285,
      "learning_rate": 5.036063888674239e-06,
      "loss": 0.3023086071014404,
      "num_input_tokens_seen": 468059360,
      "step": 7805,
      "train_runtime": 100315.9634,
      "train_tokens_per_second": 4665.851
    },
    {
      "epoch": 0.9961734693877551,
      "grad_norm": 0.6097741741301059,
      "learning_rate": 5.031055084851349e-06,
      "loss": 0.28096904754638674,
      "num_input_tokens_seen": 468363536,
      "step": 7810,
      "train_runtime": 100375.4937,
      "train_tokens_per_second": 4666.114
    },
    {
      "epoch": 0.9968112244897959,
      "grad_norm": 0.5597421027232963,
      "learning_rate": 5.0260462498625084e-06,
      "loss": 0.318403434753418,
      "num_input_tokens_seen": 468668776,
      "step": 7815,
      "train_runtime": 100438.6541,
      "train_tokens_per_second": 4666.219
    },
    {
      "epoch": 0.9974489795918368,
      "grad_norm": 0.6143193062402761,
      "learning_rate": 5.021037388734439e-06,
      "loss": 0.31057963371276853,
      "num_input_tokens_seen": 468978488,
      "step": 7820,
      "train_runtime": 100502.1577,
      "train_tokens_per_second": 4666.352
    },
    {
      "epoch": 0.9980867346938775,
      "grad_norm": 0.6202264825508076,
      "learning_rate": 5.016028506493881e-06,
      "loss": 0.29598608016967776,
      "num_input_tokens_seen": 469295688,
      "step": 7825,
      "train_runtime": 100569.2484,
      "train_tokens_per_second": 4666.394
    },
    {
      "epoch": 0.9987244897959183,
      "grad_norm": 0.5462962001030922,
      "learning_rate": 5.011019608167596e-06,
      "loss": 0.29022941589355467,
      "num_input_tokens_seen": 469598840,
      "step": 7830,
      "train_runtime": 100631.4582,
      "train_tokens_per_second": 4666.521
    },
    {
      "epoch": 0.9993622448979592,
      "grad_norm": 0.6483707288128262,
      "learning_rate": 5.006010698782365e-06,
      "loss": 0.32463507652282714,
      "num_input_tokens_seen": 469890360,
      "step": 7835,
      "train_runtime": 100695.154,
      "train_tokens_per_second": 4666.464
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.7132004143103949,
      "learning_rate": 5.001001783364978e-06,
      "loss": 0.3155714511871338,
      "num_input_tokens_seen": 470193152,
      "step": 7840,
      "train_runtime": 100761.1902,
      "train_tokens_per_second": 4666.411
    },
    {
      "epoch": 1.000637755102041,
      "grad_norm": 0.7063670221257499,
      "learning_rate": 4.995992866942232e-06,
      "loss": 0.27135615348815917,
      "num_input_tokens_seen": 470501136,
      "step": 7845,
      "train_runtime": 100855.7953,
      "train_tokens_per_second": 4665.088
    },
    {
      "epoch": 1.0012755102040816,
      "grad_norm": 0.7205766028299049,
      "learning_rate": 4.990983954540925e-06,
      "loss": 0.23918299674987792,
      "num_input_tokens_seen": 470799744,
      "step": 7850,
      "train_runtime": 100923.0157,
      "train_tokens_per_second": 4664.939
    },
    {
      "epoch": 1.0019132653061225,
      "grad_norm": 0.7440549615755528,
      "learning_rate": 4.985975051187848e-06,
      "loss": 0.2603339672088623,
      "num_input_tokens_seen": 471113560,
      "step": 7855,
      "train_runtime": 100975.9239,
      "train_tokens_per_second": 4665.603
    },
    {
      "epoch": 1.0025510204081634,
      "grad_norm": 0.6740790164198203,
      "learning_rate": 4.980966161909785e-06,
      "loss": 0.2453685760498047,
      "num_input_tokens_seen": 471406032,
      "step": 7860,
      "train_runtime": 101042.6418,
      "train_tokens_per_second": 4665.417
    },
    {
      "epoch": 1.003188775510204,
      "grad_norm": 0.5965397254378703,
      "learning_rate": 4.975957291733511e-06,
      "loss": 0.25345773696899415,
      "num_input_tokens_seen": 471722416,
      "step": 7865,
      "train_runtime": 101116.116,
      "train_tokens_per_second": 4665.156
    },
    {
      "epoch": 1.003826530612245,
      "grad_norm": 0.6354748638878189,
      "learning_rate": 4.970948445685772e-06,
      "loss": 0.22407064437866211,
      "num_input_tokens_seen": 471995736,
      "step": 7870,
      "train_runtime": 101184.7663,
      "train_tokens_per_second": 4664.692
    },
    {
      "epoch": 1.0044642857142858,
      "grad_norm": 0.6967893112176949,
      "learning_rate": 4.965939628793295e-06,
      "loss": 0.24353766441345215,
      "num_input_tokens_seen": 472305800,
      "step": 7875,
      "train_runtime": 101256.5768,
      "train_tokens_per_second": 4664.446
    },
    {
      "epoch": 1.0051020408163265,
      "grad_norm": 0.676968804710127,
      "learning_rate": 4.960930846082783e-06,
      "loss": 0.2777569770812988,
      "num_input_tokens_seen": 472613936,
      "step": 7880,
      "train_runtime": 101322.7275,
      "train_tokens_per_second": 4664.442
    },
    {
      "epoch": 1.0057397959183674,
      "grad_norm": 0.5643974833710756,
      "learning_rate": 4.955922102580894e-06,
      "loss": 0.24412152767181397,
      "num_input_tokens_seen": 472915392,
      "step": 7885,
      "train_runtime": 101395.3722,
      "train_tokens_per_second": 4664.073
    },
    {
      "epoch": 1.0063775510204083,
      "grad_norm": 0.7241753470400919,
      "learning_rate": 4.9509134033142525e-06,
      "loss": 0.23244500160217285,
      "num_input_tokens_seen": 473204720,
      "step": 7890,
      "train_runtime": 101457.5229,
      "train_tokens_per_second": 4664.067
    },
    {
      "epoch": 1.007015306122449,
      "grad_norm": 0.7318825243421886,
      "learning_rate": 4.945904753309441e-06,
      "loss": 0.27535371780395507,
      "num_input_tokens_seen": 473506616,
      "step": 7895,
      "train_runtime": 101528.1269,
      "train_tokens_per_second": 4663.797
    },
    {
      "epoch": 1.0076530612244898,
      "grad_norm": 0.6395240024520434,
      "learning_rate": 4.940896157592985e-06,
      "loss": 0.23467187881469725,
      "num_input_tokens_seen": 473805792,
      "step": 7900,
      "train_runtime": 101600.7105,
      "train_tokens_per_second": 4663.41
    },
    {
      "epoch": 1.0082908163265305,
      "grad_norm": 0.6749424965862421,
      "learning_rate": 4.935887621191364e-06,
      "loss": 0.25825228691101076,
      "num_input_tokens_seen": 474115248,
      "step": 7905,
      "train_runtime": 101661.7366,
      "train_tokens_per_second": 4663.655
    },
    {
      "epoch": 1.0089285714285714,
      "grad_norm": 0.6464308565745938,
      "learning_rate": 4.930879149130993e-06,
      "loss": 0.25428595542907717,
      "num_input_tokens_seen": 474418992,
      "step": 7910,
      "train_runtime": 101724.7789,
      "train_tokens_per_second": 4663.751
    },
    {
      "epoch": 1.0095663265306123,
      "grad_norm": 0.7086695672538507,
      "learning_rate": 4.925870746438223e-06,
      "loss": 0.2341998815536499,
      "num_input_tokens_seen": 474698872,
      "step": 7915,
      "train_runtime": 101786.7418,
      "train_tokens_per_second": 4663.661
    },
    {
      "epoch": 1.010204081632653,
      "grad_norm": 0.6683637085600683,
      "learning_rate": 4.920862418139335e-06,
      "loss": 0.2417293071746826,
      "num_input_tokens_seen": 474994944,
      "step": 7920,
      "train_runtime": 101841.7269,
      "train_tokens_per_second": 4664.05
    },
    {
      "epoch": 1.0108418367346939,
      "grad_norm": 0.6913590181544073,
      "learning_rate": 4.91585416926054e-06,
      "loss": 0.24323151111602784,
      "num_input_tokens_seen": 475289264,
      "step": 7925,
      "train_runtime": 101908.8462,
      "train_tokens_per_second": 4663.867
    },
    {
      "epoch": 1.0114795918367347,
      "grad_norm": 0.7226028388516762,
      "learning_rate": 4.91084600482796e-06,
      "loss": 0.23154821395874023,
      "num_input_tokens_seen": 475583128,
      "step": 7930,
      "train_runtime": 101981.3247,
      "train_tokens_per_second": 4663.434
    },
    {
      "epoch": 1.0121173469387754,
      "grad_norm": 0.6687960980981288,
      "learning_rate": 4.905837929867643e-06,
      "loss": 0.2589773893356323,
      "num_input_tokens_seen": 475902616,
      "step": 7935,
      "train_runtime": 102053.4561,
      "train_tokens_per_second": 4663.268
    },
    {
      "epoch": 1.0127551020408163,
      "grad_norm": 0.61616438449182,
      "learning_rate": 4.900829949405539e-06,
      "loss": 0.2532099723815918,
      "num_input_tokens_seen": 476190664,
      "step": 7940,
      "train_runtime": 102111.5179,
      "train_tokens_per_second": 4663.437
    },
    {
      "epoch": 1.0133928571428572,
      "grad_norm": 0.6123550636551225,
      "learning_rate": 4.895822068467505e-06,
      "loss": 0.2447188377380371,
      "num_input_tokens_seen": 476499112,
      "step": 7945,
      "train_runtime": 102170.5412,
      "train_tokens_per_second": 4663.762
    },
    {
      "epoch": 1.0140306122448979,
      "grad_norm": 0.6581367871078015,
      "learning_rate": 4.890814292079303e-06,
      "loss": 0.26943182945251465,
      "num_input_tokens_seen": 476815032,
      "step": 7950,
      "train_runtime": 102230.2551,
      "train_tokens_per_second": 4664.128
    },
    {
      "epoch": 1.0146683673469388,
      "grad_norm": 0.568421364437829,
      "learning_rate": 4.885806625266585e-06,
      "loss": 0.23051421642303466,
      "num_input_tokens_seen": 477098568,
      "step": 7955,
      "train_runtime": 102282.8105,
      "train_tokens_per_second": 4664.504
    },
    {
      "epoch": 1.0153061224489797,
      "grad_norm": 0.6099696747243039,
      "learning_rate": 4.8807990730548905e-06,
      "loss": 0.2716972351074219,
      "num_input_tokens_seen": 477390840,
      "step": 7960,
      "train_runtime": 102347.4804,
      "train_tokens_per_second": 4664.412
    },
    {
      "epoch": 1.0159438775510203,
      "grad_norm": 0.5391053729948323,
      "learning_rate": 4.8757916404696545e-06,
      "loss": 0.22613182067871093,
      "num_input_tokens_seen": 477687248,
      "step": 7965,
      "train_runtime": 102426.2004,
      "train_tokens_per_second": 4663.721
    },
    {
      "epoch": 1.0165816326530612,
      "grad_norm": 0.6559138211623041,
      "learning_rate": 4.870784332536181e-06,
      "loss": 0.24821786880493163,
      "num_input_tokens_seen": 477979368,
      "step": 7970,
      "train_runtime": 102486.4612,
      "train_tokens_per_second": 4663.829
    },
    {
      "epoch": 1.0172193877551021,
      "grad_norm": 0.6723493475992208,
      "learning_rate": 4.865777154279653e-06,
      "loss": 0.26028215885162354,
      "num_input_tokens_seen": 478280648,
      "step": 7975,
      "train_runtime": 102562.7945,
      "train_tokens_per_second": 4663.296
    },
    {
      "epoch": 1.0178571428571428,
      "grad_norm": 0.6210324868756046,
      "learning_rate": 4.8607701107251274e-06,
      "loss": 0.2637598514556885,
      "num_input_tokens_seen": 478594504,
      "step": 7980,
      "train_runtime": 102623.5945,
      "train_tokens_per_second": 4663.591
    },
    {
      "epoch": 1.0184948979591837,
      "grad_norm": 0.5864754377444793,
      "learning_rate": 4.855763206897516e-06,
      "loss": 0.25537154674530027,
      "num_input_tokens_seen": 478896512,
      "step": 7985,
      "train_runtime": 102685.038,
      "train_tokens_per_second": 4663.742
    },
    {
      "epoch": 1.0191326530612246,
      "grad_norm": 0.67423971781068,
      "learning_rate": 4.850756447821602e-06,
      "loss": 0.2521768569946289,
      "num_input_tokens_seen": 479186984,
      "step": 7990,
      "train_runtime": 102748.9622,
      "train_tokens_per_second": 4663.667
    },
    {
      "epoch": 1.0197704081632653,
      "grad_norm": 0.6232373731879575,
      "learning_rate": 4.845749838522016e-06,
      "loss": 0.22831923961639405,
      "num_input_tokens_seen": 479482192,
      "step": 7995,
      "train_runtime": 102822.2966,
      "train_tokens_per_second": 4663.212
    },
    {
      "epoch": 1.0204081632653061,
      "grad_norm": 0.6082933837002366,
      "learning_rate": 4.84074338402324e-06,
      "loss": 0.24029741287231446,
      "num_input_tokens_seen": 479777744,
      "step": 8000,
      "train_runtime": 102892.6301,
      "train_tokens_per_second": 4662.897
    },
    {
      "epoch": 1.021045918367347,
      "grad_norm": 0.6932153992534329,
      "learning_rate": 4.835737089349599e-06,
      "loss": 0.2580875396728516,
      "num_input_tokens_seen": 480085464,
      "step": 8005,
      "train_runtime": 102946.6047,
      "train_tokens_per_second": 4663.441
    },
    {
      "epoch": 1.0216836734693877,
      "grad_norm": 0.6398197766513816,
      "learning_rate": 4.8307309595252624e-06,
      "loss": 0.24136962890625,
      "num_input_tokens_seen": 480388920,
      "step": 8010,
      "train_runtime": 103008.574,
      "train_tokens_per_second": 4663.582
    },
    {
      "epoch": 1.0223214285714286,
      "grad_norm": 0.5655827604804655,
      "learning_rate": 4.82572499957423e-06,
      "loss": 0.22456057071685792,
      "num_input_tokens_seen": 480695024,
      "step": 8015,
      "train_runtime": 103063.8494,
      "train_tokens_per_second": 4664.051
    },
    {
      "epoch": 1.0229591836734695,
      "grad_norm": 0.6674421538864252,
      "learning_rate": 4.820719214520329e-06,
      "loss": 0.22296371459960937,
      "num_input_tokens_seen": 480982296,
      "step": 8020,
      "train_runtime": 103129.4477,
      "train_tokens_per_second": 4663.87
    },
    {
      "epoch": 1.0235969387755102,
      "grad_norm": 0.7010180058705325,
      "learning_rate": 4.815713609387221e-06,
      "loss": 0.22968318462371826,
      "num_input_tokens_seen": 481268960,
      "step": 8025,
      "train_runtime": 103200.1372,
      "train_tokens_per_second": 4663.453
    },
    {
      "epoch": 1.024234693877551,
      "grad_norm": 0.7424588837248032,
      "learning_rate": 4.810708189198374e-06,
      "loss": 0.25366778373718263,
      "num_input_tokens_seen": 481543296,
      "step": 8030,
      "train_runtime": 103274.6296,
      "train_tokens_per_second": 4662.745
    },
    {
      "epoch": 1.0248724489795917,
      "grad_norm": 0.6489336132554717,
      "learning_rate": 4.80570295897708e-06,
      "loss": 0.24692921638488768,
      "num_input_tokens_seen": 481829120,
      "step": 8035,
      "train_runtime": 103340.8353,
      "train_tokens_per_second": 4662.524
    },
    {
      "epoch": 1.0255102040816326,
      "grad_norm": 0.6126117737638476,
      "learning_rate": 4.800697923746437e-06,
      "loss": 0.24772284030914307,
      "num_input_tokens_seen": 482135760,
      "step": 8040,
      "train_runtime": 103390.5847,
      "train_tokens_per_second": 4663.246
    },
    {
      "epoch": 1.0261479591836735,
      "grad_norm": 0.6213374960515321,
      "learning_rate": 4.795693088529345e-06,
      "loss": 0.2315685510635376,
      "num_input_tokens_seen": 482444240,
      "step": 8045,
      "train_runtime": 103449.884,
      "train_tokens_per_second": 4663.555
    },
    {
      "epoch": 1.0267857142857142,
      "grad_norm": 0.6578325237121249,
      "learning_rate": 4.790688458348509e-06,
      "loss": 0.2314953088760376,
      "num_input_tokens_seen": 482724176,
      "step": 8050,
      "train_runtime": 103511.8759,
      "train_tokens_per_second": 4663.467
    },
    {
      "epoch": 1.027423469387755,
      "grad_norm": 0.6392076860059664,
      "learning_rate": 4.7856840382264234e-06,
      "loss": 0.2663999080657959,
      "num_input_tokens_seen": 483026872,
      "step": 8055,
      "train_runtime": 103584.7084,
      "train_tokens_per_second": 4663.11
    },
    {
      "epoch": 1.028061224489796,
      "grad_norm": 0.6271337962363529,
      "learning_rate": 4.780679833185372e-06,
      "loss": 0.237894868850708,
      "num_input_tokens_seen": 483309624,
      "step": 8060,
      "train_runtime": 103642.4458,
      "train_tokens_per_second": 4663.24
    },
    {
      "epoch": 1.0286989795918366,
      "grad_norm": 0.6272003459089206,
      "learning_rate": 4.775675848247427e-06,
      "loss": 0.24145734310150146,
      "num_input_tokens_seen": 483622136,
      "step": 8065,
      "train_runtime": 103720.4166,
      "train_tokens_per_second": 4662.748
    },
    {
      "epoch": 1.0293367346938775,
      "grad_norm": 0.7643285491618245,
      "learning_rate": 4.770672088434433e-06,
      "loss": 0.26503143310546873,
      "num_input_tokens_seen": 483927584,
      "step": 8070,
      "train_runtime": 103786.938,
      "train_tokens_per_second": 4662.702
    },
    {
      "epoch": 1.0299744897959184,
      "grad_norm": 0.7224211334277504,
      "learning_rate": 4.765668558768015e-06,
      "loss": 0.2647601366043091,
      "num_input_tokens_seen": 484219912,
      "step": 8075,
      "train_runtime": 103846.0738,
      "train_tokens_per_second": 4662.862
    },
    {
      "epoch": 1.030612244897959,
      "grad_norm": 0.6260556021229013,
      "learning_rate": 4.760665264269566e-06,
      "loss": 0.2593989372253418,
      "num_input_tokens_seen": 484523920,
      "step": 8080,
      "train_runtime": 103918.9874,
      "train_tokens_per_second": 4662.516
    },
    {
      "epoch": 1.03125,
      "grad_norm": 0.7384068818094177,
      "learning_rate": 4.755662209960238e-06,
      "loss": 0.2510901927947998,
      "num_input_tokens_seen": 484819144,
      "step": 8085,
      "train_runtime": 103978.2914,
      "train_tokens_per_second": 4662.696
    },
    {
      "epoch": 1.031887755102041,
      "grad_norm": 0.78521447721399,
      "learning_rate": 4.750659400860945e-06,
      "loss": 0.2539271354675293,
      "num_input_tokens_seen": 485118608,
      "step": 8090,
      "train_runtime": 104037.4872,
      "train_tokens_per_second": 4662.921
    },
    {
      "epoch": 1.0325255102040816,
      "grad_norm": 0.6250747330880316,
      "learning_rate": 4.74565684199236e-06,
      "loss": 0.24170682430267335,
      "num_input_tokens_seen": 485436096,
      "step": 8095,
      "train_runtime": 104108.6134,
      "train_tokens_per_second": 4662.785
    },
    {
      "epoch": 1.0331632653061225,
      "grad_norm": 0.6782670204566934,
      "learning_rate": 4.740654538374896e-06,
      "loss": 0.2533915042877197,
      "num_input_tokens_seen": 485738432,
      "step": 8100,
      "train_runtime": 104165.3291,
      "train_tokens_per_second": 4663.149
    },
    {
      "epoch": 1.0338010204081634,
      "grad_norm": 0.593760496801077,
      "learning_rate": 4.735652495028714e-06,
      "loss": 0.24408226013183593,
      "num_input_tokens_seen": 486036120,
      "step": 8105,
      "train_runtime": 104224.6897,
      "train_tokens_per_second": 4663.349
    },
    {
      "epoch": 1.034438775510204,
      "grad_norm": 0.6852036028617559,
      "learning_rate": 4.7306507169737176e-06,
      "loss": 0.24594330787658691,
      "num_input_tokens_seen": 486347296,
      "step": 8110,
      "train_runtime": 104289.0295,
      "train_tokens_per_second": 4663.456
    },
    {
      "epoch": 1.035076530612245,
      "grad_norm": 0.6724049261384363,
      "learning_rate": 4.725649209229537e-06,
      "loss": 0.2450767993927002,
      "num_input_tokens_seen": 486645472,
      "step": 8115,
      "train_runtime": 104347.8074,
      "train_tokens_per_second": 4663.687
    },
    {
      "epoch": 1.0357142857142858,
      "grad_norm": 0.652922414752715,
      "learning_rate": 4.720647976815536e-06,
      "loss": 0.24182796478271484,
      "num_input_tokens_seen": 486943200,
      "step": 8120,
      "train_runtime": 104411.5615,
      "train_tokens_per_second": 4663.69
    },
    {
      "epoch": 1.0363520408163265,
      "grad_norm": 0.6386399166479841,
      "learning_rate": 4.7156470247508e-06,
      "loss": 0.2487572193145752,
      "num_input_tokens_seen": 487234656,
      "step": 8125,
      "train_runtime": 104478.4581,
      "train_tokens_per_second": 4663.494
    },
    {
      "epoch": 1.0369897959183674,
      "grad_norm": 0.6375505378297279,
      "learning_rate": 4.710646358054133e-06,
      "loss": 0.2301931858062744,
      "num_input_tokens_seen": 487533672,
      "step": 8130,
      "train_runtime": 104555.1924,
      "train_tokens_per_second": 4662.931
    },
    {
      "epoch": 1.0376275510204083,
      "grad_norm": 0.690799677250621,
      "learning_rate": 4.705645981744055e-06,
      "loss": 0.2515393257141113,
      "num_input_tokens_seen": 487821880,
      "step": 8135,
      "train_runtime": 104624.4958,
      "train_tokens_per_second": 4662.597
    },
    {
      "epoch": 1.038265306122449,
      "grad_norm": 0.6805220934630078,
      "learning_rate": 4.700645900838793e-06,
      "loss": 0.2581647872924805,
      "num_input_tokens_seen": 488114280,
      "step": 8140,
      "train_runtime": 104693.1979,
      "train_tokens_per_second": 4662.33
    },
    {
      "epoch": 1.0389030612244898,
      "grad_norm": 0.6547597860715865,
      "learning_rate": 4.695646120356275e-06,
      "loss": 0.23448925018310546,
      "num_input_tokens_seen": 488397288,
      "step": 8145,
      "train_runtime": 104767.1561,
      "train_tokens_per_second": 4661.74
    },
    {
      "epoch": 1.0395408163265305,
      "grad_norm": 0.588188314591436,
      "learning_rate": 4.69064664531413e-06,
      "loss": 0.2562718868255615,
      "num_input_tokens_seen": 488707832,
      "step": 8150,
      "train_runtime": 104834.3514,
      "train_tokens_per_second": 4661.715
    },
    {
      "epoch": 1.0401785714285714,
      "grad_norm": 0.6041528110153581,
      "learning_rate": 4.685647480729684e-06,
      "loss": 0.22281389236450194,
      "num_input_tokens_seen": 489010392,
      "step": 8155,
      "train_runtime": 104903.0247,
      "train_tokens_per_second": 4661.547
    },
    {
      "epoch": 1.0408163265306123,
      "grad_norm": 0.7010735358941815,
      "learning_rate": 4.680648631619945e-06,
      "loss": 0.23440265655517578,
      "num_input_tokens_seen": 489313720,
      "step": 8160,
      "train_runtime": 104965.4902,
      "train_tokens_per_second": 4661.663
    },
    {
      "epoch": 1.041454081632653,
      "grad_norm": 0.5494106667712622,
      "learning_rate": 4.675650103001606e-06,
      "loss": 0.24059038162231444,
      "num_input_tokens_seen": 489611376,
      "step": 8165,
      "train_runtime": 105030.9282,
      "train_tokens_per_second": 4661.592
    },
    {
      "epoch": 1.0420918367346939,
      "grad_norm": 0.6547849325586217,
      "learning_rate": 4.670651899891043e-06,
      "loss": 0.26830923557281494,
      "num_input_tokens_seen": 489908088,
      "step": 8170,
      "train_runtime": 105101.5458,
      "train_tokens_per_second": 4661.283
    },
    {
      "epoch": 1.0427295918367347,
      "grad_norm": 0.7478085413499376,
      "learning_rate": 4.6656540273042986e-06,
      "loss": 0.2521498680114746,
      "num_input_tokens_seen": 490205880,
      "step": 8175,
      "train_runtime": 105167.8946,
      "train_tokens_per_second": 4661.174
    },
    {
      "epoch": 1.0433673469387754,
      "grad_norm": 0.6947748976659139,
      "learning_rate": 4.66065649025709e-06,
      "loss": 0.2561060905456543,
      "num_input_tokens_seen": 490510144,
      "step": 8180,
      "train_runtime": 105223.8519,
      "train_tokens_per_second": 4661.587
    },
    {
      "epoch": 1.0440051020408163,
      "grad_norm": 0.6422039834091763,
      "learning_rate": 4.655659293764793e-06,
      "loss": 0.23210339546203612,
      "num_input_tokens_seen": 490820784,
      "step": 8185,
      "train_runtime": 105294.0904,
      "train_tokens_per_second": 4661.428
    },
    {
      "epoch": 1.0446428571428572,
      "grad_norm": 0.5832399499471925,
      "learning_rate": 4.6506624428424405e-06,
      "loss": 0.2376014232635498,
      "num_input_tokens_seen": 491122776,
      "step": 8190,
      "train_runtime": 105363.1273,
      "train_tokens_per_second": 4661.24
    },
    {
      "epoch": 1.0452806122448979,
      "grad_norm": 0.6015679326644382,
      "learning_rate": 4.645665942504728e-06,
      "loss": 0.2396768569946289,
      "num_input_tokens_seen": 491417992,
      "step": 8195,
      "train_runtime": 105417.1187,
      "train_tokens_per_second": 4661.653
    },
    {
      "epoch": 1.0459183673469388,
      "grad_norm": 0.9163963802550906,
      "learning_rate": 4.640669797765987e-06,
      "loss": 0.2639081239700317,
      "num_input_tokens_seen": 491699496,
      "step": 8200,
      "train_runtime": 105476.645,
      "train_tokens_per_second": 4661.691
    },
    {
      "epoch": 1.0465561224489797,
      "grad_norm": 0.6607861493632479,
      "learning_rate": 4.6356740136402e-06,
      "loss": 0.2335533618927002,
      "num_input_tokens_seen": 492000240,
      "step": 8205,
      "train_runtime": 105541.6368,
      "train_tokens_per_second": 4661.67
    },
    {
      "epoch": 1.0471938775510203,
      "grad_norm": 0.6708484584467924,
      "learning_rate": 4.630678595140985e-06,
      "loss": 0.23867573738098144,
      "num_input_tokens_seen": 492284344,
      "step": 8210,
      "train_runtime": 105602.4552,
      "train_tokens_per_second": 4661.675
    },
    {
      "epoch": 1.0478316326530612,
      "grad_norm": 0.7111724217709186,
      "learning_rate": 4.625683547281592e-06,
      "loss": 0.25943105220794677,
      "num_input_tokens_seen": 492579312,
      "step": 8215,
      "train_runtime": 105662.3284,
      "train_tokens_per_second": 4661.825
    },
    {
      "epoch": 1.0484693877551021,
      "grad_norm": 1.6127182936475108,
      "learning_rate": 4.620688875074902e-06,
      "loss": 0.25869834423065186,
      "num_input_tokens_seen": 492867768,
      "step": 8220,
      "train_runtime": 105732.377,
      "train_tokens_per_second": 4661.465
    },
    {
      "epoch": 1.0491071428571428,
      "grad_norm": 0.616028641947979,
      "learning_rate": 4.615694583533418e-06,
      "loss": 0.24853596687316895,
      "num_input_tokens_seen": 493172160,
      "step": 8225,
      "train_runtime": 105806.6879,
      "train_tokens_per_second": 4661.068
    },
    {
      "epoch": 1.0497448979591837,
      "grad_norm": 0.6369125192979328,
      "learning_rate": 4.610700677669259e-06,
      "loss": 0.26491570472717285,
      "num_input_tokens_seen": 493464224,
      "step": 8230,
      "train_runtime": 105879.8915,
      "train_tokens_per_second": 4660.604
    },
    {
      "epoch": 1.0503826530612246,
      "grad_norm": 0.6476716752501955,
      "learning_rate": 4.605707162494157e-06,
      "loss": 0.255249547958374,
      "num_input_tokens_seen": 493752528,
      "step": 8235,
      "train_runtime": 105958.7546,
      "train_tokens_per_second": 4659.856
    },
    {
      "epoch": 1.0510204081632653,
      "grad_norm": 0.6636453758071511,
      "learning_rate": 4.600714043019456e-06,
      "loss": 0.24991931915283203,
      "num_input_tokens_seen": 494066096,
      "step": 8240,
      "train_runtime": 106027.2826,
      "train_tokens_per_second": 4659.802
    },
    {
      "epoch": 1.0516581632653061,
      "grad_norm": 0.632592406944963,
      "learning_rate": 4.5957213242560985e-06,
      "loss": 0.23273146152496338,
      "num_input_tokens_seen": 494363272,
      "step": 8245,
      "train_runtime": 106083.1102,
      "train_tokens_per_second": 4660.151
    },
    {
      "epoch": 1.052295918367347,
      "grad_norm": 0.6183258733887178,
      "learning_rate": 4.5907290112146265e-06,
      "loss": 0.2537068843841553,
      "num_input_tokens_seen": 494667136,
      "step": 8250,
      "train_runtime": 106145.9691,
      "train_tokens_per_second": 4660.254
    },
    {
      "epoch": 1.0529336734693877,
      "grad_norm": 0.6001141758422646,
      "learning_rate": 4.585737108905178e-06,
      "loss": 0.21852438449859618,
      "num_input_tokens_seen": 494969088,
      "step": 8255,
      "train_runtime": 106205.2281,
      "train_tokens_per_second": 4660.496
    },
    {
      "epoch": 1.0535714285714286,
      "grad_norm": 0.6329533428253195,
      "learning_rate": 4.5807456223374695e-06,
      "loss": 0.2435009717941284,
      "num_input_tokens_seen": 495274640,
      "step": 8260,
      "train_runtime": 106258.549,
      "train_tokens_per_second": 4661.033
    },
    {
      "epoch": 1.0542091836734695,
      "grad_norm": 0.6272827069368175,
      "learning_rate": 4.57575455652081e-06,
      "loss": 0.2449747085571289,
      "num_input_tokens_seen": 495584168,
      "step": 8265,
      "train_runtime": 106320.8425,
      "train_tokens_per_second": 4661.214
    },
    {
      "epoch": 1.0548469387755102,
      "grad_norm": 0.6524770493077416,
      "learning_rate": 4.570763916464084e-06,
      "loss": 0.2195876121520996,
      "num_input_tokens_seen": 495871720,
      "step": 8270,
      "train_runtime": 106380.0726,
      "train_tokens_per_second": 4661.322
    },
    {
      "epoch": 1.055484693877551,
      "grad_norm": 0.6366331214503633,
      "learning_rate": 4.565773707175743e-06,
      "loss": 0.2583183526992798,
      "num_input_tokens_seen": 496157304,
      "step": 8275,
      "train_runtime": 106456.4656,
      "train_tokens_per_second": 4660.659
    },
    {
      "epoch": 1.0561224489795917,
      "grad_norm": 0.6630599897765524,
      "learning_rate": 4.560783933663814e-06,
      "loss": 0.25582590103149416,
      "num_input_tokens_seen": 496441344,
      "step": 8280,
      "train_runtime": 106524.8375,
      "train_tokens_per_second": 4660.334
    },
    {
      "epoch": 1.0567602040816326,
      "grad_norm": 0.6226155958471633,
      "learning_rate": 4.5557946009358846e-06,
      "loss": 0.2566941261291504,
      "num_input_tokens_seen": 496736528,
      "step": 8285,
      "train_runtime": 106589.8279,
      "train_tokens_per_second": 4660.262
    },
    {
      "epoch": 1.0573979591836735,
      "grad_norm": 0.6844929404939469,
      "learning_rate": 4.550805713999094e-06,
      "loss": 0.2503832817077637,
      "num_input_tokens_seen": 497042912,
      "step": 8290,
      "train_runtime": 106659.0276,
      "train_tokens_per_second": 4660.111
    },
    {
      "epoch": 1.0580357142857142,
      "grad_norm": 0.7050648304264016,
      "learning_rate": 4.545817277860143e-06,
      "loss": 0.235514497756958,
      "num_input_tokens_seen": 497334944,
      "step": 8295,
      "train_runtime": 106719.115,
      "train_tokens_per_second": 4660.224
    },
    {
      "epoch": 1.058673469387755,
      "grad_norm": 0.6753001483023913,
      "learning_rate": 4.540829297525272e-06,
      "loss": 0.23983364105224608,
      "num_input_tokens_seen": 497616896,
      "step": 8300,
      "train_runtime": 106788.966,
      "train_tokens_per_second": 4659.816
    },
    {
      "epoch": 1.059311224489796,
      "grad_norm": 0.6633887701073338,
      "learning_rate": 4.53584177800027e-06,
      "loss": 0.25391721725463867,
      "num_input_tokens_seen": 497912912,
      "step": 8305,
      "train_runtime": 106851.0031,
      "train_tokens_per_second": 4659.881
    },
    {
      "epoch": 1.0599489795918366,
      "grad_norm": 0.5993826661676184,
      "learning_rate": 4.530854724290461e-06,
      "loss": 0.25507731437683107,
      "num_input_tokens_seen": 498206496,
      "step": 8310,
      "train_runtime": 106919.7773,
      "train_tokens_per_second": 4659.629
    },
    {
      "epoch": 1.0605867346938775,
      "grad_norm": 0.6662739749912538,
      "learning_rate": 4.5258681414007005e-06,
      "loss": 0.2637107133865356,
      "num_input_tokens_seen": 498507960,
      "step": 8315,
      "train_runtime": 106996.7552,
      "train_tokens_per_second": 4659.094
    },
    {
      "epoch": 1.0612244897959184,
      "grad_norm": 0.5679221404656976,
      "learning_rate": 4.5208820343353726e-06,
      "loss": 0.23627259731292724,
      "num_input_tokens_seen": 498809664,
      "step": 8320,
      "train_runtime": 107058.119,
      "train_tokens_per_second": 4659.242
    },
    {
      "epoch": 1.061862244897959,
      "grad_norm": 0.5786166526952957,
      "learning_rate": 4.515896408098387e-06,
      "loss": 0.2349698781967163,
      "num_input_tokens_seen": 499116280,
      "step": 8325,
      "train_runtime": 107119.6699,
      "train_tokens_per_second": 4659.427
    },
    {
      "epoch": 1.0625,
      "grad_norm": 0.608677092154444,
      "learning_rate": 4.510911267693164e-06,
      "loss": 0.2412008285522461,
      "num_input_tokens_seen": 499411432,
      "step": 8330,
      "train_runtime": 107178.991,
      "train_tokens_per_second": 4659.602
    },
    {
      "epoch": 1.063137755102041,
      "grad_norm": 0.6523443635691932,
      "learning_rate": 4.505926618122641e-06,
      "loss": 0.25432002544403076,
      "num_input_tokens_seen": 499716936,
      "step": 8335,
      "train_runtime": 107242.455,
      "train_tokens_per_second": 4659.693
    },
    {
      "epoch": 1.0637755102040816,
      "grad_norm": 0.6656181552793023,
      "learning_rate": 4.5009424643892644e-06,
      "loss": 0.26527795791625974,
      "num_input_tokens_seen": 500018064,
      "step": 8340,
      "train_runtime": 107308.2021,
      "train_tokens_per_second": 4659.644
    },
    {
      "epoch": 1.0644132653061225,
      "grad_norm": 0.6450333545005282,
      "learning_rate": 4.4959588114949785e-06,
      "loss": 0.2552632808685303,
      "num_input_tokens_seen": 500319952,
      "step": 8345,
      "train_runtime": 107373.947,
      "train_tokens_per_second": 4659.603
    },
    {
      "epoch": 1.0650510204081634,
      "grad_norm": 0.666522284543712,
      "learning_rate": 4.490975664441227e-06,
      "loss": 0.2451164960861206,
      "num_input_tokens_seen": 500615320,
      "step": 8350,
      "train_runtime": 107442.4022,
      "train_tokens_per_second": 4659.383
    },
    {
      "epoch": 1.065688775510204,
      "grad_norm": 0.6475147741320086,
      "learning_rate": 4.4859930282289475e-06,
      "loss": 0.24610958099365235,
      "num_input_tokens_seen": 500903224,
      "step": 8355,
      "train_runtime": 107517.8799,
      "train_tokens_per_second": 4658.79
    },
    {
      "epoch": 1.066326530612245,
      "grad_norm": 0.6512810536661453,
      "learning_rate": 4.48101090785856e-06,
      "loss": 0.23372960090637207,
      "num_input_tokens_seen": 501212080,
      "step": 8360,
      "train_runtime": 107574.2447,
      "train_tokens_per_second": 4659.22
    },
    {
      "epoch": 1.0669642857142858,
      "grad_norm": 0.6692692683844598,
      "learning_rate": 4.4760293083299736e-06,
      "loss": 0.2537166833877563,
      "num_input_tokens_seen": 501513856,
      "step": 8365,
      "train_runtime": 107636.0389,
      "train_tokens_per_second": 4659.349
    },
    {
      "epoch": 1.0676020408163265,
      "grad_norm": 0.6802222357751844,
      "learning_rate": 4.471048234642572e-06,
      "loss": 0.27139983177185056,
      "num_input_tokens_seen": 501825848,
      "step": 8370,
      "train_runtime": 107693.3794,
      "train_tokens_per_second": 4659.765
    },
    {
      "epoch": 1.0682397959183674,
      "grad_norm": 0.6141573801798184,
      "learning_rate": 4.466067691795208e-06,
      "loss": 0.24951858520507814,
      "num_input_tokens_seen": 502135968,
      "step": 8375,
      "train_runtime": 107748.9125,
      "train_tokens_per_second": 4660.242
    },
    {
      "epoch": 1.068877551020408,
      "grad_norm": 0.6770204722988951,
      "learning_rate": 4.4610876847862034e-06,
      "loss": 0.26896252632141116,
      "num_input_tokens_seen": 502434496,
      "step": 8380,
      "train_runtime": 107826.2483,
      "train_tokens_per_second": 4659.668
    },
    {
      "epoch": 1.069515306122449,
      "grad_norm": 0.9004020387442226,
      "learning_rate": 4.456108218613346e-06,
      "loss": 0.2850655078887939,
      "num_input_tokens_seen": 502731136,
      "step": 8385,
      "train_runtime": 107897.7267,
      "train_tokens_per_second": 4659.33
    },
    {
      "epoch": 1.0701530612244898,
      "grad_norm": 0.6583034221887463,
      "learning_rate": 4.451129298273877e-06,
      "loss": 0.25475876331329345,
      "num_input_tokens_seen": 503032912,
      "step": 8390,
      "train_runtime": 107949.7713,
      "train_tokens_per_second": 4659.879
    },
    {
      "epoch": 1.0707908163265305,
      "grad_norm": 0.6430763705170768,
      "learning_rate": 4.4461509287644885e-06,
      "loss": 0.24304118156433105,
      "num_input_tokens_seen": 503329176,
      "step": 8395,
      "train_runtime": 108011.0848,
      "train_tokens_per_second": 4659.977
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 0.5660273995465457,
      "learning_rate": 4.441173115081327e-06,
      "loss": 0.24364175796508789,
      "num_input_tokens_seen": 503632680,
      "step": 8400,
      "train_runtime": 108077.8713,
      "train_tokens_per_second": 4659.906
    },
    {
      "epoch": 1.0720663265306123,
      "grad_norm": 0.6107229313949432,
      "learning_rate": 4.43619586221997e-06,
      "loss": 0.25842485427856443,
      "num_input_tokens_seen": 503947216,
      "step": 8405,
      "train_runtime": 108138.3778,
      "train_tokens_per_second": 4660.207
    },
    {
      "epoch": 1.072704081632653,
      "grad_norm": 0.6144189248574193,
      "learning_rate": 4.4312191751754415e-06,
      "loss": 0.22999982833862304,
      "num_input_tokens_seen": 504248392,
      "step": 8410,
      "train_runtime": 108208.4975,
      "train_tokens_per_second": 4659.97
    },
    {
      "epoch": 1.0733418367346939,
      "grad_norm": 0.6687803146210762,
      "learning_rate": 4.426243058942196e-06,
      "loss": 0.2347099542617798,
      "num_input_tokens_seen": 504557584,
      "step": 8415,
      "train_runtime": 108267.1493,
      "train_tokens_per_second": 4660.302
    },
    {
      "epoch": 1.0739795918367347,
      "grad_norm": 0.5889694346744581,
      "learning_rate": 4.42126751851411e-06,
      "loss": 0.22898523807525634,
      "num_input_tokens_seen": 504857424,
      "step": 8420,
      "train_runtime": 108331.0921,
      "train_tokens_per_second": 4660.319
    },
    {
      "epoch": 1.0746173469387754,
      "grad_norm": 0.6695630492744628,
      "learning_rate": 4.4162925588844895e-06,
      "loss": 0.23672561645507811,
      "num_input_tokens_seen": 505160792,
      "step": 8425,
      "train_runtime": 108381.8974,
      "train_tokens_per_second": 4660.933
    },
    {
      "epoch": 1.0752551020408163,
      "grad_norm": 0.6419046672953492,
      "learning_rate": 4.4113181850460504e-06,
      "loss": 0.23820641040802001,
      "num_input_tokens_seen": 505469544,
      "step": 8430,
      "train_runtime": 108461.8041,
      "train_tokens_per_second": 4660.346
    },
    {
      "epoch": 1.0758928571428572,
      "grad_norm": 0.6564311883698931,
      "learning_rate": 4.406344401990927e-06,
      "loss": 0.24784979820251465,
      "num_input_tokens_seen": 505777952,
      "step": 8435,
      "train_runtime": 108526.8087,
      "train_tokens_per_second": 4660.396
    },
    {
      "epoch": 1.0765306122448979,
      "grad_norm": 0.6302951845203199,
      "learning_rate": 4.401371214710658e-06,
      "loss": 0.22909510135650635,
      "num_input_tokens_seen": 506076848,
      "step": 8440,
      "train_runtime": 108587.9026,
      "train_tokens_per_second": 4660.527
    },
    {
      "epoch": 1.0771683673469388,
      "grad_norm": 0.5525748546668868,
      "learning_rate": 4.396398628196181e-06,
      "loss": 0.24703164100646974,
      "num_input_tokens_seen": 506366536,
      "step": 8445,
      "train_runtime": 108664.3182,
      "train_tokens_per_second": 4659.915
    },
    {
      "epoch": 1.0778061224489797,
      "grad_norm": 0.5892407507277255,
      "learning_rate": 4.391426647437836e-06,
      "loss": 0.2493654489517212,
      "num_input_tokens_seen": 506671088,
      "step": 8450,
      "train_runtime": 108725.6082,
      "train_tokens_per_second": 4660.09
    },
    {
      "epoch": 1.0784438775510203,
      "grad_norm": 0.6858416107099327,
      "learning_rate": 4.386455277425355e-06,
      "loss": 0.23719313144683837,
      "num_input_tokens_seen": 506968520,
      "step": 8455,
      "train_runtime": 108795.0967,
      "train_tokens_per_second": 4659.847
    },
    {
      "epoch": 1.0790816326530612,
      "grad_norm": 0.6341771851077849,
      "learning_rate": 4.381484523147852e-06,
      "loss": 0.26437945365905763,
      "num_input_tokens_seen": 507268616,
      "step": 8460,
      "train_runtime": 108860.8914,
      "train_tokens_per_second": 4659.787
    },
    {
      "epoch": 1.0797193877551021,
      "grad_norm": 0.6230770952175335,
      "learning_rate": 4.376514389593827e-06,
      "loss": 0.25137033462524416,
      "num_input_tokens_seen": 507566088,
      "step": 8465,
      "train_runtime": 108928.4265,
      "train_tokens_per_second": 4659.629
    },
    {
      "epoch": 1.0803571428571428,
      "grad_norm": 0.6903878203003238,
      "learning_rate": 4.3715448817511565e-06,
      "loss": 0.2370760917663574,
      "num_input_tokens_seen": 507883104,
      "step": 8470,
      "train_runtime": 109000.8963,
      "train_tokens_per_second": 4659.44
    },
    {
      "epoch": 1.0809948979591837,
      "grad_norm": 0.5973717072526464,
      "learning_rate": 4.3665760046070885e-06,
      "loss": 0.23699584007263183,
      "num_input_tokens_seen": 508188696,
      "step": 8475,
      "train_runtime": 109065.5472,
      "train_tokens_per_second": 4659.48
    },
    {
      "epoch": 1.0816326530612246,
      "grad_norm": 0.5788254959377019,
      "learning_rate": 4.361607763148236e-06,
      "loss": 0.22778549194335937,
      "num_input_tokens_seen": 508495216,
      "step": 8480,
      "train_runtime": 109122.4298,
      "train_tokens_per_second": 4659.86
    },
    {
      "epoch": 1.0822704081632653,
      "grad_norm": 0.6916897496545412,
      "learning_rate": 4.356640162360582e-06,
      "loss": 0.2446908950805664,
      "num_input_tokens_seen": 508782408,
      "step": 8485,
      "train_runtime": 109183.7869,
      "train_tokens_per_second": 4659.871
    },
    {
      "epoch": 1.0829081632653061,
      "grad_norm": 0.5857273313626739,
      "learning_rate": 4.3516732072294545e-06,
      "loss": 0.2512369155883789,
      "num_input_tokens_seen": 509096616,
      "step": 8490,
      "train_runtime": 109240.8955,
      "train_tokens_per_second": 4660.312
    },
    {
      "epoch": 1.083545918367347,
      "grad_norm": 0.6386722073991092,
      "learning_rate": 4.346706902739544e-06,
      "loss": 0.25554609298706055,
      "num_input_tokens_seen": 509395648,
      "step": 8495,
      "train_runtime": 109295.429,
      "train_tokens_per_second": 4660.722
    },
    {
      "epoch": 1.0841836734693877,
      "grad_norm": 0.6861412852364408,
      "learning_rate": 4.3417412538748824e-06,
      "loss": 0.24098649024963378,
      "num_input_tokens_seen": 509699320,
      "step": 8500,
      "train_runtime": 109351.0394,
      "train_tokens_per_second": 4661.129
    },
    {
      "epoch": 1.0848214285714286,
      "grad_norm": 0.6072046279238779,
      "learning_rate": 4.336776265618844e-06,
      "loss": 0.2599900722503662,
      "num_input_tokens_seen": 510003248,
      "step": 8505,
      "train_runtime": 109425.0831,
      "train_tokens_per_second": 4660.753
    },
    {
      "epoch": 1.0854591836734695,
      "grad_norm": 0.8551260945967675,
      "learning_rate": 4.331811942954143e-06,
      "loss": 0.2413041114807129,
      "num_input_tokens_seen": 510278912,
      "step": 8510,
      "train_runtime": 109487.242,
      "train_tokens_per_second": 4660.624
    },
    {
      "epoch": 1.0860969387755102,
      "grad_norm": 0.6659236914285106,
      "learning_rate": 4.3268482908628245e-06,
      "loss": 0.24942457675933838,
      "num_input_tokens_seen": 510580200,
      "step": 8515,
      "train_runtime": 109554.1565,
      "train_tokens_per_second": 4660.528
    },
    {
      "epoch": 1.086734693877551,
      "grad_norm": 0.7739805326763189,
      "learning_rate": 4.321885314326258e-06,
      "loss": 0.2388164758682251,
      "num_input_tokens_seen": 510883392,
      "step": 8520,
      "train_runtime": 109611.0834,
      "train_tokens_per_second": 4660.873
    },
    {
      "epoch": 1.087372448979592,
      "grad_norm": 0.6225166331507451,
      "learning_rate": 4.316923018325137e-06,
      "loss": 0.22586362361907958,
      "num_input_tokens_seen": 511191600,
      "step": 8525,
      "train_runtime": 109666.9653,
      "train_tokens_per_second": 4661.309
    },
    {
      "epoch": 1.0880102040816326,
      "grad_norm": 0.6413617564403306,
      "learning_rate": 4.3119614078394756e-06,
      "loss": 0.23577985763549805,
      "num_input_tokens_seen": 511484384,
      "step": 8530,
      "train_runtime": 109728.3426,
      "train_tokens_per_second": 4661.37
    },
    {
      "epoch": 1.0886479591836735,
      "grad_norm": 0.6235517419524493,
      "learning_rate": 4.3070004878485925e-06,
      "loss": 0.24158282279968263,
      "num_input_tokens_seen": 511784280,
      "step": 8535,
      "train_runtime": 109799.3159,
      "train_tokens_per_second": 4661.088
    },
    {
      "epoch": 1.0892857142857142,
      "grad_norm": 0.5865082138653116,
      "learning_rate": 4.30204026333112e-06,
      "loss": 0.2610388994216919,
      "num_input_tokens_seen": 512085792,
      "step": 8540,
      "train_runtime": 109858.7028,
      "train_tokens_per_second": 4661.313
    },
    {
      "epoch": 1.089923469387755,
      "grad_norm": 0.6053085239154334,
      "learning_rate": 4.297080739264987e-06,
      "loss": 0.2202895164489746,
      "num_input_tokens_seen": 512393920,
      "step": 8545,
      "train_runtime": 109920.9462,
      "train_tokens_per_second": 4661.477
    },
    {
      "epoch": 1.090561224489796,
      "grad_norm": 0.7097943296568763,
      "learning_rate": 4.292121920627423e-06,
      "loss": 0.25112998485565186,
      "num_input_tokens_seen": 512688592,
      "step": 8550,
      "train_runtime": 109996.2003,
      "train_tokens_per_second": 4660.966
    },
    {
      "epoch": 1.0911989795918366,
      "grad_norm": 0.6153079940819608,
      "learning_rate": 4.287163812394952e-06,
      "loss": 0.22177939414978026,
      "num_input_tokens_seen": 512992664,
      "step": 8555,
      "train_runtime": 110064.6268,
      "train_tokens_per_second": 4660.831
    },
    {
      "epoch": 1.0918367346938775,
      "grad_norm": 0.6422902413039913,
      "learning_rate": 4.282206419543379e-06,
      "loss": 0.24580161571502684,
      "num_input_tokens_seen": 513292160,
      "step": 8560,
      "train_runtime": 110142.0474,
      "train_tokens_per_second": 4660.274
    },
    {
      "epoch": 1.0924744897959184,
      "grad_norm": 0.784318111678597,
      "learning_rate": 4.2772497470477934e-06,
      "loss": 0.2570383071899414,
      "num_input_tokens_seen": 513580216,
      "step": 8565,
      "train_runtime": 110201.4713,
      "train_tokens_per_second": 4660.375
    },
    {
      "epoch": 1.093112244897959,
      "grad_norm": 0.57623539824992,
      "learning_rate": 4.272293799882565e-06,
      "loss": 0.2587752342224121,
      "num_input_tokens_seen": 513887568,
      "step": 8570,
      "train_runtime": 110261.4572,
      "train_tokens_per_second": 4660.627
    },
    {
      "epoch": 1.09375,
      "grad_norm": 0.6942732863649478,
      "learning_rate": 4.267338583021331e-06,
      "loss": 0.24844789505004883,
      "num_input_tokens_seen": 514190592,
      "step": 8575,
      "train_runtime": 110327.7734,
      "train_tokens_per_second": 4660.573
    },
    {
      "epoch": 1.094387755102041,
      "grad_norm": 0.6538682735284778,
      "learning_rate": 4.262384101436997e-06,
      "loss": 0.24650192260742188,
      "num_input_tokens_seen": 514502072,
      "step": 8580,
      "train_runtime": 110385.3433,
      "train_tokens_per_second": 4660.964
    },
    {
      "epoch": 1.0950255102040816,
      "grad_norm": 0.6069131663138339,
      "learning_rate": 4.257430360101734e-06,
      "loss": 0.2545419454574585,
      "num_input_tokens_seen": 514818112,
      "step": 8585,
      "train_runtime": 110457.489,
      "train_tokens_per_second": 4660.781
    },
    {
      "epoch": 1.0956632653061225,
      "grad_norm": 0.5788914544539152,
      "learning_rate": 4.252477363986963e-06,
      "loss": 0.2407358169555664,
      "num_input_tokens_seen": 515124544,
      "step": 8590,
      "train_runtime": 110529.3895,
      "train_tokens_per_second": 4660.521
    },
    {
      "epoch": 1.0963010204081634,
      "grad_norm": 0.8910302050946797,
      "learning_rate": 4.247525118063366e-06,
      "loss": 0.24895682334899902,
      "num_input_tokens_seen": 515425744,
      "step": 8595,
      "train_runtime": 110593.7547,
      "train_tokens_per_second": 4660.532
    },
    {
      "epoch": 1.096938775510204,
      "grad_norm": 0.6594975719339832,
      "learning_rate": 4.242573627300866e-06,
      "loss": 0.2434098243713379,
      "num_input_tokens_seen": 515739952,
      "step": 8600,
      "train_runtime": 110652.9106,
      "train_tokens_per_second": 4660.88
    },
    {
      "epoch": 1.097576530612245,
      "grad_norm": 0.7240782170388725,
      "learning_rate": 4.237622896668628e-06,
      "loss": 0.23852176666259767,
      "num_input_tokens_seen": 516034008,
      "step": 8605,
      "train_runtime": 110709.3234,
      "train_tokens_per_second": 4661.161
    },
    {
      "epoch": 1.0982142857142858,
      "grad_norm": 0.6637953558110109,
      "learning_rate": 4.232672931135058e-06,
      "loss": 0.24601476192474364,
      "num_input_tokens_seen": 516328520,
      "step": 8610,
      "train_runtime": 110768.3536,
      "train_tokens_per_second": 4661.336
    },
    {
      "epoch": 1.0988520408163265,
      "grad_norm": 0.6789934166654039,
      "learning_rate": 4.227723735667793e-06,
      "loss": 0.2546565532684326,
      "num_input_tokens_seen": 516627344,
      "step": 8615,
      "train_runtime": 110834.7893,
      "train_tokens_per_second": 4661.238
    },
    {
      "epoch": 1.0994897959183674,
      "grad_norm": 0.6743180358260381,
      "learning_rate": 4.222775315233693e-06,
      "loss": 0.2450329303741455,
      "num_input_tokens_seen": 516926536,
      "step": 8620,
      "train_runtime": 110891.3765,
      "train_tokens_per_second": 4661.558
    },
    {
      "epoch": 1.100127551020408,
      "grad_norm": 0.6819867044598046,
      "learning_rate": 4.217827674798845e-06,
      "loss": 0.2333766460418701,
      "num_input_tokens_seen": 517208032,
      "step": 8625,
      "train_runtime": 110960.744,
      "train_tokens_per_second": 4661.18
    },
    {
      "epoch": 1.100765306122449,
      "grad_norm": 0.661119390373851,
      "learning_rate": 4.2128808193285555e-06,
      "loss": 0.25281786918640137,
      "num_input_tokens_seen": 517511480,
      "step": 8630,
      "train_runtime": 111018.3836,
      "train_tokens_per_second": 4661.494
    },
    {
      "epoch": 1.1014030612244898,
      "grad_norm": 0.6292791859153245,
      "learning_rate": 4.2079347537873325e-06,
      "loss": 0.25015289783477784,
      "num_input_tokens_seen": 517806768,
      "step": 8635,
      "train_runtime": 111074.1525,
      "train_tokens_per_second": 4661.812
    },
    {
      "epoch": 1.1020408163265305,
      "grad_norm": 0.560281951568524,
      "learning_rate": 4.2029894831389036e-06,
      "loss": 0.23009247779846193,
      "num_input_tokens_seen": 518107424,
      "step": 8640,
      "train_runtime": 111145.3314,
      "train_tokens_per_second": 4661.531
    },
    {
      "epoch": 1.1026785714285714,
      "grad_norm": 0.6251386010444134,
      "learning_rate": 4.198045012346192e-06,
      "loss": 0.24771604537963868,
      "num_input_tokens_seen": 518414256,
      "step": 8645,
      "train_runtime": 111210.8146,
      "train_tokens_per_second": 4661.545
    },
    {
      "epoch": 1.1033163265306123,
      "grad_norm": 0.7705113770677164,
      "learning_rate": 4.1931013463713165e-06,
      "loss": 0.259547758102417,
      "num_input_tokens_seen": 518702976,
      "step": 8650,
      "train_runtime": 111275.476,
      "train_tokens_per_second": 4661.431
    },
    {
      "epoch": 1.103954081632653,
      "grad_norm": 0.7054425512566621,
      "learning_rate": 4.188158490175596e-06,
      "loss": 0.22324018478393554,
      "num_input_tokens_seen": 519006576,
      "step": 8655,
      "train_runtime": 111332.1181,
      "train_tokens_per_second": 4661.787
    },
    {
      "epoch": 1.1045918367346939,
      "grad_norm": 0.6464946280216061,
      "learning_rate": 4.1832164487195295e-06,
      "loss": 0.26404922008514403,
      "num_input_tokens_seen": 519322336,
      "step": 8660,
      "train_runtime": 111406.9787,
      "train_tokens_per_second": 4661.488
    },
    {
      "epoch": 1.1052295918367347,
      "grad_norm": 0.635481471424389,
      "learning_rate": 4.1782752269627995e-06,
      "loss": 0.23730854988098143,
      "num_input_tokens_seen": 519620848,
      "step": 8665,
      "train_runtime": 111476.8516,
      "train_tokens_per_second": 4661.244
    },
    {
      "epoch": 1.1058673469387754,
      "grad_norm": 0.7693134141834581,
      "learning_rate": 4.17333482986427e-06,
      "loss": 0.2592004776000977,
      "num_input_tokens_seen": 519931392,
      "step": 8670,
      "train_runtime": 111549.1729,
      "train_tokens_per_second": 4661.006
    },
    {
      "epoch": 1.1065051020408163,
      "grad_norm": 0.5766353405067196,
      "learning_rate": 4.16839526238197e-06,
      "loss": 0.22901449203491211,
      "num_input_tokens_seen": 520241504,
      "step": 8675,
      "train_runtime": 111604.5074,
      "train_tokens_per_second": 4661.474
    },
    {
      "epoch": 1.1071428571428572,
      "grad_norm": 0.639494213084661,
      "learning_rate": 4.1634565294731045e-06,
      "loss": 0.25211329460144044,
      "num_input_tokens_seen": 520539736,
      "step": 8680,
      "train_runtime": 111660.078,
      "train_tokens_per_second": 4661.825
    },
    {
      "epoch": 1.1077806122448979,
      "grad_norm": 0.5700841107449708,
      "learning_rate": 4.158518636094037e-06,
      "loss": 0.24518826007843017,
      "num_input_tokens_seen": 520837320,
      "step": 8685,
      "train_runtime": 111716.1387,
      "train_tokens_per_second": 4662.149
    },
    {
      "epoch": 1.1084183673469388,
      "grad_norm": 0.5718162723588303,
      "learning_rate": 4.153581587200287e-06,
      "loss": 0.2576592922210693,
      "num_input_tokens_seen": 521140296,
      "step": 8690,
      "train_runtime": 111769.3873,
      "train_tokens_per_second": 4662.639
    },
    {
      "epoch": 1.1090561224489797,
      "grad_norm": 0.6218159544464157,
      "learning_rate": 4.148645387746526e-06,
      "loss": 0.2505445718765259,
      "num_input_tokens_seen": 521443008,
      "step": 8695,
      "train_runtime": 111821.7977,
      "train_tokens_per_second": 4663.161
    },
    {
      "epoch": 1.1096938775510203,
      "grad_norm": 0.6813986888146172,
      "learning_rate": 4.143710042686579e-06,
      "loss": 0.2475372314453125,
      "num_input_tokens_seen": 521750504,
      "step": 8700,
      "train_runtime": 111878.6859,
      "train_tokens_per_second": 4663.538
    },
    {
      "epoch": 1.1103316326530612,
      "grad_norm": 0.6058288226100228,
      "learning_rate": 4.138775556973406e-06,
      "loss": 0.26158604621887205,
      "num_input_tokens_seen": 522030912,
      "step": 8705,
      "train_runtime": 111942.0057,
      "train_tokens_per_second": 4663.405
    },
    {
      "epoch": 1.1109693877551021,
      "grad_norm": 0.6904469086085062,
      "learning_rate": 4.133841935559109e-06,
      "loss": 0.26089537143707275,
      "num_input_tokens_seen": 522323992,
      "step": 8710,
      "train_runtime": 112004.2566,
      "train_tokens_per_second": 4663.43
    },
    {
      "epoch": 1.1116071428571428,
      "grad_norm": 0.6707493974163103,
      "learning_rate": 4.1289091833949235e-06,
      "loss": 0.25034823417663576,
      "num_input_tokens_seen": 522620576,
      "step": 8715,
      "train_runtime": 112066.0083,
      "train_tokens_per_second": 4663.507
    },
    {
      "epoch": 1.1122448979591837,
      "grad_norm": 0.6948665945679124,
      "learning_rate": 4.123977305431209e-06,
      "loss": 0.22675719261169433,
      "num_input_tokens_seen": 522909792,
      "step": 8720,
      "train_runtime": 112138.3084,
      "train_tokens_per_second": 4663.079
    },
    {
      "epoch": 1.1128826530612246,
      "grad_norm": 0.653997402260834,
      "learning_rate": 4.119046306617449e-06,
      "loss": 0.23716330528259277,
      "num_input_tokens_seen": 523206376,
      "step": 8725,
      "train_runtime": 112199.7623,
      "train_tokens_per_second": 4663.168
    },
    {
      "epoch": 1.1135204081632653,
      "grad_norm": 0.6347530765050445,
      "learning_rate": 4.114116191902248e-06,
      "loss": 0.24915711879730223,
      "num_input_tokens_seen": 523516496,
      "step": 8730,
      "train_runtime": 112261.3015,
      "train_tokens_per_second": 4663.375
    },
    {
      "epoch": 1.1141581632653061,
      "grad_norm": 0.699603284516342,
      "learning_rate": 4.109186966233315e-06,
      "loss": 0.2517080545425415,
      "num_input_tokens_seen": 523824840,
      "step": 8735,
      "train_runtime": 112330.128,
      "train_tokens_per_second": 4663.262
    },
    {
      "epoch": 1.114795918367347,
      "grad_norm": 0.6011570474960909,
      "learning_rate": 4.104258634557478e-06,
      "loss": 0.24463653564453125,
      "num_input_tokens_seen": 524126296,
      "step": 8740,
      "train_runtime": 112390.756,
      "train_tokens_per_second": 4663.429
    },
    {
      "epoch": 1.1154336734693877,
      "grad_norm": 0.8184521177672255,
      "learning_rate": 4.09933120182066e-06,
      "loss": 0.23126814365386963,
      "num_input_tokens_seen": 524421360,
      "step": 8745,
      "train_runtime": 112462.0875,
      "train_tokens_per_second": 4663.095
    },
    {
      "epoch": 1.1160714285714286,
      "grad_norm": 0.7711043543729131,
      "learning_rate": 4.094404672967882e-06,
      "loss": 0.23586022853851318,
      "num_input_tokens_seen": 524712216,
      "step": 8750,
      "train_runtime": 112531.4826,
      "train_tokens_per_second": 4662.804
    },
    {
      "epoch": 1.1167091836734695,
      "grad_norm": 0.569997286888262,
      "learning_rate": 4.089479052943261e-06,
      "loss": 0.23141369819641114,
      "num_input_tokens_seen": 525016712,
      "step": 8755,
      "train_runtime": 112589.8613,
      "train_tokens_per_second": 4663.09
    },
    {
      "epoch": 1.1173469387755102,
      "grad_norm": 0.6029928288031483,
      "learning_rate": 4.084554346690003e-06,
      "loss": 0.25970346927642823,
      "num_input_tokens_seen": 525305040,
      "step": 8760,
      "train_runtime": 112656.111,
      "train_tokens_per_second": 4662.908
    },
    {
      "epoch": 1.117984693877551,
      "grad_norm": 0.6589543488231843,
      "learning_rate": 4.079630559150392e-06,
      "loss": 0.24691288471221923,
      "num_input_tokens_seen": 525586784,
      "step": 8765,
      "train_runtime": 112723.6392,
      "train_tokens_per_second": 4662.614
    },
    {
      "epoch": 1.118622448979592,
      "grad_norm": 0.61521781871156,
      "learning_rate": 4.074707695265792e-06,
      "loss": 0.2508359909057617,
      "num_input_tokens_seen": 525891264,
      "step": 8770,
      "train_runtime": 112787.6434,
      "train_tokens_per_second": 4662.667
    },
    {
      "epoch": 1.1192602040816326,
      "grad_norm": 0.6420620386451169,
      "learning_rate": 4.069785759976645e-06,
      "loss": 0.24450926780700682,
      "num_input_tokens_seen": 526189544,
      "step": 8775,
      "train_runtime": 112843.4565,
      "train_tokens_per_second": 4663.004
    },
    {
      "epoch": 1.1198979591836735,
      "grad_norm": 0.841776634817563,
      "learning_rate": 4.0648647582224504e-06,
      "loss": 0.2408369541168213,
      "num_input_tokens_seen": 526488680,
      "step": 8780,
      "train_runtime": 112902.7282,
      "train_tokens_per_second": 4663.206
    },
    {
      "epoch": 1.1205357142857142,
      "grad_norm": 0.6045633595489134,
      "learning_rate": 4.059944694941783e-06,
      "loss": 0.22510142326354982,
      "num_input_tokens_seen": 526800512,
      "step": 8785,
      "train_runtime": 112958.7177,
      "train_tokens_per_second": 4663.655
    },
    {
      "epoch": 1.121173469387755,
      "grad_norm": 0.6776121769528503,
      "learning_rate": 4.0550255750722674e-06,
      "loss": 0.24085683822631837,
      "num_input_tokens_seen": 527107496,
      "step": 8790,
      "train_runtime": 113032.7839,
      "train_tokens_per_second": 4663.315
    },
    {
      "epoch": 1.121811224489796,
      "grad_norm": 0.6049770917730501,
      "learning_rate": 4.050107403550582e-06,
      "loss": 0.2417008876800537,
      "num_input_tokens_seen": 527399464,
      "step": 8795,
      "train_runtime": 113098.9307,
      "train_tokens_per_second": 4663.169
    },
    {
      "epoch": 1.1224489795918366,
      "grad_norm": 0.6695859870368668,
      "learning_rate": 4.0451901853124596e-06,
      "loss": 0.24154107570648192,
      "num_input_tokens_seen": 527709368,
      "step": 8800,
      "train_runtime": 113158.2587,
      "train_tokens_per_second": 4663.463
    },
    {
      "epoch": 1.1230867346938775,
      "grad_norm": 0.6085742985049531,
      "learning_rate": 4.04027392529267e-06,
      "loss": 0.2165979862213135,
      "num_input_tokens_seen": 528015248,
      "step": 8805,
      "train_runtime": 113222.2721,
      "train_tokens_per_second": 4663.528
    },
    {
      "epoch": 1.1237244897959184,
      "grad_norm": 0.7926317096607808,
      "learning_rate": 4.035358628425022e-06,
      "loss": 0.2610020637512207,
      "num_input_tokens_seen": 528302576,
      "step": 8810,
      "train_runtime": 113289.0832,
      "train_tokens_per_second": 4663.314
    },
    {
      "epoch": 1.124362244897959,
      "grad_norm": 0.6610127980447489,
      "learning_rate": 4.030444299642363e-06,
      "loss": 0.2666980743408203,
      "num_input_tokens_seen": 528610872,
      "step": 8815,
      "train_runtime": 113353.9008,
      "train_tokens_per_second": 4663.367
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.6486416535490288,
      "learning_rate": 4.02553094387656e-06,
      "loss": 0.27113239765167235,
      "num_input_tokens_seen": 528897960,
      "step": 8820,
      "train_runtime": 113414.262,
      "train_tokens_per_second": 4663.417
    },
    {
      "epoch": 1.125637755102041,
      "grad_norm": 0.6527299179917347,
      "learning_rate": 4.020618566058514e-06,
      "loss": 0.23246688842773439,
      "num_input_tokens_seen": 529217848,
      "step": 8825,
      "train_runtime": 113477.6125,
      "train_tokens_per_second": 4663.632
    },
    {
      "epoch": 1.1262755102040816,
      "grad_norm": 0.6086829460001116,
      "learning_rate": 4.015707171118136e-06,
      "loss": 0.24707884788513185,
      "num_input_tokens_seen": 529527624,
      "step": 8830,
      "train_runtime": 113545.3715,
      "train_tokens_per_second": 4663.577
    },
    {
      "epoch": 1.1269132653061225,
      "grad_norm": 0.6135742903501101,
      "learning_rate": 4.010796763984355e-06,
      "loss": 0.2489476203918457,
      "num_input_tokens_seen": 529838424,
      "step": 8835,
      "train_runtime": 113616.4467,
      "train_tokens_per_second": 4663.395
    },
    {
      "epoch": 1.1275510204081634,
      "grad_norm": 0.6258624758361515,
      "learning_rate": 4.005887349585106e-06,
      "loss": 0.24770126342773438,
      "num_input_tokens_seen": 530131984,
      "step": 8840,
      "train_runtime": 113690.7991,
      "train_tokens_per_second": 4662.928
    },
    {
      "epoch": 1.128188775510204,
      "grad_norm": 0.6462539060419881,
      "learning_rate": 4.000978932847332e-06,
      "loss": 0.2586657047271729,
      "num_input_tokens_seen": 530430848,
      "step": 8845,
      "train_runtime": 113753.3411,
      "train_tokens_per_second": 4662.991
    },
    {
      "epoch": 1.128826530612245,
      "grad_norm": 0.560858907102757,
      "learning_rate": 3.996071518696969e-06,
      "loss": 0.24396800994873047,
      "num_input_tokens_seen": 530739048,
      "step": 8850,
      "train_runtime": 113807.1722,
      "train_tokens_per_second": 4663.494
    },
    {
      "epoch": 1.1294642857142858,
      "grad_norm": 0.678900867275871,
      "learning_rate": 3.99116511205895e-06,
      "loss": 0.2503016471862793,
      "num_input_tokens_seen": 531036064,
      "step": 8855,
      "train_runtime": 113864.5592,
      "train_tokens_per_second": 4663.752
    },
    {
      "epoch": 1.1301020408163265,
      "grad_norm": 0.6501528824998741,
      "learning_rate": 3.986259717857201e-06,
      "loss": 0.2325228452682495,
      "num_input_tokens_seen": 531345072,
      "step": 8860,
      "train_runtime": 113926.8569,
      "train_tokens_per_second": 4663.914
    },
    {
      "epoch": 1.1307397959183674,
      "grad_norm": 0.6719603416085055,
      "learning_rate": 3.9813553410146225e-06,
      "loss": 0.2344541072845459,
      "num_input_tokens_seen": 531640184,
      "step": 8865,
      "train_runtime": 113993.78,
      "train_tokens_per_second": 4663.765
    },
    {
      "epoch": 1.131377551020408,
      "grad_norm": 0.5983287801634353,
      "learning_rate": 3.9764519864531026e-06,
      "loss": 0.24439842700958253,
      "num_input_tokens_seen": 531936528,
      "step": 8870,
      "train_runtime": 114068.0639,
      "train_tokens_per_second": 4663.326
    },
    {
      "epoch": 1.132015306122449,
      "grad_norm": 0.5993162683759906,
      "learning_rate": 3.971549659093499e-06,
      "loss": 0.23204846382141114,
      "num_input_tokens_seen": 532227288,
      "step": 8875,
      "train_runtime": 114132.6913,
      "train_tokens_per_second": 4663.233
    },
    {
      "epoch": 1.1326530612244898,
      "grad_norm": 0.5948447703546732,
      "learning_rate": 3.966648363855637e-06,
      "loss": 0.2498317003250122,
      "num_input_tokens_seen": 532531048,
      "step": 8880,
      "train_runtime": 114189.1418,
      "train_tokens_per_second": 4663.587
    },
    {
      "epoch": 1.1332908163265305,
      "grad_norm": 0.6109721055890992,
      "learning_rate": 3.961748105658312e-06,
      "loss": 0.24610438346862792,
      "num_input_tokens_seen": 532823672,
      "step": 8885,
      "train_runtime": 114253.6052,
      "train_tokens_per_second": 4663.517
    },
    {
      "epoch": 1.1339285714285714,
      "grad_norm": 0.5863524541777505,
      "learning_rate": 3.956848889419276e-06,
      "loss": 0.25118041038513184,
      "num_input_tokens_seen": 533133984,
      "step": 8890,
      "train_runtime": 114322.7721,
      "train_tokens_per_second": 4663.41
    },
    {
      "epoch": 1.1345663265306123,
      "grad_norm": 0.6698500360829401,
      "learning_rate": 3.951950720055231e-06,
      "loss": 0.2555887222290039,
      "num_input_tokens_seen": 533425512,
      "step": 8895,
      "train_runtime": 114378.8561,
      "train_tokens_per_second": 4663.672
    },
    {
      "epoch": 1.135204081632653,
      "grad_norm": 0.7470990401150549,
      "learning_rate": 3.947053602481834e-06,
      "loss": 0.24155745506286622,
      "num_input_tokens_seen": 533722200,
      "step": 8900,
      "train_runtime": 114430.3179,
      "train_tokens_per_second": 4664.168
    },
    {
      "epoch": 1.1358418367346939,
      "grad_norm": 0.5784733966288881,
      "learning_rate": 3.9421575416136866e-06,
      "loss": 0.251849889755249,
      "num_input_tokens_seen": 534014344,
      "step": 8905,
      "train_runtime": 114491.7871,
      "train_tokens_per_second": 4664.215
    },
    {
      "epoch": 1.1364795918367347,
      "grad_norm": 0.6426342730063402,
      "learning_rate": 3.937262542364324e-06,
      "loss": 0.24389877319335937,
      "num_input_tokens_seen": 534309280,
      "step": 8910,
      "train_runtime": 114556.6863,
      "train_tokens_per_second": 4664.147
    },
    {
      "epoch": 1.1371173469387754,
      "grad_norm": 0.6184032214831218,
      "learning_rate": 3.932368609646223e-06,
      "loss": 0.23805222511291504,
      "num_input_tokens_seen": 534611264,
      "step": 8915,
      "train_runtime": 114623.8185,
      "train_tokens_per_second": 4664.05
    },
    {
      "epoch": 1.1377551020408163,
      "grad_norm": 0.6955281150748618,
      "learning_rate": 3.927475748370784e-06,
      "loss": 0.24289226531982422,
      "num_input_tokens_seen": 534911912,
      "step": 8920,
      "train_runtime": 114694.9732,
      "train_tokens_per_second": 4663.778
    },
    {
      "epoch": 1.1383928571428572,
      "grad_norm": 0.6444689763555554,
      "learning_rate": 3.922583963448336e-06,
      "loss": 0.2645781755447388,
      "num_input_tokens_seen": 535221712,
      "step": 8925,
      "train_runtime": 114754.4892,
      "train_tokens_per_second": 4664.059
    },
    {
      "epoch": 1.1390306122448979,
      "grad_norm": 0.7044417930873609,
      "learning_rate": 3.9176932597881286e-06,
      "loss": 0.26086721420288084,
      "num_input_tokens_seen": 535523032,
      "step": 8930,
      "train_runtime": 114819.8863,
      "train_tokens_per_second": 4664.027
    },
    {
      "epoch": 1.1396683673469388,
      "grad_norm": 0.7011005108852735,
      "learning_rate": 3.912803642298323e-06,
      "loss": 0.23372416496276854,
      "num_input_tokens_seen": 535814944,
      "step": 8935,
      "train_runtime": 114878.2483,
      "train_tokens_per_second": 4664.198
    },
    {
      "epoch": 1.1403061224489797,
      "grad_norm": 0.6347562665809505,
      "learning_rate": 3.90791511588599e-06,
      "loss": 0.25880999565124513,
      "num_input_tokens_seen": 536124032,
      "step": 8940,
      "train_runtime": 114946.6072,
      "train_tokens_per_second": 4664.114
    },
    {
      "epoch": 1.1409438775510203,
      "grad_norm": 0.6599876980833477,
      "learning_rate": 3.903027685457112e-06,
      "loss": 0.2445706844329834,
      "num_input_tokens_seen": 536418776,
      "step": 8945,
      "train_runtime": 115001.4023,
      "train_tokens_per_second": 4664.454
    },
    {
      "epoch": 1.1415816326530612,
      "grad_norm": 0.62710591246252,
      "learning_rate": 3.898141355916563e-06,
      "loss": 0.24156970977783204,
      "num_input_tokens_seen": 536713192,
      "step": 8950,
      "train_runtime": 115071.1806,
      "train_tokens_per_second": 4664.184
    },
    {
      "epoch": 1.1422193877551021,
      "grad_norm": 0.5594054910943858,
      "learning_rate": 3.893256132168117e-06,
      "loss": 0.22762279510498046,
      "num_input_tokens_seen": 537018000,
      "step": 8955,
      "train_runtime": 115137.7132,
      "train_tokens_per_second": 4664.136
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.6018653801349356,
      "learning_rate": 3.888372019114438e-06,
      "loss": 0.22532968521118163,
      "num_input_tokens_seen": 537291952,
      "step": 8960,
      "train_runtime": 115208.1882,
      "train_tokens_per_second": 4663.661
    },
    {
      "epoch": 1.1434948979591837,
      "grad_norm": 0.6906651665295294,
      "learning_rate": 3.8834890216570715e-06,
      "loss": 0.2603922367095947,
      "num_input_tokens_seen": 537589832,
      "step": 8965,
      "train_runtime": 115273.4421,
      "train_tokens_per_second": 4663.605
    },
    {
      "epoch": 1.1441326530612246,
      "grad_norm": 0.6358963163709775,
      "learning_rate": 3.878607144696449e-06,
      "loss": 0.24374897480010987,
      "num_input_tokens_seen": 537890624,
      "step": 8970,
      "train_runtime": 115346.5018,
      "train_tokens_per_second": 4663.259
    },
    {
      "epoch": 1.1447704081632653,
      "grad_norm": 0.6534673080564684,
      "learning_rate": 3.873726393131877e-06,
      "loss": 0.23672876358032227,
      "num_input_tokens_seen": 538188488,
      "step": 8975,
      "train_runtime": 115414.0543,
      "train_tokens_per_second": 4663.11
    },
    {
      "epoch": 1.1454081632653061,
      "grad_norm": 0.6379043559718519,
      "learning_rate": 3.868846771861528e-06,
      "loss": 0.2526662349700928,
      "num_input_tokens_seen": 538473736,
      "step": 8980,
      "train_runtime": 115472.0993,
      "train_tokens_per_second": 4663.237
    },
    {
      "epoch": 1.146045918367347,
      "grad_norm": 0.587786521523755,
      "learning_rate": 3.86396828578244e-06,
      "loss": 0.24660902023315429,
      "num_input_tokens_seen": 538782264,
      "step": 8985,
      "train_runtime": 115520.9504,
      "train_tokens_per_second": 4663.936
    },
    {
      "epoch": 1.1466836734693877,
      "grad_norm": 0.6518577329012045,
      "learning_rate": 3.859090939790521e-06,
      "loss": 0.23081755638122559,
      "num_input_tokens_seen": 539065056,
      "step": 8990,
      "train_runtime": 115588.1693,
      "train_tokens_per_second": 4663.67
    },
    {
      "epoch": 1.1473214285714286,
      "grad_norm": 0.6217135552266576,
      "learning_rate": 3.854214738780523e-06,
      "loss": 0.24677810668945313,
      "num_input_tokens_seen": 539369512,
      "step": 8995,
      "train_runtime": 115655.3494,
      "train_tokens_per_second": 4663.593
    },
    {
      "epoch": 1.1479591836734695,
      "grad_norm": 0.7097649318957695,
      "learning_rate": 3.849339687646055e-06,
      "loss": 0.27447986602783203,
      "num_input_tokens_seen": 539667064,
      "step": 9000,
      "train_runtime": 115716.9333,
      "train_tokens_per_second": 4663.683
    },
    {
      "epoch": 1.1485969387755102,
      "grad_norm": 0.7626611740955035,
      "learning_rate": 3.844465791279575e-06,
      "loss": 0.2431164026260376,
      "num_input_tokens_seen": 539955840,
      "step": 9005,
      "train_runtime": 115771.4757,
      "train_tokens_per_second": 4663.98
    },
    {
      "epoch": 1.149234693877551,
      "grad_norm": 0.635715227930441,
      "learning_rate": 3.839593054572374e-06,
      "loss": 0.2344665050506592,
      "num_input_tokens_seen": 540259608,
      "step": 9010,
      "train_runtime": 115828.3342,
      "train_tokens_per_second": 4664.313
    },
    {
      "epoch": 1.149872448979592,
      "grad_norm": 0.6741538632282347,
      "learning_rate": 3.834721482414586e-06,
      "loss": 0.24207234382629395,
      "num_input_tokens_seen": 540556576,
      "step": 9015,
      "train_runtime": 115892.5665,
      "train_tokens_per_second": 4664.29
    },
    {
      "epoch": 1.1505102040816326,
      "grad_norm": 0.6822093729956156,
      "learning_rate": 3.829851079695174e-06,
      "loss": 0.2660356044769287,
      "num_input_tokens_seen": 540862224,
      "step": 9020,
      "train_runtime": 115963.4675,
      "train_tokens_per_second": 4664.074
    },
    {
      "epoch": 1.1511479591836735,
      "grad_norm": 0.5855754212435335,
      "learning_rate": 3.824981851301924e-06,
      "loss": 0.2310089111328125,
      "num_input_tokens_seen": 541175464,
      "step": 9025,
      "train_runtime": 116025.638,
      "train_tokens_per_second": 4664.275
    },
    {
      "epoch": 1.1517857142857142,
      "grad_norm": 0.7456571687236758,
      "learning_rate": 3.820113802121453e-06,
      "loss": 0.23663051128387452,
      "num_input_tokens_seen": 541462456,
      "step": 9030,
      "train_runtime": 116100.1959,
      "train_tokens_per_second": 4663.751
    },
    {
      "epoch": 1.152423469387755,
      "grad_norm": 0.5594020174254829,
      "learning_rate": 3.815246937039182e-06,
      "loss": 0.22132573127746583,
      "num_input_tokens_seen": 541753776,
      "step": 9035,
      "train_runtime": 116166.2489,
      "train_tokens_per_second": 4663.607
    },
    {
      "epoch": 1.153061224489796,
      "grad_norm": 0.694202489015025,
      "learning_rate": 3.810381260939354e-06,
      "loss": 0.2147247314453125,
      "num_input_tokens_seen": 542054880,
      "step": 9040,
      "train_runtime": 116225.5582,
      "train_tokens_per_second": 4663.818
    },
    {
      "epoch": 1.1536989795918366,
      "grad_norm": 0.6691855742024077,
      "learning_rate": 3.8055167787050134e-06,
      "loss": 0.2462533950805664,
      "num_input_tokens_seen": 542359544,
      "step": 9045,
      "train_runtime": 116296.6266,
      "train_tokens_per_second": 4663.588
    },
    {
      "epoch": 1.1543367346938775,
      "grad_norm": 0.6256484008561468,
      "learning_rate": 3.8006534952180062e-06,
      "loss": 0.24401683807373048,
      "num_input_tokens_seen": 542659152,
      "step": 9050,
      "train_runtime": 116359.1845,
      "train_tokens_per_second": 4663.656
    },
    {
      "epoch": 1.1549744897959184,
      "grad_norm": 0.6430972241200976,
      "learning_rate": 3.79579141535898e-06,
      "loss": 0.24500937461853028,
      "num_input_tokens_seen": 542942632,
      "step": 9055,
      "train_runtime": 116422.6336,
      "train_tokens_per_second": 4663.549
    },
    {
      "epoch": 1.155612244897959,
      "grad_norm": 0.5955060538312635,
      "learning_rate": 3.7909305440073707e-06,
      "loss": 0.24865365028381348,
      "num_input_tokens_seen": 543240272,
      "step": 9060,
      "train_runtime": 116478.2095,
      "train_tokens_per_second": 4663.879
    },
    {
      "epoch": 1.15625,
      "grad_norm": 0.6177609398095152,
      "learning_rate": 3.786070886041401e-06,
      "loss": 0.22608916759490966,
      "num_input_tokens_seen": 543538488,
      "step": 9065,
      "train_runtime": 116542.1808,
      "train_tokens_per_second": 4663.878
    },
    {
      "epoch": 1.156887755102041,
      "grad_norm": 0.6200387597458965,
      "learning_rate": 3.7812124463380766e-06,
      "loss": 0.24038472175598144,
      "num_input_tokens_seen": 543834648,
      "step": 9070,
      "train_runtime": 116603.1814,
      "train_tokens_per_second": 4663.978
    },
    {
      "epoch": 1.1575255102040816,
      "grad_norm": 0.7038111749796568,
      "learning_rate": 3.7763552297731845e-06,
      "loss": 0.2305060863494873,
      "num_input_tokens_seen": 544124952,
      "step": 9075,
      "train_runtime": 116665.3221,
      "train_tokens_per_second": 4663.982
    },
    {
      "epoch": 1.1581632653061225,
      "grad_norm": 0.7451002695056345,
      "learning_rate": 3.771499241221277e-06,
      "loss": 0.24811456203460694,
      "num_input_tokens_seen": 544431408,
      "step": 9080,
      "train_runtime": 116728.2745,
      "train_tokens_per_second": 4664.092
    },
    {
      "epoch": 1.1588010204081634,
      "grad_norm": 0.6751730016623045,
      "learning_rate": 3.766644485555679e-06,
      "loss": 0.25781056880950926,
      "num_input_tokens_seen": 544735616,
      "step": 9085,
      "train_runtime": 116793.9754,
      "train_tokens_per_second": 4664.073
    },
    {
      "epoch": 1.159438775510204,
      "grad_norm": 0.6172387705145659,
      "learning_rate": 3.7617909676484788e-06,
      "loss": 0.22715344429016113,
      "num_input_tokens_seen": 545035608,
      "step": 9090,
      "train_runtime": 116862.9783,
      "train_tokens_per_second": 4663.886
    },
    {
      "epoch": 1.160076530612245,
      "grad_norm": 0.6562851751710456,
      "learning_rate": 3.7569386923705173e-06,
      "loss": 0.23681538105010985,
      "num_input_tokens_seen": 545335384,
      "step": 9095,
      "train_runtime": 116916.212,
      "train_tokens_per_second": 4664.326
    },
    {
      "epoch": 1.1607142857142858,
      "grad_norm": 0.602306701220074,
      "learning_rate": 3.752087664591394e-06,
      "loss": 0.24715650081634521,
      "num_input_tokens_seen": 545640128,
      "step": 9100,
      "train_runtime": 116965.1864,
      "train_tokens_per_second": 4664.979
    },
    {
      "epoch": 1.1613520408163265,
      "grad_norm": 0.6942235252239803,
      "learning_rate": 3.7472378891794537e-06,
      "loss": 0.2595963478088379,
      "num_input_tokens_seen": 545941880,
      "step": 9105,
      "train_runtime": 117029.5323,
      "train_tokens_per_second": 4664.992
    },
    {
      "epoch": 1.1619897959183674,
      "grad_norm": 0.6967520469521169,
      "learning_rate": 3.7423893710017826e-06,
      "loss": 0.24517674446105958,
      "num_input_tokens_seen": 546229888,
      "step": 9110,
      "train_runtime": 117095.5233,
      "train_tokens_per_second": 4664.823
    },
    {
      "epoch": 1.162627551020408,
      "grad_norm": 0.6292964339974995,
      "learning_rate": 3.7375421149242102e-06,
      "loss": 0.25058884620666505,
      "num_input_tokens_seen": 546533384,
      "step": 9115,
      "train_runtime": 117160.1731,
      "train_tokens_per_second": 4664.839
    },
    {
      "epoch": 1.163265306122449,
      "grad_norm": 0.6459696765253816,
      "learning_rate": 3.7326961258112963e-06,
      "loss": 0.2546118974685669,
      "num_input_tokens_seen": 546827608,
      "step": 9120,
      "train_runtime": 117226.779,
      "train_tokens_per_second": 4664.699
    },
    {
      "epoch": 1.1639030612244898,
      "grad_norm": 0.6359502309514716,
      "learning_rate": 3.7278514085263275e-06,
      "loss": 0.2721463680267334,
      "num_input_tokens_seen": 547120440,
      "step": 9125,
      "train_runtime": 117289.348,
      "train_tokens_per_second": 4664.707
    },
    {
      "epoch": 1.1645408163265305,
      "grad_norm": 0.6769281285982229,
      "learning_rate": 3.7230079679313157e-06,
      "loss": 0.256221342086792,
      "num_input_tokens_seen": 547427632,
      "step": 9130,
      "train_runtime": 117349.2818,
      "train_tokens_per_second": 4664.942
    },
    {
      "epoch": 1.1651785714285714,
      "grad_norm": 0.8581595851773528,
      "learning_rate": 3.7181658088869954e-06,
      "loss": 0.26300437450408937,
      "num_input_tokens_seen": 547739000,
      "step": 9135,
      "train_runtime": 117423.8759,
      "train_tokens_per_second": 4664.631
    },
    {
      "epoch": 1.1658163265306123,
      "grad_norm": 0.6424659767116858,
      "learning_rate": 3.7133249362528075e-06,
      "loss": 0.24833297729492188,
      "num_input_tokens_seen": 548037560,
      "step": 9140,
      "train_runtime": 117495.8573,
      "train_tokens_per_second": 4664.314
    },
    {
      "epoch": 1.166454081632653,
      "grad_norm": 0.631329380463012,
      "learning_rate": 3.708485354886906e-06,
      "loss": 0.22677249908447267,
      "num_input_tokens_seen": 548346544,
      "step": 9145,
      "train_runtime": 117559.1515,
      "train_tokens_per_second": 4664.431
    },
    {
      "epoch": 1.1670918367346939,
      "grad_norm": 0.6714517377518626,
      "learning_rate": 3.7036470696461535e-06,
      "loss": 0.25949141979217527,
      "num_input_tokens_seen": 548637168,
      "step": 9150,
      "train_runtime": 117622.3477,
      "train_tokens_per_second": 4664.396
    },
    {
      "epoch": 1.1677295918367347,
      "grad_norm": 0.6835736808563946,
      "learning_rate": 3.698810085386101e-06,
      "loss": 0.2544286012649536,
      "num_input_tokens_seen": 548931232,
      "step": 9155,
      "train_runtime": 117689.8689,
      "train_tokens_per_second": 4664.218
    },
    {
      "epoch": 1.1683673469387754,
      "grad_norm": 0.6110788024764375,
      "learning_rate": 3.693974406961004e-06,
      "loss": 0.2577073335647583,
      "num_input_tokens_seen": 549245168,
      "step": 9160,
      "train_runtime": 117750.8411,
      "train_tokens_per_second": 4664.469
    },
    {
      "epoch": 1.1690051020408163,
      "grad_norm": 0.6770064378364575,
      "learning_rate": 3.689140039223802e-06,
      "loss": 0.25592479705810545,
      "num_input_tokens_seen": 549548056,
      "step": 9165,
      "train_runtime": 117824.0662,
      "train_tokens_per_second": 4664.141
    },
    {
      "epoch": 1.1696428571428572,
      "grad_norm": 0.6187611265175217,
      "learning_rate": 3.6843069870261196e-06,
      "loss": 0.2484119176864624,
      "num_input_tokens_seen": 549839520,
      "step": 9170,
      "train_runtime": 117888.4783,
      "train_tokens_per_second": 4664.065
    },
    {
      "epoch": 1.1702806122448979,
      "grad_norm": 0.5803643727511468,
      "learning_rate": 3.6794752552182654e-06,
      "loss": 0.24884932041168212,
      "num_input_tokens_seen": 550147808,
      "step": 9175,
      "train_runtime": 117950.1268,
      "train_tokens_per_second": 4664.241
    },
    {
      "epoch": 1.1709183673469388,
      "grad_norm": 0.6528384407806481,
      "learning_rate": 3.674644848649217e-06,
      "loss": 0.24987993240356446,
      "num_input_tokens_seen": 550460824,
      "step": 9180,
      "train_runtime": 118009.1496,
      "train_tokens_per_second": 4664.561
    },
    {
      "epoch": 1.1715561224489797,
      "grad_norm": 0.6322431619920269,
      "learning_rate": 3.669815772166625e-06,
      "loss": 0.23595900535583497,
      "num_input_tokens_seen": 550767152,
      "step": 9185,
      "train_runtime": 118081.2508,
      "train_tokens_per_second": 4664.307
    },
    {
      "epoch": 1.1721938775510203,
      "grad_norm": 0.6027450146415512,
      "learning_rate": 3.6649880306168073e-06,
      "loss": 0.25257282257080077,
      "num_input_tokens_seen": 551070096,
      "step": 9190,
      "train_runtime": 118152.1277,
      "train_tokens_per_second": 4664.073
    },
    {
      "epoch": 1.1728316326530612,
      "grad_norm": 0.6259146317279678,
      "learning_rate": 3.6601616288447352e-06,
      "loss": 0.22191522121429444,
      "num_input_tokens_seen": 551382240,
      "step": 9195,
      "train_runtime": 118221.1076,
      "train_tokens_per_second": 4663.991
    },
    {
      "epoch": 1.1734693877551021,
      "grad_norm": 0.7630515997330095,
      "learning_rate": 3.655336571694044e-06,
      "loss": 0.26834137439727784,
      "num_input_tokens_seen": 551668544,
      "step": 9200,
      "train_runtime": 118284.8524,
      "train_tokens_per_second": 4663.898
    },
    {
      "epoch": 1.1741071428571428,
      "grad_norm": 0.597978151035573,
      "learning_rate": 3.6505128640070152e-06,
      "loss": 0.24073524475097657,
      "num_input_tokens_seen": 551963936,
      "step": 9205,
      "train_runtime": 118361.3459,
      "train_tokens_per_second": 4663.38
    },
    {
      "epoch": 1.1747448979591837,
      "grad_norm": 0.6130406112564567,
      "learning_rate": 3.6456905106245743e-06,
      "loss": 0.253322958946228,
      "num_input_tokens_seen": 552254864,
      "step": 9210,
      "train_runtime": 118421.5701,
      "train_tokens_per_second": 4663.465
    },
    {
      "epoch": 1.1753826530612246,
      "grad_norm": 0.6039484361070391,
      "learning_rate": 3.6408695163862895e-06,
      "loss": 0.22835330963134765,
      "num_input_tokens_seen": 552565152,
      "step": 9215,
      "train_runtime": 118482.4192,
      "train_tokens_per_second": 4663.689
    },
    {
      "epoch": 1.1760204081632653,
      "grad_norm": 0.7558607931592242,
      "learning_rate": 3.6360498861303693e-06,
      "loss": 0.2553338766098022,
      "num_input_tokens_seen": 552865136,
      "step": 9220,
      "train_runtime": 118547.5556,
      "train_tokens_per_second": 4663.657
    },
    {
      "epoch": 1.1766581632653061,
      "grad_norm": 0.6296071342408717,
      "learning_rate": 3.631231624693645e-06,
      "loss": 0.24725809097290039,
      "num_input_tokens_seen": 553177424,
      "step": 9225,
      "train_runtime": 118624.0642,
      "train_tokens_per_second": 4663.282
    },
    {
      "epoch": 1.177295918367347,
      "grad_norm": 0.6134551301136502,
      "learning_rate": 3.6264147369115788e-06,
      "loss": 0.2447408676147461,
      "num_input_tokens_seen": 553478216,
      "step": 9230,
      "train_runtime": 118695.0777,
      "train_tokens_per_second": 4663.026
    },
    {
      "epoch": 1.1779336734693877,
      "grad_norm": 0.609506357508338,
      "learning_rate": 3.621599227618258e-06,
      "loss": 0.24174644947052001,
      "num_input_tokens_seen": 553766792,
      "step": 9235,
      "train_runtime": 118759.8057,
      "train_tokens_per_second": 4662.914
    },
    {
      "epoch": 1.1785714285714286,
      "grad_norm": 0.6881325119035495,
      "learning_rate": 3.6167851016463762e-06,
      "loss": 0.25118389129638674,
      "num_input_tokens_seen": 554078144,
      "step": 9240,
      "train_runtime": 118833.6779,
      "train_tokens_per_second": 4662.636
    },
    {
      "epoch": 1.1792091836734695,
      "grad_norm": 0.6949850142071543,
      "learning_rate": 3.61197236382725e-06,
      "loss": 0.264749813079834,
      "num_input_tokens_seen": 554353128,
      "step": 9245,
      "train_runtime": 118901.1332,
      "train_tokens_per_second": 4662.303
    },
    {
      "epoch": 1.1798469387755102,
      "grad_norm": 0.6237410584708641,
      "learning_rate": 3.6071610189907975e-06,
      "loss": 0.2633744955062866,
      "num_input_tokens_seen": 554654464,
      "step": 9250,
      "train_runtime": 118961.4601,
      "train_tokens_per_second": 4662.472
    },
    {
      "epoch": 1.180484693877551,
      "grad_norm": 0.607375416487404,
      "learning_rate": 3.602351071965536e-06,
      "loss": 0.2352853775024414,
      "num_input_tokens_seen": 554957296,
      "step": 9255,
      "train_runtime": 119024.8538,
      "train_tokens_per_second": 4662.533
    },
    {
      "epoch": 1.181122448979592,
      "grad_norm": 0.6229404929317026,
      "learning_rate": 3.597542527578586e-06,
      "loss": 0.2447524070739746,
      "num_input_tokens_seen": 555268560,
      "step": 9260,
      "train_runtime": 119074.8666,
      "train_tokens_per_second": 4663.189
    },
    {
      "epoch": 1.1817602040816326,
      "grad_norm": 0.6978467524958565,
      "learning_rate": 3.5927353906556583e-06,
      "loss": 0.25916457176208496,
      "num_input_tokens_seen": 555550816,
      "step": 9265,
      "train_runtime": 119142.412,
      "train_tokens_per_second": 4662.914
    },
    {
      "epoch": 1.1823979591836735,
      "grad_norm": 0.6531276949703494,
      "learning_rate": 3.587929666021048e-06,
      "loss": 0.25164358615875243,
      "num_input_tokens_seen": 555873480,
      "step": 9270,
      "train_runtime": 119195.4844,
      "train_tokens_per_second": 4663.545
    },
    {
      "epoch": 1.1830357142857142,
      "grad_norm": 0.7364330224536769,
      "learning_rate": 3.5831253584976377e-06,
      "loss": 0.24759864807128906,
      "num_input_tokens_seen": 556177128,
      "step": 9275,
      "train_runtime": 119260.0804,
      "train_tokens_per_second": 4663.565
    },
    {
      "epoch": 1.183673469387755,
      "grad_norm": 0.6462113257712689,
      "learning_rate": 3.578322472906882e-06,
      "loss": 0.2553884506225586,
      "num_input_tokens_seen": 556473376,
      "step": 9280,
      "train_runtime": 119327.0084,
      "train_tokens_per_second": 4663.432
    },
    {
      "epoch": 1.184311224489796,
      "grad_norm": 0.629273389281804,
      "learning_rate": 3.5735210140688163e-06,
      "loss": 0.23726444244384765,
      "num_input_tokens_seen": 556768616,
      "step": 9285,
      "train_runtime": 119397.3057,
      "train_tokens_per_second": 4663.159
    },
    {
      "epoch": 1.1849489795918366,
      "grad_norm": 0.6166447669029329,
      "learning_rate": 3.568720986802039e-06,
      "loss": 0.23703157901763916,
      "num_input_tokens_seen": 557075072,
      "step": 9290,
      "train_runtime": 119464.9214,
      "train_tokens_per_second": 4663.085
    },
    {
      "epoch": 1.1855867346938775,
      "grad_norm": 0.5888052487622764,
      "learning_rate": 3.56392239592371e-06,
      "loss": 0.23610968589782716,
      "num_input_tokens_seen": 557373416,
      "step": 9295,
      "train_runtime": 119541.318,
      "train_tokens_per_second": 4662.601
    },
    {
      "epoch": 1.1862244897959184,
      "grad_norm": 0.7006106595864101,
      "learning_rate": 3.559125246249552e-06,
      "loss": 0.24983363151550292,
      "num_input_tokens_seen": 557676320,
      "step": 9300,
      "train_runtime": 119611.277,
      "train_tokens_per_second": 4662.406
    },
    {
      "epoch": 1.186862244897959,
      "grad_norm": 0.6302775441592273,
      "learning_rate": 3.554329542593841e-06,
      "loss": 0.24187581539154052,
      "num_input_tokens_seen": 557962864,
      "step": 9305,
      "train_runtime": 119668.3273,
      "train_tokens_per_second": 4662.578
    },
    {
      "epoch": 1.1875,
      "grad_norm": 0.6658184462450729,
      "learning_rate": 3.549535289769399e-06,
      "loss": 0.24613499641418457,
      "num_input_tokens_seen": 558233360,
      "step": 9310,
      "train_runtime": 119737.4592,
      "train_tokens_per_second": 4662.145
    },
    {
      "epoch": 1.188137755102041,
      "grad_norm": 0.726298207769721,
      "learning_rate": 3.544742492587593e-06,
      "loss": 0.28460307121276857,
      "num_input_tokens_seen": 558524488,
      "step": 9315,
      "train_runtime": 119805.6747,
      "train_tokens_per_second": 4661.92
    },
    {
      "epoch": 1.1887755102040816,
      "grad_norm": 0.690872033354097,
      "learning_rate": 3.5399511558583323e-06,
      "loss": 0.24558644294738768,
      "num_input_tokens_seen": 558822904,
      "step": 9320,
      "train_runtime": 119877.3924,
      "train_tokens_per_second": 4661.62
    },
    {
      "epoch": 1.1894132653061225,
      "grad_norm": 0.636094550133446,
      "learning_rate": 3.5351612843900555e-06,
      "loss": 0.23512711524963378,
      "num_input_tokens_seen": 559123576,
      "step": 9325,
      "train_runtime": 119947.4983,
      "train_tokens_per_second": 4661.403
    },
    {
      "epoch": 1.1900510204081634,
      "grad_norm": 0.6683601460308425,
      "learning_rate": 3.5303728829897336e-06,
      "loss": 0.25034427642822266,
      "num_input_tokens_seen": 559425440,
      "step": 9330,
      "train_runtime": 120013.1594,
      "train_tokens_per_second": 4661.367
    },
    {
      "epoch": 1.190688775510204,
      "grad_norm": 0.6695643430497313,
      "learning_rate": 3.5255859564628637e-06,
      "loss": 0.24240822792053224,
      "num_input_tokens_seen": 559730616,
      "step": 9335,
      "train_runtime": 120076.7622,
      "train_tokens_per_second": 4661.44
    },
    {
      "epoch": 1.191326530612245,
      "grad_norm": 0.6623470970724884,
      "learning_rate": 3.5208005096134567e-06,
      "loss": 0.2593933820724487,
      "num_input_tokens_seen": 560028952,
      "step": 9340,
      "train_runtime": 120146.2545,
      "train_tokens_per_second": 4661.227
    },
    {
      "epoch": 1.1919642857142858,
      "grad_norm": 0.6678385536760265,
      "learning_rate": 3.516016547244047e-06,
      "loss": 0.21670892238616943,
      "num_input_tokens_seen": 560343528,
      "step": 9345,
      "train_runtime": 120212.6634,
      "train_tokens_per_second": 4661.269
    },
    {
      "epoch": 1.1926020408163265,
      "grad_norm": 0.8182870131267127,
      "learning_rate": 3.5112340741556737e-06,
      "loss": 0.26225972175598145,
      "num_input_tokens_seen": 560643024,
      "step": 9350,
      "train_runtime": 120270.7753,
      "train_tokens_per_second": 4661.507
    },
    {
      "epoch": 1.1932397959183674,
      "grad_norm": 0.5997101290632552,
      "learning_rate": 3.5064530951478822e-06,
      "loss": 0.23266761302947997,
      "num_input_tokens_seen": 560939344,
      "step": 9355,
      "train_runtime": 120341.6154,
      "train_tokens_per_second": 4661.225
    },
    {
      "epoch": 1.193877551020408,
      "grad_norm": 0.6497956114394874,
      "learning_rate": 3.501673615018717e-06,
      "loss": 0.24074554443359375,
      "num_input_tokens_seen": 561249864,
      "step": 9360,
      "train_runtime": 120409.5106,
      "train_tokens_per_second": 4661.176
    },
    {
      "epoch": 1.194515306122449,
      "grad_norm": 0.6099208940229317,
      "learning_rate": 3.496895638564724e-06,
      "loss": 0.247788667678833,
      "num_input_tokens_seen": 561551416,
      "step": 9365,
      "train_runtime": 120465.6494,
      "train_tokens_per_second": 4661.507
    },
    {
      "epoch": 1.1951530612244898,
      "grad_norm": 0.6974162082537729,
      "learning_rate": 3.492119170580933e-06,
      "loss": 0.2576225996017456,
      "num_input_tokens_seen": 561860984,
      "step": 9370,
      "train_runtime": 120538.6994,
      "train_tokens_per_second": 4661.25
    },
    {
      "epoch": 1.1957908163265305,
      "grad_norm": 0.6831801746858304,
      "learning_rate": 3.4873442158608638e-06,
      "loss": 0.24086074829101561,
      "num_input_tokens_seen": 562138512,
      "step": 9375,
      "train_runtime": 120602.265,
      "train_tokens_per_second": 4661.094
    },
    {
      "epoch": 1.1964285714285714,
      "grad_norm": 0.7347750564493745,
      "learning_rate": 3.4825707791965213e-06,
      "loss": 0.25907833576202394,
      "num_input_tokens_seen": 562444504,
      "step": 9380,
      "train_runtime": 120667.4375,
      "train_tokens_per_second": 4661.113
    },
    {
      "epoch": 1.1970663265306123,
      "grad_norm": 0.6830529434820147,
      "learning_rate": 3.477798865378375e-06,
      "loss": 0.2647451877593994,
      "num_input_tokens_seen": 562736128,
      "step": 9385,
      "train_runtime": 120744.8089,
      "train_tokens_per_second": 4660.541
    },
    {
      "epoch": 1.197704081632653,
      "grad_norm": 0.6699661918602079,
      "learning_rate": 3.4730284791953793e-06,
      "loss": 0.23806190490722656,
      "num_input_tokens_seen": 563036856,
      "step": 9390,
      "train_runtime": 120815.7666,
      "train_tokens_per_second": 4660.293
    },
    {
      "epoch": 1.1983418367346939,
      "grad_norm": 0.6660307024039173,
      "learning_rate": 3.4682596254349487e-06,
      "loss": 0.2621174335479736,
      "num_input_tokens_seen": 563341440,
      "step": 9395,
      "train_runtime": 120879.9135,
      "train_tokens_per_second": 4660.34
    },
    {
      "epoch": 1.1989795918367347,
      "grad_norm": 0.6888233895005237,
      "learning_rate": 3.4634923088829586e-06,
      "loss": 0.23306708335876464,
      "num_input_tokens_seen": 563643520,
      "step": 9400,
      "train_runtime": 120954.1091,
      "train_tokens_per_second": 4659.978
    },
    {
      "epoch": 1.1996173469387754,
      "grad_norm": 0.6150800837168472,
      "learning_rate": 3.4587265343237474e-06,
      "loss": 0.2485969066619873,
      "num_input_tokens_seen": 563936920,
      "step": 9405,
      "train_runtime": 121018.3573,
      "train_tokens_per_second": 4659.929
    },
    {
      "epoch": 1.2002551020408163,
      "grad_norm": 0.6353422200146611,
      "learning_rate": 3.4539623065401014e-06,
      "loss": 0.2417161464691162,
      "num_input_tokens_seen": 564243456,
      "step": 9410,
      "train_runtime": 121084.8256,
      "train_tokens_per_second": 4659.902
    },
    {
      "epoch": 1.2008928571428572,
      "grad_norm": 0.6336374806099082,
      "learning_rate": 3.4491996303132553e-06,
      "loss": 0.2554072380065918,
      "num_input_tokens_seen": 564555928,
      "step": 9415,
      "train_runtime": 121138.3299,
      "train_tokens_per_second": 4660.424
    },
    {
      "epoch": 1.2015306122448979,
      "grad_norm": 0.6884006889013108,
      "learning_rate": 3.444438510422889e-06,
      "loss": 0.22291643619537355,
      "num_input_tokens_seen": 564840192,
      "step": 9420,
      "train_runtime": 121194.2743,
      "train_tokens_per_second": 4660.618
    },
    {
      "epoch": 1.2021683673469388,
      "grad_norm": 0.5912748419669106,
      "learning_rate": 3.4396789516471152e-06,
      "loss": 0.23480658531188964,
      "num_input_tokens_seen": 565145792,
      "step": 9425,
      "train_runtime": 121257.7308,
      "train_tokens_per_second": 4660.699
    },
    {
      "epoch": 1.2028061224489797,
      "grad_norm": 0.6174977455663048,
      "learning_rate": 3.4349209587624876e-06,
      "loss": 0.259614086151123,
      "num_input_tokens_seen": 565446576,
      "step": 9430,
      "train_runtime": 121325.7457,
      "train_tokens_per_second": 4660.565
    },
    {
      "epoch": 1.2034438775510203,
      "grad_norm": 0.6502749099140333,
      "learning_rate": 3.4301645365439827e-06,
      "loss": 0.26551547050476076,
      "num_input_tokens_seen": 565764464,
      "step": 9435,
      "train_runtime": 121386.7255,
      "train_tokens_per_second": 4660.843
    },
    {
      "epoch": 1.2040816326530612,
      "grad_norm": 0.6233623788941873,
      "learning_rate": 3.4254096897650026e-06,
      "loss": 0.23483748435974122,
      "num_input_tokens_seen": 566077560,
      "step": 9440,
      "train_runtime": 121445.9811,
      "train_tokens_per_second": 4661.147
    },
    {
      "epoch": 1.2047193877551021,
      "grad_norm": 0.6713930657422693,
      "learning_rate": 3.4206564231973664e-06,
      "loss": 0.23852186203002929,
      "num_input_tokens_seen": 566381952,
      "step": 9445,
      "train_runtime": 121502.5826,
      "train_tokens_per_second": 4661.481
    },
    {
      "epoch": 1.2053571428571428,
      "grad_norm": 0.5998381766696496,
      "learning_rate": 3.4159047416113133e-06,
      "loss": 0.23655376434326172,
      "num_input_tokens_seen": 566690112,
      "step": 9450,
      "train_runtime": 121568.6943,
      "train_tokens_per_second": 4661.481
    },
    {
      "epoch": 1.2059948979591837,
      "grad_norm": 0.585610590757165,
      "learning_rate": 3.4111546497754834e-06,
      "loss": 0.2396397113800049,
      "num_input_tokens_seen": 566993576,
      "step": 9455,
      "train_runtime": 121623.1231,
      "train_tokens_per_second": 4661.89
    },
    {
      "epoch": 1.2066326530612246,
      "grad_norm": 0.6394254926611324,
      "learning_rate": 3.406406152456926e-06,
      "loss": 0.25543689727783203,
      "num_input_tokens_seen": 567296048,
      "step": 9460,
      "train_runtime": 121692.9237,
      "train_tokens_per_second": 4661.701
    },
    {
      "epoch": 1.2072704081632653,
      "grad_norm": 0.6736076741793873,
      "learning_rate": 3.4016592544210937e-06,
      "loss": 0.22780184745788573,
      "num_input_tokens_seen": 567611232,
      "step": 9465,
      "train_runtime": 121748.117,
      "train_tokens_per_second": 4662.177
    },
    {
      "epoch": 1.2079081632653061,
      "grad_norm": 0.724909092864873,
      "learning_rate": 3.3969139604318257e-06,
      "loss": 0.2502995491027832,
      "num_input_tokens_seen": 567904160,
      "step": 9470,
      "train_runtime": 121819.3603,
      "train_tokens_per_second": 4661.855
    },
    {
      "epoch": 1.208545918367347,
      "grad_norm": 0.6354101488056316,
      "learning_rate": 3.3921702752513576e-06,
      "loss": 0.23016157150268554,
      "num_input_tokens_seen": 568200096,
      "step": 9475,
      "train_runtime": 121870.4569,
      "train_tokens_per_second": 4662.328
    },
    {
      "epoch": 1.2091836734693877,
      "grad_norm": 0.7218693513273385,
      "learning_rate": 3.3874282036403106e-06,
      "loss": 0.23276865482330322,
      "num_input_tokens_seen": 568498312,
      "step": 9480,
      "train_runtime": 121943.7495,
      "train_tokens_per_second": 4661.972
    },
    {
      "epoch": 1.2098214285714286,
      "grad_norm": 0.7254054278864226,
      "learning_rate": 3.3826877503576805e-06,
      "loss": 0.24601442813873292,
      "num_input_tokens_seen": 568788904,
      "step": 9485,
      "train_runtime": 122015.1089,
      "train_tokens_per_second": 4661.627
    },
    {
      "epoch": 1.2104591836734695,
      "grad_norm": 0.7746722846285617,
      "learning_rate": 3.3779489201608468e-06,
      "loss": 0.25545225143432615,
      "num_input_tokens_seen": 569073328,
      "step": 9490,
      "train_runtime": 122069.6447,
      "train_tokens_per_second": 4661.874
    },
    {
      "epoch": 1.2110969387755102,
      "grad_norm": 0.634040013325104,
      "learning_rate": 3.373211717805557e-06,
      "loss": 0.24503130912780763,
      "num_input_tokens_seen": 569371048,
      "step": 9495,
      "train_runtime": 122136.9593,
      "train_tokens_per_second": 4661.742
    },
    {
      "epoch": 1.211734693877551,
      "grad_norm": 0.5463735961398527,
      "learning_rate": 3.3684761480459218e-06,
      "loss": 0.24413666725158692,
      "num_input_tokens_seen": 569682568,
      "step": 9500,
      "train_runtime": 122204.3646,
      "train_tokens_per_second": 4661.72
    },
    {
      "epoch": 1.212372448979592,
      "grad_norm": 0.640346597606397,
      "learning_rate": 3.363742215634416e-06,
      "loss": 0.22815101146697997,
      "num_input_tokens_seen": 569986536,
      "step": 9505,
      "train_runtime": 122259.3282,
      "train_tokens_per_second": 4662.111
    },
    {
      "epoch": 1.2130102040816326,
      "grad_norm": 0.5562411538974917,
      "learning_rate": 3.3590099253218745e-06,
      "loss": 0.24046213626861573,
      "num_input_tokens_seen": 570289152,
      "step": 9510,
      "train_runtime": 122326.6957,
      "train_tokens_per_second": 4662.017
    },
    {
      "epoch": 1.2136479591836735,
      "grad_norm": 0.594256768739644,
      "learning_rate": 3.3542792818574793e-06,
      "loss": 0.2501716136932373,
      "num_input_tokens_seen": 570600240,
      "step": 9515,
      "train_runtime": 122391.128,
      "train_tokens_per_second": 4662.105
    },
    {
      "epoch": 1.2142857142857142,
      "grad_norm": 0.6750620716967806,
      "learning_rate": 3.3495502899887624e-06,
      "loss": 0.26310560703277586,
      "num_input_tokens_seen": 570889384,
      "step": 9520,
      "train_runtime": 122450.4879,
      "train_tokens_per_second": 4662.206
    },
    {
      "epoch": 1.214923469387755,
      "grad_norm": 0.7131524670740134,
      "learning_rate": 3.344822954461595e-06,
      "loss": 0.22837610244750978,
      "num_input_tokens_seen": 571197592,
      "step": 9525,
      "train_runtime": 122510.2591,
      "train_tokens_per_second": 4662.447
    },
    {
      "epoch": 1.215561224489796,
      "grad_norm": 0.6539464488458759,
      "learning_rate": 3.34009728002019e-06,
      "loss": 0.23309762477874757,
      "num_input_tokens_seen": 571495144,
      "step": 9530,
      "train_runtime": 122579.2496,
      "train_tokens_per_second": 4662.25
    },
    {
      "epoch": 1.2161989795918366,
      "grad_norm": 0.6845398888019838,
      "learning_rate": 3.3353732714070933e-06,
      "loss": 0.23382034301757812,
      "num_input_tokens_seen": 571790752,
      "step": 9535,
      "train_runtime": 122636.2501,
      "train_tokens_per_second": 4662.494
    },
    {
      "epoch": 1.2168367346938775,
      "grad_norm": 0.5880402846387599,
      "learning_rate": 3.3306509333631755e-06,
      "loss": 0.24842290878295897,
      "num_input_tokens_seen": 572077008,
      "step": 9540,
      "train_runtime": 122703.9193,
      "train_tokens_per_second": 4662.255
    },
    {
      "epoch": 1.2174744897959184,
      "grad_norm": 0.6439471879577157,
      "learning_rate": 3.325930270627632e-06,
      "loss": 0.2351449489593506,
      "num_input_tokens_seen": 572344136,
      "step": 9545,
      "train_runtime": 122760.4623,
      "train_tokens_per_second": 4662.284
    },
    {
      "epoch": 1.218112244897959,
      "grad_norm": 0.6941666628201706,
      "learning_rate": 3.3212112879379806e-06,
      "loss": 0.2537204742431641,
      "num_input_tokens_seen": 572636944,
      "step": 9550,
      "train_runtime": 122825.5499,
      "train_tokens_per_second": 4662.197
    },
    {
      "epoch": 1.21875,
      "grad_norm": 0.8305207909209866,
      "learning_rate": 3.3164939900300467e-06,
      "loss": 0.25160627365112304,
      "num_input_tokens_seen": 572928200,
      "step": 9555,
      "train_runtime": 122889.2114,
      "train_tokens_per_second": 4662.152
    },
    {
      "epoch": 1.219387755102041,
      "grad_norm": 0.6425826189693363,
      "learning_rate": 3.31177838163797e-06,
      "loss": 0.2512808084487915,
      "num_input_tokens_seen": 573228480,
      "step": 9560,
      "train_runtime": 122954.6929,
      "train_tokens_per_second": 4662.111
    },
    {
      "epoch": 1.2200255102040816,
      "grad_norm": 0.7235417645152831,
      "learning_rate": 3.3070644674941938e-06,
      "loss": 0.2483961582183838,
      "num_input_tokens_seen": 573525744,
      "step": 9565,
      "train_runtime": 123017.8877,
      "train_tokens_per_second": 4662.133
    },
    {
      "epoch": 1.2206632653061225,
      "grad_norm": 0.676437011184346,
      "learning_rate": 3.302352252329456e-06,
      "loss": 0.2219179630279541,
      "num_input_tokens_seen": 573826040,
      "step": 9570,
      "train_runtime": 123089.1066,
      "train_tokens_per_second": 4661.875
    },
    {
      "epoch": 1.2213010204081634,
      "grad_norm": 0.7021298551423343,
      "learning_rate": 3.2976417408727978e-06,
      "loss": 0.2485231876373291,
      "num_input_tokens_seen": 574124728,
      "step": 9575,
      "train_runtime": 123151.3361,
      "train_tokens_per_second": 4661.945
    },
    {
      "epoch": 1.221938775510204,
      "grad_norm": 0.5885944348296309,
      "learning_rate": 3.2929329378515466e-06,
      "loss": 0.2570828914642334,
      "num_input_tokens_seen": 574445152,
      "step": 9580,
      "train_runtime": 123215.1794,
      "train_tokens_per_second": 4662.13
    },
    {
      "epoch": 1.222576530612245,
      "grad_norm": 0.719852957071995,
      "learning_rate": 3.288225847991312e-06,
      "loss": 0.249613356590271,
      "num_input_tokens_seen": 574736216,
      "step": 9585,
      "train_runtime": 123285.219,
      "train_tokens_per_second": 4661.842
    },
    {
      "epoch": 1.2232142857142858,
      "grad_norm": 0.6467967168188589,
      "learning_rate": 3.2835204760159883e-06,
      "loss": 0.2640578269958496,
      "num_input_tokens_seen": 575033336,
      "step": 9590,
      "train_runtime": 123339.5056,
      "train_tokens_per_second": 4662.199
    },
    {
      "epoch": 1.2238520408163265,
      "grad_norm": 0.618876874848355,
      "learning_rate": 3.278816826647747e-06,
      "loss": 0.22925953865051268,
      "num_input_tokens_seen": 575322848,
      "step": 9595,
      "train_runtime": 123403.1543,
      "train_tokens_per_second": 4662.141
    },
    {
      "epoch": 1.2244897959183674,
      "grad_norm": 1.0537863884932905,
      "learning_rate": 3.2741149046070274e-06,
      "loss": 0.247985315322876,
      "num_input_tokens_seen": 575622176,
      "step": 9600,
      "train_runtime": 123469.0493,
      "train_tokens_per_second": 4662.077
    },
    {
      "epoch": 1.225127551020408,
      "grad_norm": 0.672465506512946,
      "learning_rate": 3.269414714612534e-06,
      "loss": 0.2552971839904785,
      "num_input_tokens_seen": 575936872,
      "step": 9605,
      "train_runtime": 123537.8016,
      "train_tokens_per_second": 4662.029
    },
    {
      "epoch": 1.225765306122449,
      "grad_norm": 0.6439160971463344,
      "learning_rate": 3.2647162613812435e-06,
      "loss": 0.25532970428466795,
      "num_input_tokens_seen": 576231408,
      "step": 9610,
      "train_runtime": 123598.8492,
      "train_tokens_per_second": 4662.11
    },
    {
      "epoch": 1.2264030612244898,
      "grad_norm": 0.6707144940046731,
      "learning_rate": 3.2600195496283716e-06,
      "loss": 0.2427670478820801,
      "num_input_tokens_seen": 576540112,
      "step": 9615,
      "train_runtime": 123647.5813,
      "train_tokens_per_second": 4662.769
    },
    {
      "epoch": 1.2270408163265305,
      "grad_norm": 0.6483414717759375,
      "learning_rate": 3.2553245840674037e-06,
      "loss": 0.2417902946472168,
      "num_input_tokens_seen": 576838640,
      "step": 9620,
      "train_runtime": 123700.2175,
      "train_tokens_per_second": 4663.198
    },
    {
      "epoch": 1.2276785714285714,
      "grad_norm": 0.5917535571007025,
      "learning_rate": 3.2506313694100643e-06,
      "loss": 0.2310344696044922,
      "num_input_tokens_seen": 577150248,
      "step": 9625,
      "train_runtime": 123781.3019,
      "train_tokens_per_second": 4662.661
    },
    {
      "epoch": 1.2283163265306123,
      "grad_norm": 0.6605121606704848,
      "learning_rate": 3.245939910366319e-06,
      "loss": 0.22664246559143067,
      "num_input_tokens_seen": 577424432,
      "step": 9630,
      "train_runtime": 123848.237,
      "train_tokens_per_second": 4662.355
    },
    {
      "epoch": 1.228954081632653,
      "grad_norm": 0.6414315445805147,
      "learning_rate": 3.2412502116443783e-06,
      "loss": 0.2320406436920166,
      "num_input_tokens_seen": 577731528,
      "step": 9635,
      "train_runtime": 123904.2092,
      "train_tokens_per_second": 4662.727
    },
    {
      "epoch": 1.2295918367346939,
      "grad_norm": 0.610721756241424,
      "learning_rate": 3.236562277950682e-06,
      "loss": 0.2285013198852539,
      "num_input_tokens_seen": 578022736,
      "step": 9640,
      "train_runtime": 123955.2853,
      "train_tokens_per_second": 4663.155
    },
    {
      "epoch": 1.2302295918367347,
      "grad_norm": 0.6632598272231298,
      "learning_rate": 3.231876113989897e-06,
      "loss": 0.24210233688354493,
      "num_input_tokens_seen": 578316872,
      "step": 9645,
      "train_runtime": 124019.0156,
      "train_tokens_per_second": 4663.131
    },
    {
      "epoch": 1.2308673469387754,
      "grad_norm": 0.6290910465594164,
      "learning_rate": 3.227191724464918e-06,
      "loss": 0.23908476829528807,
      "num_input_tokens_seen": 578603112,
      "step": 9650,
      "train_runtime": 124083.7123,
      "train_tokens_per_second": 4663.006
    },
    {
      "epoch": 1.2315051020408163,
      "grad_norm": 0.6028383692432482,
      "learning_rate": 3.2225091140768537e-06,
      "loss": 0.23353929519653321,
      "num_input_tokens_seen": 578893808,
      "step": 9655,
      "train_runtime": 124146.8512,
      "train_tokens_per_second": 4662.976
    },
    {
      "epoch": 1.2321428571428572,
      "grad_norm": 0.6940282199334723,
      "learning_rate": 3.2178282875250356e-06,
      "loss": 0.24055807590484618,
      "num_input_tokens_seen": 579190224,
      "step": 9660,
      "train_runtime": 124213.7144,
      "train_tokens_per_second": 4662.852
    },
    {
      "epoch": 1.2327806122448979,
      "grad_norm": 0.6351038537015795,
      "learning_rate": 3.213149249506997e-06,
      "loss": 0.2548715114593506,
      "num_input_tokens_seen": 579491968,
      "step": 9665,
      "train_runtime": 124276.8311,
      "train_tokens_per_second": 4662.912
    },
    {
      "epoch": 1.2334183673469388,
      "grad_norm": 0.7180850012402774,
      "learning_rate": 3.208472004718479e-06,
      "loss": 0.2566064834594727,
      "num_input_tokens_seen": 579772312,
      "step": 9670,
      "train_runtime": 124341.8525,
      "train_tokens_per_second": 4662.729
    },
    {
      "epoch": 1.2340561224489797,
      "grad_norm": 0.5798785800653484,
      "learning_rate": 3.2037965578534245e-06,
      "loss": 0.26011550426483154,
      "num_input_tokens_seen": 580074776,
      "step": 9675,
      "train_runtime": 124406.4191,
      "train_tokens_per_second": 4662.74
    },
    {
      "epoch": 1.2346938775510203,
      "grad_norm": 0.8496378610750186,
      "learning_rate": 3.199122913603972e-06,
      "loss": 0.2577479362487793,
      "num_input_tokens_seen": 580362904,
      "step": 9680,
      "train_runtime": 124469.1193,
      "train_tokens_per_second": 4662.706
    },
    {
      "epoch": 1.2353316326530612,
      "grad_norm": 0.675406051927215,
      "learning_rate": 3.1944510766604473e-06,
      "loss": 0.2648852109909058,
      "num_input_tokens_seen": 580649824,
      "step": 9685,
      "train_runtime": 124543.1743,
      "train_tokens_per_second": 4662.237
    },
    {
      "epoch": 1.2359693877551021,
      "grad_norm": 0.6516245491378782,
      "learning_rate": 3.1897810517113657e-06,
      "loss": 0.23095903396606446,
      "num_input_tokens_seen": 580943592,
      "step": 9690,
      "train_runtime": 124614.8243,
      "train_tokens_per_second": 4661.914
    },
    {
      "epoch": 1.2366071428571428,
      "grad_norm": 0.6487833784761903,
      "learning_rate": 3.1851128434434263e-06,
      "loss": 0.271923828125,
      "num_input_tokens_seen": 581232392,
      "step": 9695,
      "train_runtime": 124673.8991,
      "train_tokens_per_second": 4662.021
    },
    {
      "epoch": 1.2372448979591837,
      "grad_norm": 0.5968849582919264,
      "learning_rate": 3.1804464565414993e-06,
      "loss": 0.23573975563049315,
      "num_input_tokens_seen": 581531312,
      "step": 9700,
      "train_runtime": 124738.5785,
      "train_tokens_per_second": 4662.0
    },
    {
      "epoch": 1.2378826530612246,
      "grad_norm": 0.5346215637862628,
      "learning_rate": 3.17578189568863e-06,
      "loss": 0.23364415168762206,
      "num_input_tokens_seen": 581835456,
      "step": 9705,
      "train_runtime": 124807.7962,
      "train_tokens_per_second": 4661.852
    },
    {
      "epoch": 1.2385204081632653,
      "grad_norm": 0.5783827962680886,
      "learning_rate": 3.1711191655660327e-06,
      "loss": 0.24957633018493652,
      "num_input_tokens_seen": 582138432,
      "step": 9710,
      "train_runtime": 124870.5723,
      "train_tokens_per_second": 4661.935
    },
    {
      "epoch": 1.2391581632653061,
      "grad_norm": 0.6103049110622445,
      "learning_rate": 3.1664582708530796e-06,
      "loss": 0.23300728797912598,
      "num_input_tokens_seen": 582440344,
      "step": 9715,
      "train_runtime": 124940.0731,
      "train_tokens_per_second": 4661.758
    },
    {
      "epoch": 1.239795918367347,
      "grad_norm": 0.6181695465068429,
      "learning_rate": 3.161799216227307e-06,
      "loss": 0.24103116989135742,
      "num_input_tokens_seen": 582754056,
      "step": 9720,
      "train_runtime": 125004.2274,
      "train_tokens_per_second": 4661.875
    },
    {
      "epoch": 1.2404336734693877,
      "grad_norm": 0.5879967426439205,
      "learning_rate": 3.157142006364401e-06,
      "loss": 0.22790446281433105,
      "num_input_tokens_seen": 583052224,
      "step": 9725,
      "train_runtime": 125059.7885,
      "train_tokens_per_second": 4662.188
    },
    {
      "epoch": 1.2410714285714286,
      "grad_norm": 0.6791849966276937,
      "learning_rate": 3.1524866459381966e-06,
      "loss": 0.2671624183654785,
      "num_input_tokens_seen": 583361360,
      "step": 9730,
      "train_runtime": 125122.1602,
      "train_tokens_per_second": 4662.334
    },
    {
      "epoch": 1.2417091836734695,
      "grad_norm": 0.6784696421808819,
      "learning_rate": 3.147833139620671e-06,
      "loss": 0.2518662452697754,
      "num_input_tokens_seen": 583658360,
      "step": 9735,
      "train_runtime": 125188.8562,
      "train_tokens_per_second": 4662.223
    },
    {
      "epoch": 1.2423469387755102,
      "grad_norm": 0.632734661193053,
      "learning_rate": 3.143181492081947e-06,
      "loss": 0.2688909530639648,
      "num_input_tokens_seen": 583959968,
      "step": 9740,
      "train_runtime": 125251.7817,
      "train_tokens_per_second": 4662.289
    },
    {
      "epoch": 1.242984693877551,
      "grad_norm": 0.6506809875514474,
      "learning_rate": 3.1385317079902743e-06,
      "loss": 0.24842522144317628,
      "num_input_tokens_seen": 584273608,
      "step": 9745,
      "train_runtime": 125319.983,
      "train_tokens_per_second": 4662.254
    },
    {
      "epoch": 1.243622448979592,
      "grad_norm": 0.6404448969075249,
      "learning_rate": 3.133883792012037e-06,
      "loss": 0.23932576179504395,
      "num_input_tokens_seen": 584569088,
      "step": 9750,
      "train_runtime": 125386.4018,
      "train_tokens_per_second": 4662.141
    },
    {
      "epoch": 1.2442602040816326,
      "grad_norm": 0.6622136313376177,
      "learning_rate": 3.1292377488117466e-06,
      "loss": 0.26366333961486815,
      "num_input_tokens_seen": 584886040,
      "step": 9755,
      "train_runtime": 125442.8655,
      "train_tokens_per_second": 4662.569
    },
    {
      "epoch": 1.2448979591836735,
      "grad_norm": 0.617229978107778,
      "learning_rate": 3.124593583052026e-06,
      "loss": 0.2389538288116455,
      "num_input_tokens_seen": 585186440,
      "step": 9760,
      "train_runtime": 125498.6491,
      "train_tokens_per_second": 4662.89
    },
    {
      "epoch": 1.2455357142857142,
      "grad_norm": 0.5766138574003634,
      "learning_rate": 3.119951299393625e-06,
      "loss": 0.2251434564590454,
      "num_input_tokens_seen": 585503184,
      "step": 9765,
      "train_runtime": 125559.3874,
      "train_tokens_per_second": 4663.157
    },
    {
      "epoch": 1.246173469387755,
      "grad_norm": 0.6146959484283944,
      "learning_rate": 3.115310902495396e-06,
      "loss": 0.24179534912109374,
      "num_input_tokens_seen": 585820496,
      "step": 9770,
      "train_runtime": 125606.6063,
      "train_tokens_per_second": 4663.931
    },
    {
      "epoch": 1.246811224489796,
      "grad_norm": 0.6731739894245796,
      "learning_rate": 3.1106723970143016e-06,
      "loss": 0.2561976909637451,
      "num_input_tokens_seen": 586120200,
      "step": 9775,
      "train_runtime": 125674.9063,
      "train_tokens_per_second": 4663.781
    },
    {
      "epoch": 1.2474489795918366,
      "grad_norm": 0.6484276544348772,
      "learning_rate": 3.10603578760541e-06,
      "loss": 0.24519357681274415,
      "num_input_tokens_seen": 586423160,
      "step": 9780,
      "train_runtime": 125730.5778,
      "train_tokens_per_second": 4664.125
    },
    {
      "epoch": 1.2480867346938775,
      "grad_norm": 0.713590031586681,
      "learning_rate": 3.101401078921878e-06,
      "loss": 0.23312654495239257,
      "num_input_tokens_seen": 586721840,
      "step": 9785,
      "train_runtime": 125787.6749,
      "train_tokens_per_second": 4664.383
    },
    {
      "epoch": 1.2487244897959184,
      "grad_norm": 0.8124111308788121,
      "learning_rate": 3.0967682756149607e-06,
      "loss": 0.2430281162261963,
      "num_input_tokens_seen": 587008192,
      "step": 9790,
      "train_runtime": 125845.6817,
      "train_tokens_per_second": 4664.508
    },
    {
      "epoch": 1.249362244897959,
      "grad_norm": 0.7656218296472863,
      "learning_rate": 3.0921373823340028e-06,
      "loss": 0.25926847457885743,
      "num_input_tokens_seen": 587306120,
      "step": 9795,
      "train_runtime": 125907.5491,
      "train_tokens_per_second": 4664.582
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.6525768502255174,
      "learning_rate": 3.0875084037264237e-06,
      "loss": 0.22389230728149415,
      "num_input_tokens_seen": 587616640,
      "step": 9800,
      "train_runtime": 125971.5723,
      "train_tokens_per_second": 4664.677
    },
    {
      "epoch": 1.250637755102041,
      "grad_norm": 0.5941419954954219,
      "learning_rate": 3.0828813444377304e-06,
      "loss": 0.2395017623901367,
      "num_input_tokens_seen": 587929192,
      "step": 9805,
      "train_runtime": 126036.5653,
      "train_tokens_per_second": 4664.751
    },
    {
      "epoch": 1.2512755102040816,
      "grad_norm": 0.6360207891616095,
      "learning_rate": 3.0782562091114997e-06,
      "loss": 0.24663848876953126,
      "num_input_tokens_seen": 588232144,
      "step": 9810,
      "train_runtime": 126113.3717,
      "train_tokens_per_second": 4664.312
    },
    {
      "epoch": 1.2519132653061225,
      "grad_norm": 0.6308596792698608,
      "learning_rate": 3.0736330023893774e-06,
      "loss": 0.2444248914718628,
      "num_input_tokens_seen": 588534920,
      "step": 9815,
      "train_runtime": 126181.2622,
      "train_tokens_per_second": 4664.202
    },
    {
      "epoch": 1.2525510204081631,
      "grad_norm": 0.6706195268447501,
      "learning_rate": 3.0690117289110732e-06,
      "loss": 0.2598260879516602,
      "num_input_tokens_seen": 588838288,
      "step": 9820,
      "train_runtime": 126252.054,
      "train_tokens_per_second": 4663.99
    },
    {
      "epoch": 1.253188775510204,
      "grad_norm": 0.7299696350067582,
      "learning_rate": 3.0643923933143603e-06,
      "loss": 0.2867970705032349,
      "num_input_tokens_seen": 589134928,
      "step": 9825,
      "train_runtime": 126314.8642,
      "train_tokens_per_second": 4664.019
    },
    {
      "epoch": 1.253826530612245,
      "grad_norm": 0.6862653727619418,
      "learning_rate": 3.059775000235063e-06,
      "loss": 0.239058780670166,
      "num_input_tokens_seen": 589430168,
      "step": 9830,
      "train_runtime": 126393.4287,
      "train_tokens_per_second": 4663.456
    },
    {
      "epoch": 1.2544642857142856,
      "grad_norm": 0.6818486572676211,
      "learning_rate": 3.0551595543070567e-06,
      "loss": 0.25689897537231443,
      "num_input_tokens_seen": 589728240,
      "step": 9835,
      "train_runtime": 126468.1225,
      "train_tokens_per_second": 4663.058
    },
    {
      "epoch": 1.2551020408163265,
      "grad_norm": 0.7183256813859021,
      "learning_rate": 3.050546060162267e-06,
      "loss": 0.25381813049316404,
      "num_input_tokens_seen": 590026288,
      "step": 9840,
      "train_runtime": 126538.942,
      "train_tokens_per_second": 4662.804
    },
    {
      "epoch": 1.2557397959183674,
      "grad_norm": 0.6800326696163166,
      "learning_rate": 3.045934522430655e-06,
      "loss": 0.2316105842590332,
      "num_input_tokens_seen": 590321544,
      "step": 9845,
      "train_runtime": 126605.0016,
      "train_tokens_per_second": 4662.703
    },
    {
      "epoch": 1.256377551020408,
      "grad_norm": 0.609686847508778,
      "learning_rate": 3.0413249457402206e-06,
      "loss": 0.25052900314331056,
      "num_input_tokens_seen": 590609080,
      "step": 9850,
      "train_runtime": 126656.5957,
      "train_tokens_per_second": 4663.074
    },
    {
      "epoch": 1.257015306122449,
      "grad_norm": 0.6319635813899378,
      "learning_rate": 3.0367173347169978e-06,
      "loss": 0.24621901512145997,
      "num_input_tokens_seen": 590900952,
      "step": 9855,
      "train_runtime": 126723.2199,
      "train_tokens_per_second": 4662.926
    },
    {
      "epoch": 1.2576530612244898,
      "grad_norm": 0.6273263453820251,
      "learning_rate": 3.0321116939850435e-06,
      "loss": 0.23445494174957277,
      "num_input_tokens_seen": 591213904,
      "step": 9860,
      "train_runtime": 126789.5774,
      "train_tokens_per_second": 4662.954
    },
    {
      "epoch": 1.2582908163265305,
      "grad_norm": 0.63648803262082,
      "learning_rate": 3.0275080281664417e-06,
      "loss": 0.24305615425109864,
      "num_input_tokens_seen": 591498472,
      "step": 9865,
      "train_runtime": 126859.9434,
      "train_tokens_per_second": 4662.61
    },
    {
      "epoch": 1.2589285714285714,
      "grad_norm": 0.7284924231470784,
      "learning_rate": 3.0229063418812932e-06,
      "loss": 0.26299324035644533,
      "num_input_tokens_seen": 591797136,
      "step": 9870,
      "train_runtime": 126931.7558,
      "train_tokens_per_second": 4662.325
    },
    {
      "epoch": 1.2595663265306123,
      "grad_norm": 0.6795771148313055,
      "learning_rate": 3.0183066397477103e-06,
      "loss": 0.25070414543151853,
      "num_input_tokens_seen": 592087736,
      "step": 9875,
      "train_runtime": 126997.2662,
      "train_tokens_per_second": 4662.209
    },
    {
      "epoch": 1.260204081632653,
      "grad_norm": 0.6313372192986613,
      "learning_rate": 3.013708926381814e-06,
      "loss": 0.28244891166687014,
      "num_input_tokens_seen": 592391680,
      "step": 9880,
      "train_runtime": 127070.1969,
      "train_tokens_per_second": 4661.925
    },
    {
      "epoch": 1.2608418367346939,
      "grad_norm": 0.695641878360283,
      "learning_rate": 3.0091132063977344e-06,
      "loss": 0.2593372821807861,
      "num_input_tokens_seen": 592690432,
      "step": 9885,
      "train_runtime": 127142.576,
      "train_tokens_per_second": 4661.62
    },
    {
      "epoch": 1.2614795918367347,
      "grad_norm": 0.582783347348586,
      "learning_rate": 3.004519484407594e-06,
      "loss": 0.21239848136901857,
      "num_input_tokens_seen": 592998984,
      "step": 9890,
      "train_runtime": 127206.6827,
      "train_tokens_per_second": 4661.697
    },
    {
      "epoch": 1.2621173469387754,
      "grad_norm": 0.6447943995639185,
      "learning_rate": 2.9999277650215164e-06,
      "loss": 0.2596427917480469,
      "num_input_tokens_seen": 593285784,
      "step": 9895,
      "train_runtime": 127276.7131,
      "train_tokens_per_second": 4661.385
    },
    {
      "epoch": 1.2627551020408163,
      "grad_norm": 0.5925802211805856,
      "learning_rate": 2.9953380528476105e-06,
      "loss": 0.23733255863189698,
      "num_input_tokens_seen": 593594176,
      "step": 9900,
      "train_runtime": 127340.1794,
      "train_tokens_per_second": 4661.484
    },
    {
      "epoch": 1.2633928571428572,
      "grad_norm": 0.6552263140926321,
      "learning_rate": 2.9907503524919734e-06,
      "loss": 0.24280138015747071,
      "num_input_tokens_seen": 593895328,
      "step": 9905,
      "train_runtime": 127408.2113,
      "train_tokens_per_second": 4661.358
    },
    {
      "epoch": 1.2640306122448979,
      "grad_norm": 0.5794674407213174,
      "learning_rate": 2.9861646685586843e-06,
      "loss": 0.2521806001663208,
      "num_input_tokens_seen": 594176032,
      "step": 9910,
      "train_runtime": 127469.668,
      "train_tokens_per_second": 4661.313
    },
    {
      "epoch": 1.2646683673469388,
      "grad_norm": 0.6096879784672814,
      "learning_rate": 2.981581005649795e-06,
      "loss": 0.23649122714996337,
      "num_input_tokens_seen": 594480448,
      "step": 9915,
      "train_runtime": 127537.1602,
      "train_tokens_per_second": 4661.233
    },
    {
      "epoch": 1.2653061224489797,
      "grad_norm": 0.5472430801576244,
      "learning_rate": 2.9769993683653324e-06,
      "loss": 0.23223299980163575,
      "num_input_tokens_seen": 594783312,
      "step": 9920,
      "train_runtime": 127595.4696,
      "train_tokens_per_second": 4661.477
    },
    {
      "epoch": 1.2659438775510203,
      "grad_norm": 0.7011267883947306,
      "learning_rate": 2.9724197613032917e-06,
      "loss": 0.243508243560791,
      "num_input_tokens_seen": 595082792,
      "step": 9925,
      "train_runtime": 127663.453,
      "train_tokens_per_second": 4661.34
    },
    {
      "epoch": 1.2665816326530612,
      "grad_norm": 0.697210984126257,
      "learning_rate": 2.9678421890596254e-06,
      "loss": 0.25273904800415037,
      "num_input_tokens_seen": 595365824,
      "step": 9930,
      "train_runtime": 127724.9957,
      "train_tokens_per_second": 4661.31
    },
    {
      "epoch": 1.2672193877551021,
      "grad_norm": 0.667946641017842,
      "learning_rate": 2.9632666562282477e-06,
      "loss": 0.24409079551696777,
      "num_input_tokens_seen": 595664960,
      "step": 9935,
      "train_runtime": 127777.4952,
      "train_tokens_per_second": 4661.736
    },
    {
      "epoch": 1.2678571428571428,
      "grad_norm": 0.5915450005846132,
      "learning_rate": 2.9586931674010265e-06,
      "loss": 0.23822576999664308,
      "num_input_tokens_seen": 595952024,
      "step": 9940,
      "train_runtime": 127841.3064,
      "train_tokens_per_second": 4661.655
    },
    {
      "epoch": 1.2684948979591837,
      "grad_norm": 0.6275940897430338,
      "learning_rate": 2.9541217271677747e-06,
      "loss": 0.26365227699279786,
      "num_input_tokens_seen": 596253880,
      "step": 9945,
      "train_runtime": 127906.4577,
      "train_tokens_per_second": 4661.64
    },
    {
      "epoch": 1.2691326530612246,
      "grad_norm": 0.728477385611455,
      "learning_rate": 2.949552340116254e-06,
      "loss": 0.2271937370300293,
      "num_input_tokens_seen": 596557368,
      "step": 9950,
      "train_runtime": 127975.1504,
      "train_tokens_per_second": 4661.509
    },
    {
      "epoch": 1.2697704081632653,
      "grad_norm": 0.6084134412677387,
      "learning_rate": 2.9449850108321644e-06,
      "loss": 0.25845656394958494,
      "num_input_tokens_seen": 596855520,
      "step": 9955,
      "train_runtime": 128050.2822,
      "train_tokens_per_second": 4661.103
    },
    {
      "epoch": 1.2704081632653061,
      "grad_norm": 0.7342224938631051,
      "learning_rate": 2.9404197438991377e-06,
      "loss": 0.23968310356140138,
      "num_input_tokens_seen": 597154400,
      "step": 9960,
      "train_runtime": 128101.07,
      "train_tokens_per_second": 4661.588
    },
    {
      "epoch": 1.271045918367347,
      "grad_norm": 0.679852672100427,
      "learning_rate": 2.935856543898737e-06,
      "loss": 0.2415492057800293,
      "num_input_tokens_seen": 597437920,
      "step": 9965,
      "train_runtime": 128168.4646,
      "train_tokens_per_second": 4661.349
    },
    {
      "epoch": 1.2716836734693877,
      "grad_norm": 0.597171871665114,
      "learning_rate": 2.9312954154104563e-06,
      "loss": 0.23209569454193116,
      "num_input_tokens_seen": 597733832,
      "step": 9970,
      "train_runtime": 128226.6002,
      "train_tokens_per_second": 4661.543
    },
    {
      "epoch": 1.2723214285714286,
      "grad_norm": 0.6379672125616391,
      "learning_rate": 2.9267363630117028e-06,
      "loss": 0.24382576942443848,
      "num_input_tokens_seen": 598036272,
      "step": 9975,
      "train_runtime": 128296.5862,
      "train_tokens_per_second": 4661.358
    },
    {
      "epoch": 1.2729591836734695,
      "grad_norm": 0.6247185193777421,
      "learning_rate": 2.9221793912778053e-06,
      "loss": 0.248250150680542,
      "num_input_tokens_seen": 598344848,
      "step": 9980,
      "train_runtime": 128360.3081,
      "train_tokens_per_second": 4661.448
    },
    {
      "epoch": 1.2735969387755102,
      "grad_norm": 0.700121315011332,
      "learning_rate": 2.9176245047820064e-06,
      "loss": 0.2316591739654541,
      "num_input_tokens_seen": 598635984,
      "step": 9985,
      "train_runtime": 128425.9021,
      "train_tokens_per_second": 4661.334
    },
    {
      "epoch": 1.274234693877551,
      "grad_norm": 0.67605563420079,
      "learning_rate": 2.913071708095446e-06,
      "loss": 0.246818208694458,
      "num_input_tokens_seen": 598932032,
      "step": 9990,
      "train_runtime": 128490.184,
      "train_tokens_per_second": 4661.306
    },
    {
      "epoch": 1.274872448979592,
      "grad_norm": 0.6372114225203562,
      "learning_rate": 2.908521005787177e-06,
      "loss": 0.2229018211364746,
      "num_input_tokens_seen": 599238504,
      "step": 9995,
      "train_runtime": 128547.5302,
      "train_tokens_per_second": 4661.61
    },
    {
      "epoch": 1.2755102040816326,
      "grad_norm": 0.6774852915168257,
      "learning_rate": 2.903972402424151e-06,
      "loss": 0.24406712055206298,
      "num_input_tokens_seen": 599539168,
      "step": 10000,
      "train_runtime": 128600.7006,
      "train_tokens_per_second": 4662.021
    },
    {
      "epoch": 1.2761479591836735,
      "grad_norm": 0.6895370298799267,
      "learning_rate": 2.899425902571203e-06,
      "loss": 0.2546832084655762,
      "num_input_tokens_seen": 599831168,
      "step": 10005,
      "train_runtime": 128664.2372,
      "train_tokens_per_second": 4661.988
    },
    {
      "epoch": 1.2767857142857144,
      "grad_norm": 0.670430839752296,
      "learning_rate": 2.8948815107910653e-06,
      "loss": 0.24214091300964355,
      "num_input_tokens_seen": 600126312,
      "step": 10010,
      "train_runtime": 128721.48,
      "train_tokens_per_second": 4662.208
    },
    {
      "epoch": 1.277423469387755,
      "grad_norm": 0.5791025050862594,
      "learning_rate": 2.8903392316443534e-06,
      "loss": 0.2257810115814209,
      "num_input_tokens_seen": 600428096,
      "step": 10015,
      "train_runtime": 128785.2796,
      "train_tokens_per_second": 4662.242
    },
    {
      "epoch": 1.278061224489796,
      "grad_norm": 0.6670510559237663,
      "learning_rate": 2.885799069689561e-06,
      "loss": 0.2611087322235107,
      "num_input_tokens_seen": 600732176,
      "step": 10020,
      "train_runtime": 128856.5089,
      "train_tokens_per_second": 4662.024
    },
    {
      "epoch": 1.2786989795918369,
      "grad_norm": 0.6567534118948043,
      "learning_rate": 2.8812610294830568e-06,
      "loss": 0.2763176202774048,
      "num_input_tokens_seen": 601035232,
      "step": 10025,
      "train_runtime": 128917.9532,
      "train_tokens_per_second": 4662.153
    },
    {
      "epoch": 1.2793367346938775,
      "grad_norm": 0.6770265112304864,
      "learning_rate": 2.8767251155790826e-06,
      "loss": 0.24007635116577147,
      "num_input_tokens_seen": 601336112,
      "step": 10030,
      "train_runtime": 128981.0975,
      "train_tokens_per_second": 4662.203
    },
    {
      "epoch": 1.2799744897959184,
      "grad_norm": 0.6954544127397374,
      "learning_rate": 2.8721913325297434e-06,
      "loss": 0.23859257698059083,
      "num_input_tokens_seen": 601621800,
      "step": 10035,
      "train_runtime": 129045.5017,
      "train_tokens_per_second": 4662.09
    },
    {
      "epoch": 1.280612244897959,
      "grad_norm": 0.669167366755921,
      "learning_rate": 2.8676596848850084e-06,
      "loss": 0.2485593318939209,
      "num_input_tokens_seen": 601930960,
      "step": 10040,
      "train_runtime": 129104.7849,
      "train_tokens_per_second": 4662.344
    },
    {
      "epoch": 1.28125,
      "grad_norm": 0.6860955602793227,
      "learning_rate": 2.8631301771927024e-06,
      "loss": 0.2497274398803711,
      "num_input_tokens_seen": 602225936,
      "step": 10045,
      "train_runtime": 129164.441,
      "train_tokens_per_second": 4662.475
    },
    {
      "epoch": 1.281887755102041,
      "grad_norm": 0.6278297822403494,
      "learning_rate": 2.858602813998501e-06,
      "loss": 0.2408839702606201,
      "num_input_tokens_seen": 602528808,
      "step": 10050,
      "train_runtime": 129226.8313,
      "train_tokens_per_second": 4662.567
    },
    {
      "epoch": 1.2825255102040816,
      "grad_norm": 0.7195956124045154,
      "learning_rate": 2.8540775998459336e-06,
      "loss": 0.24351954460144043,
      "num_input_tokens_seen": 602812944,
      "step": 10055,
      "train_runtime": 129284.2969,
      "train_tokens_per_second": 4662.693
    },
    {
      "epoch": 1.2831632653061225,
      "grad_norm": 0.637166229994117,
      "learning_rate": 2.849554539276361e-06,
      "loss": 0.25557701587677,
      "num_input_tokens_seen": 603111072,
      "step": 10060,
      "train_runtime": 129340.9474,
      "train_tokens_per_second": 4662.955
    },
    {
      "epoch": 1.2838010204081631,
      "grad_norm": 0.6242924195871741,
      "learning_rate": 2.845033636828998e-06,
      "loss": 0.27079105377197266,
      "num_input_tokens_seen": 603416472,
      "step": 10065,
      "train_runtime": 129408.2548,
      "train_tokens_per_second": 4662.89
    },
    {
      "epoch": 1.284438775510204,
      "grad_norm": 0.6355558778919869,
      "learning_rate": 2.840514897040884e-06,
      "loss": 0.24806299209594726,
      "num_input_tokens_seen": 603706776,
      "step": 10070,
      "train_runtime": 129463.2937,
      "train_tokens_per_second": 4663.15
    },
    {
      "epoch": 1.285076530612245,
      "grad_norm": 0.6190180007774851,
      "learning_rate": 2.835998324446887e-06,
      "loss": 0.25780978202819826,
      "num_input_tokens_seen": 603991600,
      "step": 10075,
      "train_runtime": 129537.7133,
      "train_tokens_per_second": 4662.67
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 0.68576106924587,
      "learning_rate": 2.831483923579703e-06,
      "loss": 0.2651859521865845,
      "num_input_tokens_seen": 604272496,
      "step": 10080,
      "train_runtime": 129600.5575,
      "train_tokens_per_second": 4662.576
    },
    {
      "epoch": 1.2863520408163265,
      "grad_norm": 0.7292708466163627,
      "learning_rate": 2.8269716989698547e-06,
      "loss": 0.23357632160186767,
      "num_input_tokens_seen": 604576032,
      "step": 10085,
      "train_runtime": 129655.3566,
      "train_tokens_per_second": 4662.947
    },
    {
      "epoch": 1.2869897959183674,
      "grad_norm": 0.7127516238098663,
      "learning_rate": 2.822461655145669e-06,
      "loss": 0.23446400165557862,
      "num_input_tokens_seen": 604883880,
      "step": 10090,
      "train_runtime": 129725.2588,
      "train_tokens_per_second": 4662.807
    },
    {
      "epoch": 1.287627551020408,
      "grad_norm": 0.6253778520629037,
      "learning_rate": 2.817953796633289e-06,
      "loss": 0.2427058696746826,
      "num_input_tokens_seen": 605182136,
      "step": 10095,
      "train_runtime": 129781.7084,
      "train_tokens_per_second": 4663.077
    },
    {
      "epoch": 1.288265306122449,
      "grad_norm": 0.6655998352951639,
      "learning_rate": 2.813448127956672e-06,
      "loss": 0.2528971195220947,
      "num_input_tokens_seen": 605487416,
      "step": 10100,
      "train_runtime": 129855.7841,
      "train_tokens_per_second": 4662.768
    },
    {
      "epoch": 1.2889030612244898,
      "grad_norm": 0.6268315385907461,
      "learning_rate": 2.8089446536375642e-06,
      "loss": 0.2563582181930542,
      "num_input_tokens_seen": 605790784,
      "step": 10105,
      "train_runtime": 129915.9073,
      "train_tokens_per_second": 4662.945
    },
    {
      "epoch": 1.2895408163265305,
      "grad_norm": 0.6868501808514035,
      "learning_rate": 2.8044433781955196e-06,
      "loss": 0.2638521671295166,
      "num_input_tokens_seen": 606089200,
      "step": 10110,
      "train_runtime": 129978.0063,
      "train_tokens_per_second": 4663.014
    },
    {
      "epoch": 1.2901785714285714,
      "grad_norm": 0.6533381998889802,
      "learning_rate": 2.7999443061478827e-06,
      "loss": 0.2558642148971558,
      "num_input_tokens_seen": 606392880,
      "step": 10115,
      "train_runtime": 130040.8954,
      "train_tokens_per_second": 4663.094
    },
    {
      "epoch": 1.2908163265306123,
      "grad_norm": 0.603724501405975,
      "learning_rate": 2.795447442009786e-06,
      "loss": 0.2480170249938965,
      "num_input_tokens_seen": 606707200,
      "step": 10120,
      "train_runtime": 130108.2241,
      "train_tokens_per_second": 4663.096
    },
    {
      "epoch": 1.291454081632653,
      "grad_norm": 0.6253420037521669,
      "learning_rate": 2.7909527902941467e-06,
      "loss": 0.2574573516845703,
      "num_input_tokens_seen": 607015712,
      "step": 10125,
      "train_runtime": 130180.4124,
      "train_tokens_per_second": 4662.881
    },
    {
      "epoch": 1.2920918367346939,
      "grad_norm": 0.6389694583553371,
      "learning_rate": 2.786460355511662e-06,
      "loss": 0.23451039791107178,
      "num_input_tokens_seen": 607325304,
      "step": 10130,
      "train_runtime": 130241.697,
      "train_tokens_per_second": 4663.064
    },
    {
      "epoch": 1.2927295918367347,
      "grad_norm": 0.6737598501065684,
      "learning_rate": 2.7819701421708047e-06,
      "loss": 0.2471921443939209,
      "num_input_tokens_seen": 607629136,
      "step": 10135,
      "train_runtime": 130299.5059,
      "train_tokens_per_second": 4663.326
    },
    {
      "epoch": 1.2933673469387754,
      "grad_norm": 0.6231189021705574,
      "learning_rate": 2.7774821547778176e-06,
      "loss": 0.240172815322876,
      "num_input_tokens_seen": 607925944,
      "step": 10140,
      "train_runtime": 130358.8521,
      "train_tokens_per_second": 4663.48
    },
    {
      "epoch": 1.2940051020408163,
      "grad_norm": 0.6042268538241072,
      "learning_rate": 2.772996397836704e-06,
      "loss": 0.23002943992614747,
      "num_input_tokens_seen": 608223488,
      "step": 10145,
      "train_runtime": 130429.0103,
      "train_tokens_per_second": 4663.253
    },
    {
      "epoch": 1.2946428571428572,
      "grad_norm": 0.6317798878311383,
      "learning_rate": 2.768512875849242e-06,
      "loss": 0.23823139667510987,
      "num_input_tokens_seen": 608519144,
      "step": 10150,
      "train_runtime": 130489.7291,
      "train_tokens_per_second": 4663.349
    },
    {
      "epoch": 1.2952806122448979,
      "grad_norm": 0.6300964049422993,
      "learning_rate": 2.7640315933149568e-06,
      "loss": 0.24309196472167968,
      "num_input_tokens_seen": 608803176,
      "step": 10155,
      "train_runtime": 130546.3195,
      "train_tokens_per_second": 4663.503
    },
    {
      "epoch": 1.2959183673469388,
      "grad_norm": 0.6491467073046057,
      "learning_rate": 2.7595525547311257e-06,
      "loss": 0.23561034202575684,
      "num_input_tokens_seen": 609090424,
      "step": 10160,
      "train_runtime": 130612.8179,
      "train_tokens_per_second": 4663.328
    },
    {
      "epoch": 1.2965561224489797,
      "grad_norm": 0.6629929111370925,
      "learning_rate": 2.755075764592777e-06,
      "loss": 0.24338645935058595,
      "num_input_tokens_seen": 609398400,
      "step": 10165,
      "train_runtime": 130659.6588,
      "train_tokens_per_second": 4664.013
    },
    {
      "epoch": 1.2971938775510203,
      "grad_norm": 0.6547865944820238,
      "learning_rate": 2.750601227392687e-06,
      "loss": 0.24423551559448242,
      "num_input_tokens_seen": 609687192,
      "step": 10170,
      "train_runtime": 130713.6987,
      "train_tokens_per_second": 4664.295
    },
    {
      "epoch": 1.2978316326530612,
      "grad_norm": 0.6243956844176494,
      "learning_rate": 2.7461289476213617e-06,
      "loss": 0.2264718532562256,
      "num_input_tokens_seen": 609985088,
      "step": 10175,
      "train_runtime": 130779.8484,
      "train_tokens_per_second": 4664.213
    },
    {
      "epoch": 1.2984693877551021,
      "grad_norm": 0.631937100184549,
      "learning_rate": 2.741658929767045e-06,
      "loss": 0.22623286247253419,
      "num_input_tokens_seen": 610288152,
      "step": 10180,
      "train_runtime": 130847.9555,
      "train_tokens_per_second": 4664.102
    },
    {
      "epoch": 1.2991071428571428,
      "grad_norm": 0.6201612686743121,
      "learning_rate": 2.7371911783157178e-06,
      "loss": 0.22912931442260742,
      "num_input_tokens_seen": 610592400,
      "step": 10185,
      "train_runtime": 130905.1217,
      "train_tokens_per_second": 4664.389
    },
    {
      "epoch": 1.2997448979591837,
      "grad_norm": 0.5935421690630087,
      "learning_rate": 2.7327256977510764e-06,
      "loss": 0.24432737827301027,
      "num_input_tokens_seen": 610905200,
      "step": 10190,
      "train_runtime": 130961.7409,
      "train_tokens_per_second": 4664.761
    },
    {
      "epoch": 1.3003826530612246,
      "grad_norm": 0.7009681464172817,
      "learning_rate": 2.7282624925545432e-06,
      "loss": 0.22099885940551758,
      "num_input_tokens_seen": 611203784,
      "step": 10195,
      "train_runtime": 131019.5135,
      "train_tokens_per_second": 4664.983
    },
    {
      "epoch": 1.3010204081632653,
      "grad_norm": 0.6131125029672629,
      "learning_rate": 2.723801567205256e-06,
      "loss": 0.22445275783538818,
      "num_input_tokens_seen": 611490048,
      "step": 10200,
      "train_runtime": 131088.2208,
      "train_tokens_per_second": 4664.722
    },
    {
      "epoch": 1.3016581632653061,
      "grad_norm": 0.7160100618511935,
      "learning_rate": 2.7193429261800657e-06,
      "loss": 0.24154410362243653,
      "num_input_tokens_seen": 611776816,
      "step": 10205,
      "train_runtime": 131150.4569,
      "train_tokens_per_second": 4664.695
    },
    {
      "epoch": 1.302295918367347,
      "grad_norm": 0.6989238755659404,
      "learning_rate": 2.7148865739535303e-06,
      "loss": 0.20784764289855956,
      "num_input_tokens_seen": 612075208,
      "step": 10210,
      "train_runtime": 131213.6782,
      "train_tokens_per_second": 4664.721
    },
    {
      "epoch": 1.3029336734693877,
      "grad_norm": 0.6930432977491332,
      "learning_rate": 2.710432514997909e-06,
      "loss": 0.24748585224151612,
      "num_input_tokens_seen": 612367480,
      "step": 10215,
      "train_runtime": 131276.4751,
      "train_tokens_per_second": 4664.716
    },
    {
      "epoch": 1.3035714285714286,
      "grad_norm": 0.5908520343258692,
      "learning_rate": 2.7059807537831624e-06,
      "loss": 0.2211530923843384,
      "num_input_tokens_seen": 612670920,
      "step": 10220,
      "train_runtime": 131344.1066,
      "train_tokens_per_second": 4664.624
    },
    {
      "epoch": 1.3042091836734695,
      "grad_norm": 0.6649156510386195,
      "learning_rate": 2.7015312947769436e-06,
      "loss": 0.25975823402404785,
      "num_input_tokens_seen": 612973912,
      "step": 10225,
      "train_runtime": 131405.1473,
      "train_tokens_per_second": 4664.763
    },
    {
      "epoch": 1.3048469387755102,
      "grad_norm": 0.6940272303736259,
      "learning_rate": 2.6970841424445947e-06,
      "loss": 0.2528405666351318,
      "num_input_tokens_seen": 613260264,
      "step": 10230,
      "train_runtime": 131479.6405,
      "train_tokens_per_second": 4664.298
    },
    {
      "epoch": 1.305484693877551,
      "grad_norm": 0.6716976167451483,
      "learning_rate": 2.692639301249144e-06,
      "loss": 0.23514809608459472,
      "num_input_tokens_seen": 613563136,
      "step": 10235,
      "train_runtime": 131550.143,
      "train_tokens_per_second": 4664.101
    },
    {
      "epoch": 1.306122448979592,
      "grad_norm": 0.6638311072307366,
      "learning_rate": 2.6881967756513004e-06,
      "loss": 0.22921185493469237,
      "num_input_tokens_seen": 613870040,
      "step": 10240,
      "train_runtime": 131619.2851,
      "train_tokens_per_second": 4663.982
    },
    {
      "epoch": 1.3067602040816326,
      "grad_norm": 0.6539008630229388,
      "learning_rate": 2.6837565701094505e-06,
      "loss": 0.24595980644226073,
      "num_input_tokens_seen": 614171488,
      "step": 10245,
      "train_runtime": 131690.4363,
      "train_tokens_per_second": 4663.752
    },
    {
      "epoch": 1.3073979591836735,
      "grad_norm": 0.6766006041244161,
      "learning_rate": 2.679318689079645e-06,
      "loss": 0.23799505233764648,
      "num_input_tokens_seen": 614454400,
      "step": 10250,
      "train_runtime": 131756.9971,
      "train_tokens_per_second": 4663.543
    },
    {
      "epoch": 1.3080357142857144,
      "grad_norm": 0.7066404687080433,
      "learning_rate": 2.6748831370156148e-06,
      "loss": 0.22920284271240235,
      "num_input_tokens_seen": 614741568,
      "step": 10255,
      "train_runtime": 131830.9757,
      "train_tokens_per_second": 4663.104
    },
    {
      "epoch": 1.308673469387755,
      "grad_norm": 0.677234380061771,
      "learning_rate": 2.6704499183687404e-06,
      "loss": 0.24149069786071778,
      "num_input_tokens_seen": 615034424,
      "step": 10260,
      "train_runtime": 131902.8326,
      "train_tokens_per_second": 4662.784
    },
    {
      "epoch": 1.309311224489796,
      "grad_norm": 0.6132148524381859,
      "learning_rate": 2.6660190375880656e-06,
      "loss": 0.26289279460906984,
      "num_input_tokens_seen": 615344688,
      "step": 10265,
      "train_runtime": 131975.4333,
      "train_tokens_per_second": 4662.57
    },
    {
      "epoch": 1.3099489795918369,
      "grad_norm": 0.9180387390052087,
      "learning_rate": 2.661590499120296e-06,
      "loss": 0.24758877754211425,
      "num_input_tokens_seen": 615641624,
      "step": 10270,
      "train_runtime": 132029.5093,
      "train_tokens_per_second": 4662.909
    },
    {
      "epoch": 1.3105867346938775,
      "grad_norm": 0.6479016863042246,
      "learning_rate": 2.657164307409773e-06,
      "loss": 0.2354414939880371,
      "num_input_tokens_seen": 615933960,
      "step": 10275,
      "train_runtime": 132082.9803,
      "train_tokens_per_second": 4663.235
    },
    {
      "epoch": 1.3112244897959184,
      "grad_norm": 0.5807944678730947,
      "learning_rate": 2.6527404668984907e-06,
      "loss": 0.23063254356384277,
      "num_input_tokens_seen": 616234048,
      "step": 10280,
      "train_runtime": 132150.066,
      "train_tokens_per_second": 4663.138
    },
    {
      "epoch": 1.311862244897959,
      "grad_norm": 0.7136028083098015,
      "learning_rate": 2.648318982026082e-06,
      "loss": 0.2652731418609619,
      "num_input_tokens_seen": 616536328,
      "step": 10285,
      "train_runtime": 132224.6383,
      "train_tokens_per_second": 4662.795
    },
    {
      "epoch": 1.3125,
      "grad_norm": 0.6356233393225613,
      "learning_rate": 2.6438998572298157e-06,
      "loss": 0.25313777923583985,
      "num_input_tokens_seen": 616850912,
      "step": 10290,
      "train_runtime": 132293.9362,
      "train_tokens_per_second": 4662.73
    },
    {
      "epoch": 1.313137755102041,
      "grad_norm": 0.6442941013212176,
      "learning_rate": 2.639483096944593e-06,
      "loss": 0.25389604568481444,
      "num_input_tokens_seen": 617141152,
      "step": 10295,
      "train_runtime": 132361.2256,
      "train_tokens_per_second": 4662.552
    },
    {
      "epoch": 1.3137755102040816,
      "grad_norm": 0.6479525122954625,
      "learning_rate": 2.63506870560294e-06,
      "loss": 0.249885892868042,
      "num_input_tokens_seen": 617437848,
      "step": 10300,
      "train_runtime": 132425.0939,
      "train_tokens_per_second": 4662.544
    },
    {
      "epoch": 1.3144132653061225,
      "grad_norm": 0.5994254511342957,
      "learning_rate": 2.6306566876350072e-06,
      "loss": 0.24343361854553222,
      "num_input_tokens_seen": 617729280,
      "step": 10305,
      "train_runtime": 132492.0753,
      "train_tokens_per_second": 4662.387
    },
    {
      "epoch": 1.3150510204081631,
      "grad_norm": 0.7121766419687681,
      "learning_rate": 2.626247047468563e-06,
      "loss": 0.24100232124328613,
      "num_input_tokens_seen": 618037360,
      "step": 10310,
      "train_runtime": 132548.4205,
      "train_tokens_per_second": 4662.729
    },
    {
      "epoch": 1.315688775510204,
      "grad_norm": 0.6699447714105571,
      "learning_rate": 2.6218397895289893e-06,
      "loss": 0.25350179672241213,
      "num_input_tokens_seen": 618337752,
      "step": 10315,
      "train_runtime": 132611.4158,
      "train_tokens_per_second": 4662.779
    },
    {
      "epoch": 1.316326530612245,
      "grad_norm": 0.7024153498979675,
      "learning_rate": 2.617434918239278e-06,
      "loss": 0.2486175775527954,
      "num_input_tokens_seen": 618618520,
      "step": 10320,
      "train_runtime": 132681.8293,
      "train_tokens_per_second": 4662.421
    },
    {
      "epoch": 1.3169642857142856,
      "grad_norm": 0.6137446766230643,
      "learning_rate": 2.6130324380200234e-06,
      "loss": 0.22817761898040773,
      "num_input_tokens_seen": 618916832,
      "step": 10325,
      "train_runtime": 132748.6683,
      "train_tokens_per_second": 4662.32
    },
    {
      "epoch": 1.3176020408163265,
      "grad_norm": 0.7594642166546341,
      "learning_rate": 2.6086323532894225e-06,
      "loss": 0.23662309646606444,
      "num_input_tokens_seen": 619203704,
      "step": 10330,
      "train_runtime": 132813.1886,
      "train_tokens_per_second": 4662.215
    },
    {
      "epoch": 1.3182397959183674,
      "grad_norm": 0.5828183887503053,
      "learning_rate": 2.604234668463268e-06,
      "loss": 0.2248455047607422,
      "num_input_tokens_seen": 619499320,
      "step": 10335,
      "train_runtime": 132877.5761,
      "train_tokens_per_second": 4662.181
    },
    {
      "epoch": 1.318877551020408,
      "grad_norm": 0.6646006757856918,
      "learning_rate": 2.5998393879549444e-06,
      "loss": 0.2405029296875,
      "num_input_tokens_seen": 619778688,
      "step": 10340,
      "train_runtime": 132946.09,
      "train_tokens_per_second": 4661.88
    },
    {
      "epoch": 1.319515306122449,
      "grad_norm": 0.7006720415594211,
      "learning_rate": 2.5954465161754226e-06,
      "loss": 0.25348052978515623,
      "num_input_tokens_seen": 620077992,
      "step": 10345,
      "train_runtime": 133004.7128,
      "train_tokens_per_second": 4662.075
    },
    {
      "epoch": 1.3201530612244898,
      "grad_norm": 0.6607904031477573,
      "learning_rate": 2.5910560575332524e-06,
      "loss": 0.2627995491027832,
      "num_input_tokens_seen": 620382504,
      "step": 10350,
      "train_runtime": 133074.7034,
      "train_tokens_per_second": 4661.912
    },
    {
      "epoch": 1.3207908163265305,
      "grad_norm": 0.6634275943477552,
      "learning_rate": 2.5866680164345704e-06,
      "loss": 0.24571120738983154,
      "num_input_tokens_seen": 620684224,
      "step": 10355,
      "train_runtime": 133150.8163,
      "train_tokens_per_second": 4661.513
    },
    {
      "epoch": 1.3214285714285714,
      "grad_norm": 0.6313433861607889,
      "learning_rate": 2.5822823972830825e-06,
      "loss": 0.2279362201690674,
      "num_input_tokens_seen": 620975264,
      "step": 10360,
      "train_runtime": 133221.831,
      "train_tokens_per_second": 4661.212
    },
    {
      "epoch": 1.3220663265306123,
      "grad_norm": 0.6906529316647421,
      "learning_rate": 2.5778992044800597e-06,
      "loss": 0.2523262739181519,
      "num_input_tokens_seen": 621291808,
      "step": 10365,
      "train_runtime": 133291.0334,
      "train_tokens_per_second": 4661.167
    },
    {
      "epoch": 1.322704081632653,
      "grad_norm": 0.6690653556788901,
      "learning_rate": 2.573518442424343e-06,
      "loss": 0.24666938781738282,
      "num_input_tokens_seen": 621598720,
      "step": 10370,
      "train_runtime": 133350.6628,
      "train_tokens_per_second": 4661.385
    },
    {
      "epoch": 1.3233418367346939,
      "grad_norm": 0.6196881118927071,
      "learning_rate": 2.569140115512338e-06,
      "loss": 0.2609063148498535,
      "num_input_tokens_seen": 621905856,
      "step": 10375,
      "train_runtime": 133415.4702,
      "train_tokens_per_second": 4661.422
    },
    {
      "epoch": 1.3239795918367347,
      "grad_norm": 0.6112347227065951,
      "learning_rate": 2.564764228137996e-06,
      "loss": 0.26653361320495605,
      "num_input_tokens_seen": 622199408,
      "step": 10380,
      "train_runtime": 133478.0955,
      "train_tokens_per_second": 4661.435
    },
    {
      "epoch": 1.3246173469387754,
      "grad_norm": 0.7720781550947031,
      "learning_rate": 2.5603907846928277e-06,
      "loss": 0.2378302812576294,
      "num_input_tokens_seen": 622480856,
      "step": 10385,
      "train_runtime": 133546.1368,
      "train_tokens_per_second": 4661.167
    },
    {
      "epoch": 1.3252551020408163,
      "grad_norm": 0.6975726873758948,
      "learning_rate": 2.556019789565889e-06,
      "loss": 0.2469649314880371,
      "num_input_tokens_seen": 622788032,
      "step": 10390,
      "train_runtime": 133613.1506,
      "train_tokens_per_second": 4661.128
    },
    {
      "epoch": 1.3258928571428572,
      "grad_norm": 0.689804716227359,
      "learning_rate": 2.5516512471437783e-06,
      "loss": 0.23420460224151612,
      "num_input_tokens_seen": 623092288,
      "step": 10395,
      "train_runtime": 133674.9404,
      "train_tokens_per_second": 4661.25
    },
    {
      "epoch": 1.3265306122448979,
      "grad_norm": 0.6182418828390731,
      "learning_rate": 2.547285161810634e-06,
      "loss": 0.24218957424163817,
      "num_input_tokens_seen": 623398424,
      "step": 10400,
      "train_runtime": 133738.6364,
      "train_tokens_per_second": 4661.319
    },
    {
      "epoch": 1.3271683673469388,
      "grad_norm": 0.5704768998072005,
      "learning_rate": 2.5429215379481267e-06,
      "loss": 0.26021337509155273,
      "num_input_tokens_seen": 623707200,
      "step": 10405,
      "train_runtime": 133796.5872,
      "train_tokens_per_second": 4661.608
    },
    {
      "epoch": 1.3278061224489797,
      "grad_norm": 0.6628471151287494,
      "learning_rate": 2.5385603799354574e-06,
      "loss": 0.24724078178405762,
      "num_input_tokens_seen": 624007680,
      "step": 10410,
      "train_runtime": 133876.0981,
      "train_tokens_per_second": 4661.084
    },
    {
      "epoch": 1.3284438775510203,
      "grad_norm": 0.6670291717928468,
      "learning_rate": 2.534201692149354e-06,
      "loss": 0.245704984664917,
      "num_input_tokens_seen": 624311856,
      "step": 10415,
      "train_runtime": 133933.8722,
      "train_tokens_per_second": 4661.344
    },
    {
      "epoch": 1.3290816326530612,
      "grad_norm": 0.6604817818216152,
      "learning_rate": 2.5298454789640636e-06,
      "loss": 0.2439598560333252,
      "num_input_tokens_seen": 624609024,
      "step": 10420,
      "train_runtime": 133991.0659,
      "train_tokens_per_second": 4661.572
    },
    {
      "epoch": 1.3297193877551021,
      "grad_norm": 0.6446840990830751,
      "learning_rate": 2.5254917447513503e-06,
      "loss": 0.2502993106842041,
      "num_input_tokens_seen": 624917624,
      "step": 10425,
      "train_runtime": 134047.6686,
      "train_tokens_per_second": 4661.906
    },
    {
      "epoch": 1.3303571428571428,
      "grad_norm": 0.5953900030540359,
      "learning_rate": 2.5211404938804917e-06,
      "loss": 0.2397331953048706,
      "num_input_tokens_seen": 625217040,
      "step": 10430,
      "train_runtime": 134098.6507,
      "train_tokens_per_second": 4662.366
    },
    {
      "epoch": 1.3309948979591837,
      "grad_norm": 0.6145028923979782,
      "learning_rate": 2.5167917307182675e-06,
      "loss": 0.26725015640258787,
      "num_input_tokens_seen": 625517304,
      "step": 10435,
      "train_runtime": 134161.939,
      "train_tokens_per_second": 4662.405
    },
    {
      "epoch": 1.3316326530612246,
      "grad_norm": 0.7190894950304159,
      "learning_rate": 2.5124454596289704e-06,
      "loss": 0.22888376712799072,
      "num_input_tokens_seen": 625814712,
      "step": 10440,
      "train_runtime": 134220.795,
      "train_tokens_per_second": 4662.576
    },
    {
      "epoch": 1.3322704081632653,
      "grad_norm": 0.6073485728861909,
      "learning_rate": 2.508101684974387e-06,
      "loss": 0.22669432163238526,
      "num_input_tokens_seen": 626111760,
      "step": 10445,
      "train_runtime": 134280.5503,
      "train_tokens_per_second": 4662.714
    },
    {
      "epoch": 1.3329081632653061,
      "grad_norm": 0.5863204642161498,
      "learning_rate": 2.5037604111137935e-06,
      "loss": 0.25122799873352053,
      "num_input_tokens_seen": 626416344,
      "step": 10450,
      "train_runtime": 134337.0694,
      "train_tokens_per_second": 4663.019
    },
    {
      "epoch": 1.333545918367347,
      "grad_norm": 0.6400215793325248,
      "learning_rate": 2.4994216424039637e-06,
      "loss": 0.2533901214599609,
      "num_input_tokens_seen": 626702664,
      "step": 10455,
      "train_runtime": 134399.6104,
      "train_tokens_per_second": 4662.98
    },
    {
      "epoch": 1.3341836734693877,
      "grad_norm": 0.6302628166382904,
      "learning_rate": 2.495085383199159e-06,
      "loss": 0.23258514404296876,
      "num_input_tokens_seen": 627006184,
      "step": 10460,
      "train_runtime": 134479.1024,
      "train_tokens_per_second": 4662.48
    },
    {
      "epoch": 1.3348214285714286,
      "grad_norm": 0.650789937664513,
      "learning_rate": 2.4907516378511137e-06,
      "loss": 0.2518965005874634,
      "num_input_tokens_seen": 627310752,
      "step": 10465,
      "train_runtime": 134536.4031,
      "train_tokens_per_second": 4662.758
    },
    {
      "epoch": 1.3354591836734695,
      "grad_norm": 0.6174199968443884,
      "learning_rate": 2.486420410709043e-06,
      "loss": 0.2388655424118042,
      "num_input_tokens_seen": 627617320,
      "step": 10470,
      "train_runtime": 134602.9343,
      "train_tokens_per_second": 4662.731
    },
    {
      "epoch": 1.3360969387755102,
      "grad_norm": 0.6387655861896523,
      "learning_rate": 2.482091706119642e-06,
      "loss": 0.24844703674316407,
      "num_input_tokens_seen": 627911064,
      "step": 10475,
      "train_runtime": 134668.1398,
      "train_tokens_per_second": 4662.655
    },
    {
      "epoch": 1.336734693877551,
      "grad_norm": 0.6694278283384967,
      "learning_rate": 2.4777655284270623e-06,
      "loss": 0.2570790767669678,
      "num_input_tokens_seen": 628219896,
      "step": 10480,
      "train_runtime": 134724.0819,
      "train_tokens_per_second": 4663.011
    },
    {
      "epoch": 1.337372448979592,
      "grad_norm": 0.633413740687077,
      "learning_rate": 2.473441881972927e-06,
      "loss": 0.2255763053894043,
      "num_input_tokens_seen": 628507776,
      "step": 10485,
      "train_runtime": 134786.9096,
      "train_tokens_per_second": 4662.973
    },
    {
      "epoch": 1.3380102040816326,
      "grad_norm": 0.7613036807692822,
      "learning_rate": 2.4691207710963185e-06,
      "loss": 0.25465667247772217,
      "num_input_tokens_seen": 628810656,
      "step": 10490,
      "train_runtime": 134851.944,
      "train_tokens_per_second": 4662.971
    },
    {
      "epoch": 1.3386479591836735,
      "grad_norm": 0.6566965367313716,
      "learning_rate": 2.464802200133772e-06,
      "loss": 0.25304470062255857,
      "num_input_tokens_seen": 629109008,
      "step": 10495,
      "train_runtime": 134909.6987,
      "train_tokens_per_second": 4663.186
    },
    {
      "epoch": 1.3392857142857144,
      "grad_norm": 0.711865578943311,
      "learning_rate": 2.4604861734192755e-06,
      "loss": 0.24025123119354247,
      "num_input_tokens_seen": 629411016,
      "step": 10500,
      "train_runtime": 134983.6166,
      "train_tokens_per_second": 4662.87
    },
    {
      "epoch": 1.339923469387755,
      "grad_norm": 0.6440249213267886,
      "learning_rate": 2.456172695284263e-06,
      "loss": 0.22090387344360352,
      "num_input_tokens_seen": 629693256,
      "step": 10505,
      "train_runtime": 135048.5715,
      "train_tokens_per_second": 4662.717
    },
    {
      "epoch": 1.340561224489796,
      "grad_norm": 0.6416653929775141,
      "learning_rate": 2.4518617700576116e-06,
      "loss": 0.22885353565216066,
      "num_input_tokens_seen": 629992856,
      "step": 10510,
      "train_runtime": 135114.8606,
      "train_tokens_per_second": 4662.647
    },
    {
      "epoch": 1.3411989795918369,
      "grad_norm": 0.6815679127747125,
      "learning_rate": 2.447553402065637e-06,
      "loss": 0.22661955356597902,
      "num_input_tokens_seen": 630288352,
      "step": 10515,
      "train_runtime": 135181.9621,
      "train_tokens_per_second": 4662.518
    },
    {
      "epoch": 1.3418367346938775,
      "grad_norm": 0.5811742789874202,
      "learning_rate": 2.4432475956320824e-06,
      "loss": 0.23862118721008302,
      "num_input_tokens_seen": 630594376,
      "step": 10520,
      "train_runtime": 135247.0115,
      "train_tokens_per_second": 4662.538
    },
    {
      "epoch": 1.3424744897959184,
      "grad_norm": 0.629641888521553,
      "learning_rate": 2.4389443550781304e-06,
      "loss": 0.24364774227142333,
      "num_input_tokens_seen": 630908936,
      "step": 10525,
      "train_runtime": 135301.6324,
      "train_tokens_per_second": 4662.981
    },
    {
      "epoch": 1.343112244897959,
      "grad_norm": 0.6889188070778034,
      "learning_rate": 2.4346436847223836e-06,
      "loss": 0.24048988819122313,
      "num_input_tokens_seen": 631206888,
      "step": 10530,
      "train_runtime": 135363.3059,
      "train_tokens_per_second": 4663.058
    },
    {
      "epoch": 1.34375,
      "grad_norm": 0.5503005271499407,
      "learning_rate": 2.4303455888808618e-06,
      "loss": 0.23441269397735595,
      "num_input_tokens_seen": 631507840,
      "step": 10535,
      "train_runtime": 135426.053,
      "train_tokens_per_second": 4663.119
    },
    {
      "epoch": 1.344387755102041,
      "grad_norm": 0.5943642128916533,
      "learning_rate": 2.4260500718670036e-06,
      "loss": 0.21130878925323487,
      "num_input_tokens_seen": 631812336,
      "step": 10540,
      "train_runtime": 135487.2071,
      "train_tokens_per_second": 4663.262
    },
    {
      "epoch": 1.3450255102040816,
      "grad_norm": 0.7419471964944657,
      "learning_rate": 2.4217571379916673e-06,
      "loss": 0.24795031547546387,
      "num_input_tokens_seen": 632100056,
      "step": 10545,
      "train_runtime": 135556.5893,
      "train_tokens_per_second": 4662.998
    },
    {
      "epoch": 1.3456632653061225,
      "grad_norm": 0.726479295167947,
      "learning_rate": 2.4174667915631038e-06,
      "loss": 0.23710207939147948,
      "num_input_tokens_seen": 632371248,
      "step": 10550,
      "train_runtime": 135622.3325,
      "train_tokens_per_second": 4662.737
    },
    {
      "epoch": 1.3463010204081631,
      "grad_norm": 0.6392188009075147,
      "learning_rate": 2.413179036886978e-06,
      "loss": 0.24065628051757812,
      "num_input_tokens_seen": 632679304,
      "step": 10555,
      "train_runtime": 135680.8874,
      "train_tokens_per_second": 4662.995
    },
    {
      "epoch": 1.346938775510204,
      "grad_norm": 0.6288809188937013,
      "learning_rate": 2.408893878266355e-06,
      "loss": 0.23537743091583252,
      "num_input_tokens_seen": 632968968,
      "step": 10560,
      "train_runtime": 135736.8122,
      "train_tokens_per_second": 4663.208
    },
    {
      "epoch": 1.347576530612245,
      "grad_norm": 0.5858486552796457,
      "learning_rate": 2.404611320001686e-06,
      "loss": 0.24897806644439696,
      "num_input_tokens_seen": 633263232,
      "step": 10565,
      "train_runtime": 135803.3521,
      "train_tokens_per_second": 4663.09
    },
    {
      "epoch": 1.3482142857142856,
      "grad_norm": 0.6096113101348661,
      "learning_rate": 2.4003313663908182e-06,
      "loss": 0.25921201705932617,
      "num_input_tokens_seen": 633555128,
      "step": 10570,
      "train_runtime": 135863.5392,
      "train_tokens_per_second": 4663.173
    },
    {
      "epoch": 1.3488520408163265,
      "grad_norm": 0.7193883193154333,
      "learning_rate": 2.3960540217289856e-06,
      "loss": 0.25274574756622314,
      "num_input_tokens_seen": 633860472,
      "step": 10575,
      "train_runtime": 135938.6281,
      "train_tokens_per_second": 4662.843
    },
    {
      "epoch": 1.3494897959183674,
      "grad_norm": 0.6737362160482706,
      "learning_rate": 2.3917792903088013e-06,
      "loss": 0.24594407081604003,
      "num_input_tokens_seen": 634153200,
      "step": 10580,
      "train_runtime": 136000.8431,
      "train_tokens_per_second": 4662.862
    },
    {
      "epoch": 1.350127551020408,
      "grad_norm": 0.8204822910763974,
      "learning_rate": 2.387507176420256e-06,
      "loss": 0.22988698482513428,
      "num_input_tokens_seen": 634449192,
      "step": 10585,
      "train_runtime": 136061.1127,
      "train_tokens_per_second": 4662.972
    },
    {
      "epoch": 1.350765306122449,
      "grad_norm": 0.6617703724129987,
      "learning_rate": 2.383237684350716e-06,
      "loss": 0.24116842746734618,
      "num_input_tokens_seen": 634753368,
      "step": 10590,
      "train_runtime": 136134.8489,
      "train_tokens_per_second": 4662.681
    },
    {
      "epoch": 1.3514030612244898,
      "grad_norm": 0.6979442026854079,
      "learning_rate": 2.378970818384914e-06,
      "loss": 0.25265040397644045,
      "num_input_tokens_seen": 635055304,
      "step": 10595,
      "train_runtime": 136203.1716,
      "train_tokens_per_second": 4662.559
    },
    {
      "epoch": 1.3520408163265305,
      "grad_norm": 0.658357782720236,
      "learning_rate": 2.374706582804948e-06,
      "loss": 0.2619814395904541,
      "num_input_tokens_seen": 635363696,
      "step": 10600,
      "train_runtime": 136275.2005,
      "train_tokens_per_second": 4662.357
    },
    {
      "epoch": 1.3526785714285714,
      "grad_norm": 0.6510269641150819,
      "learning_rate": 2.3704449818902763e-06,
      "loss": 0.24418268203735352,
      "num_input_tokens_seen": 635657176,
      "step": 10605,
      "train_runtime": 136343.0031,
      "train_tokens_per_second": 4662.191
    },
    {
      "epoch": 1.3533163265306123,
      "grad_norm": 0.6124389034198514,
      "learning_rate": 2.366186019917713e-06,
      "loss": 0.2479926109313965,
      "num_input_tokens_seen": 635970392,
      "step": 10610,
      "train_runtime": 136415.4066,
      "train_tokens_per_second": 4662.013
    },
    {
      "epoch": 1.353954081632653,
      "grad_norm": 0.6788895914890781,
      "learning_rate": 2.3619297011614236e-06,
      "loss": 0.25220503807067873,
      "num_input_tokens_seen": 636274720,
      "step": 10615,
      "train_runtime": 136469.385,
      "train_tokens_per_second": 4662.399
    },
    {
      "epoch": 1.3545918367346939,
      "grad_norm": 0.6662601504754513,
      "learning_rate": 2.3576760298929236e-06,
      "loss": 0.2631206512451172,
      "num_input_tokens_seen": 636585848,
      "step": 10620,
      "train_runtime": 136540.6342,
      "train_tokens_per_second": 4662.245
    },
    {
      "epoch": 1.3552295918367347,
      "grad_norm": 0.6569193427518427,
      "learning_rate": 2.353425010381063e-06,
      "loss": 0.23456768989562987,
      "num_input_tokens_seen": 636902600,
      "step": 10625,
      "train_runtime": 136587.5461,
      "train_tokens_per_second": 4662.962
    },
    {
      "epoch": 1.3558673469387754,
      "grad_norm": 0.5908036238213901,
      "learning_rate": 2.349176646892045e-06,
      "loss": 0.2443864107131958,
      "num_input_tokens_seen": 637205080,
      "step": 10630,
      "train_runtime": 136654.3475,
      "train_tokens_per_second": 4662.897
    },
    {
      "epoch": 1.3565051020408163,
      "grad_norm": 0.6739611775553626,
      "learning_rate": 2.344930943689393e-06,
      "loss": 0.23467619419097902,
      "num_input_tokens_seen": 637486312,
      "step": 10635,
      "train_runtime": 136724.48,
      "train_tokens_per_second": 4662.562
    },
    {
      "epoch": 1.3571428571428572,
      "grad_norm": 0.639161234478945,
      "learning_rate": 2.3406879050339664e-06,
      "loss": 0.25684666633605957,
      "num_input_tokens_seen": 637788312,
      "step": 10640,
      "train_runtime": 136780.9908,
      "train_tokens_per_second": 4662.843
    },
    {
      "epoch": 1.3577806122448979,
      "grad_norm": 0.6199280486615185,
      "learning_rate": 2.3364475351839573e-06,
      "loss": 0.23010401725769042,
      "num_input_tokens_seen": 638080600,
      "step": 10645,
      "train_runtime": 136850.3767,
      "train_tokens_per_second": 4662.615
    },
    {
      "epoch": 1.3584183673469388,
      "grad_norm": 0.7195215166221605,
      "learning_rate": 2.3322098383948666e-06,
      "loss": 0.26816473007202146,
      "num_input_tokens_seen": 638383856,
      "step": 10650,
      "train_runtime": 136914.3956,
      "train_tokens_per_second": 4662.65
    },
    {
      "epoch": 1.3590561224489797,
      "grad_norm": 0.6682298338800575,
      "learning_rate": 2.327974818919521e-06,
      "loss": 0.24030332565307616,
      "num_input_tokens_seen": 638676280,
      "step": 10655,
      "train_runtime": 136986.5008,
      "train_tokens_per_second": 4662.33
    },
    {
      "epoch": 1.3596938775510203,
      "grad_norm": 0.6530466542090518,
      "learning_rate": 2.323742481008059e-06,
      "loss": 0.2469203233718872,
      "num_input_tokens_seen": 638982936,
      "step": 10660,
      "train_runtime": 137053.4071,
      "train_tokens_per_second": 4662.292
    },
    {
      "epoch": 1.3603316326530612,
      "grad_norm": 0.646929425391726,
      "learning_rate": 2.3195128289079265e-06,
      "loss": 0.2459321975708008,
      "num_input_tokens_seen": 639278712,
      "step": 10665,
      "train_runtime": 137127.2866,
      "train_tokens_per_second": 4661.937
    },
    {
      "epoch": 1.3609693877551021,
      "grad_norm": 0.7074087517351287,
      "learning_rate": 2.315285866863875e-06,
      "loss": 0.2579158306121826,
      "num_input_tokens_seen": 639577152,
      "step": 10670,
      "train_runtime": 137182.26,
      "train_tokens_per_second": 4662.244
    },
    {
      "epoch": 1.3616071428571428,
      "grad_norm": 0.6926672430737997,
      "learning_rate": 2.3110615991179556e-06,
      "loss": 0.24455657005310058,
      "num_input_tokens_seen": 639864128,
      "step": 10675,
      "train_runtime": 137237.5477,
      "train_tokens_per_second": 4662.457
    },
    {
      "epoch": 1.3622448979591837,
      "grad_norm": 0.6265849100101015,
      "learning_rate": 2.3068400299095172e-06,
      "loss": 0.2326984405517578,
      "num_input_tokens_seen": 640165968,
      "step": 10680,
      "train_runtime": 137301.6068,
      "train_tokens_per_second": 4662.48
    },
    {
      "epoch": 1.3628826530612246,
      "grad_norm": 0.5785532785540959,
      "learning_rate": 2.302621163475198e-06,
      "loss": 0.22290825843811035,
      "num_input_tokens_seen": 640467656,
      "step": 10685,
      "train_runtime": 137377.494,
      "train_tokens_per_second": 4662.1
    },
    {
      "epoch": 1.3635204081632653,
      "grad_norm": 0.6051768783529228,
      "learning_rate": 2.2984050040489266e-06,
      "loss": 0.2333277702331543,
      "num_input_tokens_seen": 640758208,
      "step": 10690,
      "train_runtime": 137446.7759,
      "train_tokens_per_second": 4661.864
    },
    {
      "epoch": 1.3641581632653061,
      "grad_norm": 0.6097162866857359,
      "learning_rate": 2.2941915558619123e-06,
      "loss": 0.24473967552185058,
      "num_input_tokens_seen": 641051696,
      "step": 10695,
      "train_runtime": 137513.7724,
      "train_tokens_per_second": 4661.727
    },
    {
      "epoch": 1.364795918367347,
      "grad_norm": 0.6645129467960652,
      "learning_rate": 2.2899808231426447e-06,
      "loss": 0.24537134170532227,
      "num_input_tokens_seen": 641360072,
      "step": 10700,
      "train_runtime": 137582.2569,
      "train_tokens_per_second": 4661.648
    },
    {
      "epoch": 1.3654336734693877,
      "grad_norm": 0.5901411951603963,
      "learning_rate": 2.28577281011689e-06,
      "loss": 0.25689105987548827,
      "num_input_tokens_seen": 641670816,
      "step": 10705,
      "train_runtime": 137648.1907,
      "train_tokens_per_second": 4661.673
    },
    {
      "epoch": 1.3660714285714286,
      "grad_norm": 0.6625252011700808,
      "learning_rate": 2.281567521007678e-06,
      "loss": 0.23990726470947266,
      "num_input_tokens_seen": 641974704,
      "step": 10710,
      "train_runtime": 137715.534,
      "train_tokens_per_second": 4661.6
    },
    {
      "epoch": 1.3667091836734695,
      "grad_norm": 0.5962854350741728,
      "learning_rate": 2.277364960035316e-06,
      "loss": 0.2552059173583984,
      "num_input_tokens_seen": 642271216,
      "step": 10715,
      "train_runtime": 137774.8607,
      "train_tokens_per_second": 4661.745
    },
    {
      "epoch": 1.3673469387755102,
      "grad_norm": 0.599624008965094,
      "learning_rate": 2.2731651314173665e-06,
      "loss": 0.2545493125915527,
      "num_input_tokens_seen": 642578080,
      "step": 10720,
      "train_runtime": 137832.0187,
      "train_tokens_per_second": 4662.038
    },
    {
      "epoch": 1.367984693877551,
      "grad_norm": 0.625085449046815,
      "learning_rate": 2.2689680393686457e-06,
      "loss": 0.2324829578399658,
      "num_input_tokens_seen": 642872696,
      "step": 10725,
      "train_runtime": 137903.51,
      "train_tokens_per_second": 4661.757
    },
    {
      "epoch": 1.368622448979592,
      "grad_norm": 0.5746890026603176,
      "learning_rate": 2.2647736881012337e-06,
      "loss": 0.2443406343460083,
      "num_input_tokens_seen": 643173696,
      "step": 10730,
      "train_runtime": 137972.705,
      "train_tokens_per_second": 4661.601
    },
    {
      "epoch": 1.3692602040816326,
      "grad_norm": 0.5818243039191208,
      "learning_rate": 2.2605820818244563e-06,
      "loss": 0.21316955089569092,
      "num_input_tokens_seen": 643471536,
      "step": 10735,
      "train_runtime": 138025.5767,
      "train_tokens_per_second": 4661.973
    },
    {
      "epoch": 1.3698979591836735,
      "grad_norm": 0.5937263833030431,
      "learning_rate": 2.256393224744878e-06,
      "loss": 0.27692992687225343,
      "num_input_tokens_seen": 643787880,
      "step": 10740,
      "train_runtime": 138088.0402,
      "train_tokens_per_second": 4662.155
    },
    {
      "epoch": 1.3705357142857144,
      "grad_norm": 0.6711663853709126,
      "learning_rate": 2.252207121066311e-06,
      "loss": 0.22586865425109864,
      "num_input_tokens_seen": 644091064,
      "step": 10745,
      "train_runtime": 138150.7791,
      "train_tokens_per_second": 4662.233
    },
    {
      "epoch": 1.371173469387755,
      "grad_norm": 0.6624536991372643,
      "learning_rate": 2.248023774989804e-06,
      "loss": 0.23344554901123046,
      "num_input_tokens_seen": 644385424,
      "step": 10750,
      "train_runtime": 138217.7475,
      "train_tokens_per_second": 4662.103
    },
    {
      "epoch": 1.371811224489796,
      "grad_norm": 0.645300765417727,
      "learning_rate": 2.2438431907136365e-06,
      "loss": 0.25888876914978026,
      "num_input_tokens_seen": 644697888,
      "step": 10755,
      "train_runtime": 138284.0196,
      "train_tokens_per_second": 4662.129
    },
    {
      "epoch": 1.3724489795918369,
      "grad_norm": 0.6695649211368374,
      "learning_rate": 2.2396653724333167e-06,
      "loss": 0.23562912940979003,
      "num_input_tokens_seen": 644996744,
      "step": 10760,
      "train_runtime": 138354.819,
      "train_tokens_per_second": 4661.903
    },
    {
      "epoch": 1.3730867346938775,
      "grad_norm": 0.7433713974810475,
      "learning_rate": 2.235490324341577e-06,
      "loss": 0.23277924060821534,
      "num_input_tokens_seen": 645295648,
      "step": 10765,
      "train_runtime": 138424.9162,
      "train_tokens_per_second": 4661.702
    },
    {
      "epoch": 1.3737244897959184,
      "grad_norm": 0.6513759630040247,
      "learning_rate": 2.2313180506283695e-06,
      "loss": 0.24211931228637695,
      "num_input_tokens_seen": 645597200,
      "step": 10770,
      "train_runtime": 138493.9833,
      "train_tokens_per_second": 4661.554
    },
    {
      "epoch": 1.374362244897959,
      "grad_norm": 0.6582279634304187,
      "learning_rate": 2.227148555480863e-06,
      "loss": 0.25272645950317385,
      "num_input_tokens_seen": 645908304,
      "step": 10775,
      "train_runtime": 138549.7575,
      "train_tokens_per_second": 4661.923
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.703956439443315,
      "learning_rate": 2.222981843083436e-06,
      "loss": 0.25603179931640624,
      "num_input_tokens_seen": 646217384,
      "step": 10780,
      "train_runtime": 138610.6027,
      "train_tokens_per_second": 4662.106
    },
    {
      "epoch": 1.375637755102041,
      "grad_norm": 0.6925112911296832,
      "learning_rate": 2.2188179176176767e-06,
      "loss": 0.2590804100036621,
      "num_input_tokens_seen": 646527384,
      "step": 10785,
      "train_runtime": 138672.479,
      "train_tokens_per_second": 4662.262
    },
    {
      "epoch": 1.3762755102040816,
      "grad_norm": 0.6069876444123631,
      "learning_rate": 2.2146567832623737e-06,
      "loss": 0.25036039352416994,
      "num_input_tokens_seen": 646839656,
      "step": 10790,
      "train_runtime": 138734.7916,
      "train_tokens_per_second": 4662.418
    },
    {
      "epoch": 1.3769132653061225,
      "grad_norm": 0.7297829836920665,
      "learning_rate": 2.2104984441935167e-06,
      "loss": 0.24140830039978028,
      "num_input_tokens_seen": 647139928,
      "step": 10795,
      "train_runtime": 138792.1608,
      "train_tokens_per_second": 4662.655
    },
    {
      "epoch": 1.3775510204081631,
      "grad_norm": 0.6625224490748896,
      "learning_rate": 2.20634290458429e-06,
      "loss": 0.24470853805541992,
      "num_input_tokens_seen": 647447272,
      "step": 10800,
      "train_runtime": 138857.2637,
      "train_tokens_per_second": 4662.682
    },
    {
      "epoch": 1.378188775510204,
      "grad_norm": 0.7128958812980214,
      "learning_rate": 2.2021901686050684e-06,
      "loss": 0.23196766376495362,
      "num_input_tokens_seen": 647735072,
      "step": 10805,
      "train_runtime": 138921.8986,
      "train_tokens_per_second": 4662.584
    },
    {
      "epoch": 1.378826530612245,
      "grad_norm": 0.5795062646451145,
      "learning_rate": 2.1980402404234074e-06,
      "loss": 0.25746376514434816,
      "num_input_tokens_seen": 648041280,
      "step": 10810,
      "train_runtime": 138983.4149,
      "train_tokens_per_second": 4662.724
    },
    {
      "epoch": 1.3794642857142856,
      "grad_norm": 0.6885537391435483,
      "learning_rate": 2.1938931242040557e-06,
      "loss": 0.2460865259170532,
      "num_input_tokens_seen": 648341640,
      "step": 10815,
      "train_runtime": 139041.6036,
      "train_tokens_per_second": 4662.933
    },
    {
      "epoch": 1.3801020408163265,
      "grad_norm": 0.618919758545103,
      "learning_rate": 2.1897488241089336e-06,
      "loss": 0.2416177749633789,
      "num_input_tokens_seen": 648661488,
      "step": 10820,
      "train_runtime": 139101.7281,
      "train_tokens_per_second": 4663.217
    },
    {
      "epoch": 1.3807397959183674,
      "grad_norm": 0.7381930866970408,
      "learning_rate": 2.1856073442971322e-06,
      "loss": 0.2605881690979004,
      "num_input_tokens_seen": 648955528,
      "step": 10825,
      "train_runtime": 139161.4825,
      "train_tokens_per_second": 4663.327
    },
    {
      "epoch": 1.381377551020408,
      "grad_norm": 0.5969600365520998,
      "learning_rate": 2.181468688924916e-06,
      "loss": 0.23219058513641358,
      "num_input_tokens_seen": 649244072,
      "step": 10830,
      "train_runtime": 139228.1126,
      "train_tokens_per_second": 4663.168
    },
    {
      "epoch": 1.382015306122449,
      "grad_norm": 0.6805677768313665,
      "learning_rate": 2.1773328621457196e-06,
      "loss": 0.24557528495788575,
      "num_input_tokens_seen": 649529976,
      "step": 10835,
      "train_runtime": 139295.5333,
      "train_tokens_per_second": 4662.963
    },
    {
      "epoch": 1.3826530612244898,
      "grad_norm": 0.6479006672381705,
      "learning_rate": 2.1731998681101293e-06,
      "loss": 0.22532806396484376,
      "num_input_tokens_seen": 649833488,
      "step": 10840,
      "train_runtime": 139348.8616,
      "train_tokens_per_second": 4663.357
    },
    {
      "epoch": 1.3832908163265305,
      "grad_norm": 0.646316324356889,
      "learning_rate": 2.169069710965892e-06,
      "loss": 0.2405627727508545,
      "num_input_tokens_seen": 650140848,
      "step": 10845,
      "train_runtime": 139405.2632,
      "train_tokens_per_second": 4663.675
    },
    {
      "epoch": 1.3839285714285714,
      "grad_norm": 0.5367163420276531,
      "learning_rate": 2.1649423948579157e-06,
      "loss": 0.2120504379272461,
      "num_input_tokens_seen": 650433136,
      "step": 10850,
      "train_runtime": 139461.4156,
      "train_tokens_per_second": 4663.893
    },
    {
      "epoch": 1.3845663265306123,
      "grad_norm": 0.7034689899429091,
      "learning_rate": 2.160817923928244e-06,
      "loss": 0.24123435020446776,
      "num_input_tokens_seen": 650725704,
      "step": 10855,
      "train_runtime": 139528.9814,
      "train_tokens_per_second": 4663.731
    },
    {
      "epoch": 1.385204081632653,
      "grad_norm": 0.6566511350086877,
      "learning_rate": 2.156696302316074e-06,
      "loss": 0.2437469482421875,
      "num_input_tokens_seen": 651034544,
      "step": 10860,
      "train_runtime": 139602.7219,
      "train_tokens_per_second": 4663.48
    },
    {
      "epoch": 1.3858418367346939,
      "grad_norm": 0.6316114508662404,
      "learning_rate": 2.1525775341577404e-06,
      "loss": 0.25299577713012694,
      "num_input_tokens_seen": 651327096,
      "step": 10865,
      "train_runtime": 139665.5038,
      "train_tokens_per_second": 4663.479
    },
    {
      "epoch": 1.3864795918367347,
      "grad_norm": 0.6568636171902431,
      "learning_rate": 2.1484616235867152e-06,
      "loss": 0.22642965316772462,
      "num_input_tokens_seen": 651630216,
      "step": 10870,
      "train_runtime": 139737.9928,
      "train_tokens_per_second": 4663.229
    },
    {
      "epoch": 1.3871173469387754,
      "grad_norm": 0.7497381278526272,
      "learning_rate": 2.1443485747336014e-06,
      "loss": 0.2759822368621826,
      "num_input_tokens_seen": 651916104,
      "step": 10875,
      "train_runtime": 139801.0212,
      "train_tokens_per_second": 4663.171
    },
    {
      "epoch": 1.3877551020408163,
      "grad_norm": 0.6809589351702348,
      "learning_rate": 2.140238391726131e-06,
      "loss": 0.24073576927185059,
      "num_input_tokens_seen": 652222272,
      "step": 10880,
      "train_runtime": 139862.9104,
      "train_tokens_per_second": 4663.297
    },
    {
      "epoch": 1.3883928571428572,
      "grad_norm": 0.6546587477534579,
      "learning_rate": 2.1361310786891596e-06,
      "loss": 0.24782133102416992,
      "num_input_tokens_seen": 652528072,
      "step": 10885,
      "train_runtime": 139924.7952,
      "train_tokens_per_second": 4663.42
    },
    {
      "epoch": 1.3890306122448979,
      "grad_norm": 0.7041156445836272,
      "learning_rate": 2.1320266397446655e-06,
      "loss": 0.22001242637634277,
      "num_input_tokens_seen": 652839480,
      "step": 10890,
      "train_runtime": 139994.7978,
      "train_tokens_per_second": 4663.312
    },
    {
      "epoch": 1.3896683673469388,
      "grad_norm": 0.6044581168478713,
      "learning_rate": 2.1279250790117327e-06,
      "loss": 0.2347689151763916,
      "num_input_tokens_seen": 653131656,
      "step": 10895,
      "train_runtime": 140056.3181,
      "train_tokens_per_second": 4663.35
    },
    {
      "epoch": 1.3903061224489797,
      "grad_norm": 0.679840143816703,
      "learning_rate": 2.123826400606571e-06,
      "loss": 0.25978033542633056,
      "num_input_tokens_seen": 653444432,
      "step": 10900,
      "train_runtime": 140116.3695,
      "train_tokens_per_second": 4663.584
    },
    {
      "epoch": 1.3909438775510203,
      "grad_norm": 0.7119887067114313,
      "learning_rate": 2.119730608642489e-06,
      "loss": 0.22347233295440674,
      "num_input_tokens_seen": 653719088,
      "step": 10905,
      "train_runtime": 140180.1575,
      "train_tokens_per_second": 4663.421
    },
    {
      "epoch": 1.3915816326530612,
      "grad_norm": 0.6970403555829308,
      "learning_rate": 2.1156377072298972e-06,
      "loss": 0.23890743255615235,
      "num_input_tokens_seen": 654002784,
      "step": 10910,
      "train_runtime": 140246.4078,
      "train_tokens_per_second": 4663.241
    },
    {
      "epoch": 1.3922193877551021,
      "grad_norm": 0.7448028660534038,
      "learning_rate": 2.111547700476308e-06,
      "loss": 0.23602609634399413,
      "num_input_tokens_seen": 654315184,
      "step": 10915,
      "train_runtime": 140318.9851,
      "train_tokens_per_second": 4663.055
    },
    {
      "epoch": 1.3928571428571428,
      "grad_norm": 0.6584720165435018,
      "learning_rate": 2.107460592486333e-06,
      "loss": 0.2450235366821289,
      "num_input_tokens_seen": 654609464,
      "step": 10920,
      "train_runtime": 140376.0996,
      "train_tokens_per_second": 4663.254
    },
    {
      "epoch": 1.3934948979591837,
      "grad_norm": 0.6020681394599366,
      "learning_rate": 2.1033763873616675e-06,
      "loss": 0.23806238174438477,
      "num_input_tokens_seen": 654919048,
      "step": 10925,
      "train_runtime": 140451.687,
      "train_tokens_per_second": 4662.949
    },
    {
      "epoch": 1.3941326530612246,
      "grad_norm": 0.6544156838796378,
      "learning_rate": 2.0992950892010948e-06,
      "loss": 0.2461150884628296,
      "num_input_tokens_seen": 655220344,
      "step": 10930,
      "train_runtime": 140523.0406,
      "train_tokens_per_second": 4662.725
    },
    {
      "epoch": 1.3947704081632653,
      "grad_norm": 0.7281168804343539,
      "learning_rate": 2.0952167021004887e-06,
      "loss": 0.2532359600067139,
      "num_input_tokens_seen": 655510768,
      "step": 10935,
      "train_runtime": 140586.3961,
      "train_tokens_per_second": 4662.69
    },
    {
      "epoch": 1.3954081632653061,
      "grad_norm": 0.6733824759771405,
      "learning_rate": 2.0911412301527895e-06,
      "loss": 0.24455971717834474,
      "num_input_tokens_seen": 655818816,
      "step": 10940,
      "train_runtime": 140651.5694,
      "train_tokens_per_second": 4662.72
    },
    {
      "epoch": 1.396045918367347,
      "grad_norm": 0.6348471200624451,
      "learning_rate": 2.08706867744802e-06,
      "loss": 0.24333462715148926,
      "num_input_tokens_seen": 656117408,
      "step": 10945,
      "train_runtime": 140723.0535,
      "train_tokens_per_second": 4662.473
    },
    {
      "epoch": 1.3966836734693877,
      "grad_norm": 0.6655750157123884,
      "learning_rate": 2.0829990480732715e-06,
      "loss": 0.2534806251525879,
      "num_input_tokens_seen": 656417616,
      "step": 10950,
      "train_runtime": 140786.0205,
      "train_tokens_per_second": 4662.52
    },
    {
      "epoch": 1.3973214285714286,
      "grad_norm": 0.714398834957036,
      "learning_rate": 2.0789323461127015e-06,
      "loss": 0.23948030471801757,
      "num_input_tokens_seen": 656714808,
      "step": 10955,
      "train_runtime": 140854.2996,
      "train_tokens_per_second": 4662.37
    },
    {
      "epoch": 1.3979591836734695,
      "grad_norm": 0.6979327937979668,
      "learning_rate": 2.0748685756475286e-06,
      "loss": 0.2529648542404175,
      "num_input_tokens_seen": 657030600,
      "step": 10960,
      "train_runtime": 140901.2888,
      "train_tokens_per_second": 4663.056
    },
    {
      "epoch": 1.3985969387755102,
      "grad_norm": 0.715220965630344,
      "learning_rate": 2.0708077407560305e-06,
      "loss": 0.2523340225219727,
      "num_input_tokens_seen": 657334728,
      "step": 10965,
      "train_runtime": 140966.7956,
      "train_tokens_per_second": 4663.047
    },
    {
      "epoch": 1.399234693877551,
      "grad_norm": 0.6419444224663815,
      "learning_rate": 2.06674984551354e-06,
      "loss": 0.26148099899291993,
      "num_input_tokens_seen": 657631672,
      "step": 10970,
      "train_runtime": 141032.4216,
      "train_tokens_per_second": 4662.982
    },
    {
      "epoch": 1.399872448979592,
      "grad_norm": 0.7152631913428876,
      "learning_rate": 2.062694893992436e-06,
      "loss": 0.26262149810791013,
      "num_input_tokens_seen": 657947016,
      "step": 10975,
      "train_runtime": 141104.216,
      "train_tokens_per_second": 4662.844
    },
    {
      "epoch": 1.4005102040816326,
      "grad_norm": 0.6295128760306162,
      "learning_rate": 2.0586428902621474e-06,
      "loss": 0.2281444787979126,
      "num_input_tokens_seen": 658238832,
      "step": 10980,
      "train_runtime": 141173.076,
      "train_tokens_per_second": 4662.637
    },
    {
      "epoch": 1.4011479591836735,
      "grad_norm": 0.6108999780425177,
      "learning_rate": 2.0545938383891427e-06,
      "loss": 0.2359621524810791,
      "num_input_tokens_seen": 658543584,
      "step": 10985,
      "train_runtime": 141241.7291,
      "train_tokens_per_second": 4662.528
    },
    {
      "epoch": 1.4017857142857144,
      "grad_norm": 0.7402467771476116,
      "learning_rate": 2.050547742436929e-06,
      "loss": 0.23925113677978516,
      "num_input_tokens_seen": 658829344,
      "step": 10990,
      "train_runtime": 141307.5423,
      "train_tokens_per_second": 4662.379
    },
    {
      "epoch": 1.402423469387755,
      "grad_norm": 0.6596778939128919,
      "learning_rate": 2.0465046064660437e-06,
      "loss": 0.24088129997253419,
      "num_input_tokens_seen": 659133976,
      "step": 10995,
      "train_runtime": 141371.9924,
      "train_tokens_per_second": 4662.408
    },
    {
      "epoch": 1.403061224489796,
      "grad_norm": 0.6713856462192935,
      "learning_rate": 2.0424644345340554e-06,
      "loss": 0.25354528427124023,
      "num_input_tokens_seen": 659443248,
      "step": 11000,
      "train_runtime": 141447.1026,
      "train_tokens_per_second": 4662.119
    },
    {
      "epoch": 1.4036989795918369,
      "grad_norm": 0.707561358316077,
      "learning_rate": 2.038427230695565e-06,
      "loss": 0.2335648536682129,
      "num_input_tokens_seen": 659738752,
      "step": 11005,
      "train_runtime": 141518.5743,
      "train_tokens_per_second": 4661.853
    },
    {
      "epoch": 1.4043367346938775,
      "grad_norm": 0.6662093402085221,
      "learning_rate": 2.034392999002182e-06,
      "loss": 0.2513861179351807,
      "num_input_tokens_seen": 660044640,
      "step": 11010,
      "train_runtime": 141592.8124,
      "train_tokens_per_second": 4661.569
    },
    {
      "epoch": 1.4049744897959184,
      "grad_norm": 0.5934658664915485,
      "learning_rate": 2.0303617435025407e-06,
      "loss": 0.2568632125854492,
      "num_input_tokens_seen": 660343304,
      "step": 11015,
      "train_runtime": 141651.6033,
      "train_tokens_per_second": 4661.743
    },
    {
      "epoch": 1.405612244897959,
      "grad_norm": 0.6909332334251601,
      "learning_rate": 2.0263334682422923e-06,
      "loss": 0.2585726022720337,
      "num_input_tokens_seen": 660645536,
      "step": 11020,
      "train_runtime": 141714.1814,
      "train_tokens_per_second": 4661.817
    },
    {
      "epoch": 1.40625,
      "grad_norm": 0.6146840244682565,
      "learning_rate": 2.0223081772640867e-06,
      "loss": 0.2446803092956543,
      "num_input_tokens_seen": 660959280,
      "step": 11025,
      "train_runtime": 141773.005,
      "train_tokens_per_second": 4662.095
    },
    {
      "epoch": 1.406887755102041,
      "grad_norm": 0.6161975066320161,
      "learning_rate": 2.0182858746075873e-06,
      "loss": 0.23802344799041747,
      "num_input_tokens_seen": 661259400,
      "step": 11030,
      "train_runtime": 141844.9776,
      "train_tokens_per_second": 4661.846
    },
    {
      "epoch": 1.4075255102040816,
      "grad_norm": 0.7001020761944587,
      "learning_rate": 2.014266564309455e-06,
      "loss": 0.23745441436767578,
      "num_input_tokens_seen": 661554896,
      "step": 11035,
      "train_runtime": 141917.0231,
      "train_tokens_per_second": 4661.561
    },
    {
      "epoch": 1.4081632653061225,
      "grad_norm": 0.7623689546765444,
      "learning_rate": 2.010250250403349e-06,
      "loss": 0.23623900413513182,
      "num_input_tokens_seen": 661857168,
      "step": 11040,
      "train_runtime": 141968.1928,
      "train_tokens_per_second": 4662.01
    },
    {
      "epoch": 1.4088010204081631,
      "grad_norm": 0.6716639665075063,
      "learning_rate": 2.00623693691992e-06,
      "loss": 0.230047869682312,
      "num_input_tokens_seen": 662161992,
      "step": 11045,
      "train_runtime": 142023.881,
      "train_tokens_per_second": 4662.329
    },
    {
      "epoch": 1.409438775510204,
      "grad_norm": 0.6404711583431383,
      "learning_rate": 2.0022266278868086e-06,
      "loss": 0.22629778385162352,
      "num_input_tokens_seen": 662457600,
      "step": 11050,
      "train_runtime": 142080.0043,
      "train_tokens_per_second": 4662.567
    },
    {
      "epoch": 1.410076530612245,
      "grad_norm": 0.7615641341041465,
      "learning_rate": 1.9982193273286404e-06,
      "loss": 0.2338866949081421,
      "num_input_tokens_seen": 662748816,
      "step": 11055,
      "train_runtime": 142140.8984,
      "train_tokens_per_second": 4662.619
    },
    {
      "epoch": 1.4107142857142856,
      "grad_norm": 0.6127624693919943,
      "learning_rate": 1.9942150392670207e-06,
      "loss": 0.24228882789611816,
      "num_input_tokens_seen": 663046120,
      "step": 11060,
      "train_runtime": 142196.6147,
      "train_tokens_per_second": 4662.883
    },
    {
      "epoch": 1.4113520408163265,
      "grad_norm": 0.863624612549951,
      "learning_rate": 1.990213767720533e-06,
      "loss": 0.24905591011047362,
      "num_input_tokens_seen": 663346040,
      "step": 11065,
      "train_runtime": 142268.3757,
      "train_tokens_per_second": 4662.639
    },
    {
      "epoch": 1.4119897959183674,
      "grad_norm": 0.7022959171251992,
      "learning_rate": 1.9862155167047324e-06,
      "loss": 0.23888988494873048,
      "num_input_tokens_seen": 663647624,
      "step": 11070,
      "train_runtime": 142328.3859,
      "train_tokens_per_second": 4662.792
    },
    {
      "epoch": 1.412627551020408,
      "grad_norm": 0.6573008811431766,
      "learning_rate": 1.982220290232143e-06,
      "loss": 0.2574120044708252,
      "num_input_tokens_seen": 663942832,
      "step": 11075,
      "train_runtime": 142389.623,
      "train_tokens_per_second": 4662.86
    },
    {
      "epoch": 1.413265306122449,
      "grad_norm": 0.5892065478241137,
      "learning_rate": 1.9782280923122555e-06,
      "loss": 0.23816120624542236,
      "num_input_tokens_seen": 664233128,
      "step": 11080,
      "train_runtime": 142462.6762,
      "train_tokens_per_second": 4662.506
    },
    {
      "epoch": 1.4139030612244898,
      "grad_norm": 0.6275382672962257,
      "learning_rate": 1.9742389269515137e-06,
      "loss": 0.22530632019042968,
      "num_input_tokens_seen": 664539784,
      "step": 11085,
      "train_runtime": 142523.3778,
      "train_tokens_per_second": 4662.672
    },
    {
      "epoch": 1.4145408163265305,
      "grad_norm": 0.6834563719217182,
      "learning_rate": 1.970252798153329e-06,
      "loss": 0.2329627513885498,
      "num_input_tokens_seen": 664831808,
      "step": 11090,
      "train_runtime": 142587.9089,
      "train_tokens_per_second": 4662.61
    },
    {
      "epoch": 1.4151785714285714,
      "grad_norm": 0.627341426713708,
      "learning_rate": 1.9662697099180607e-06,
      "loss": 0.2375498056411743,
      "num_input_tokens_seen": 665123032,
      "step": 11095,
      "train_runtime": 142654.8248,
      "train_tokens_per_second": 4662.464
    },
    {
      "epoch": 1.4158163265306123,
      "grad_norm": 0.6226154279247583,
      "learning_rate": 1.9622896662430098e-06,
      "loss": 0.24169931411743165,
      "num_input_tokens_seen": 665418896,
      "step": 11100,
      "train_runtime": 142710.0802,
      "train_tokens_per_second": 4662.732
    },
    {
      "epoch": 1.416454081632653,
      "grad_norm": 0.671738072276809,
      "learning_rate": 1.9583126711224342e-06,
      "loss": 0.23000378608703614,
      "num_input_tokens_seen": 665694288,
      "step": 11105,
      "train_runtime": 142775.9821,
      "train_tokens_per_second": 4662.509
    },
    {
      "epoch": 1.4170918367346939,
      "grad_norm": 0.7467038466884931,
      "learning_rate": 1.954338728547526e-06,
      "loss": 0.2455808162689209,
      "num_input_tokens_seen": 665983488,
      "step": 11110,
      "train_runtime": 142838.2102,
      "train_tokens_per_second": 4662.502
    },
    {
      "epoch": 1.4177295918367347,
      "grad_norm": 0.7049373808422603,
      "learning_rate": 1.9503678425064097e-06,
      "loss": 0.2417128086090088,
      "num_input_tokens_seen": 666291952,
      "step": 11115,
      "train_runtime": 142890.7179,
      "train_tokens_per_second": 4662.948
    },
    {
      "epoch": 1.4183673469387754,
      "grad_norm": 0.658525510300182,
      "learning_rate": 1.9464000169841495e-06,
      "loss": 0.23396573066711426,
      "num_input_tokens_seen": 666593128,
      "step": 11120,
      "train_runtime": 142947.188,
      "train_tokens_per_second": 4663.213
    },
    {
      "epoch": 1.4190051020408163,
      "grad_norm": 0.6272706114853188,
      "learning_rate": 1.942435255962736e-06,
      "loss": 0.22563095092773439,
      "num_input_tokens_seen": 666904440,
      "step": 11125,
      "train_runtime": 143004.4183,
      "train_tokens_per_second": 4663.523
    },
    {
      "epoch": 1.4196428571428572,
      "grad_norm": 0.6525816640017443,
      "learning_rate": 1.938473563421082e-06,
      "loss": 0.26526608467102053,
      "num_input_tokens_seen": 667199696,
      "step": 11130,
      "train_runtime": 143072.3482,
      "train_tokens_per_second": 4663.373
    },
    {
      "epoch": 1.4202806122448979,
      "grad_norm": 0.6338108193389211,
      "learning_rate": 1.934514943335024e-06,
      "loss": 0.2452698230743408,
      "num_input_tokens_seen": 667504816,
      "step": 11135,
      "train_runtime": 143143.4247,
      "train_tokens_per_second": 4663.189
    },
    {
      "epoch": 1.4209183673469388,
      "grad_norm": 0.6479257772530378,
      "learning_rate": 1.930559399677313e-06,
      "loss": 0.25869481563568114,
      "num_input_tokens_seen": 667813184,
      "step": 11140,
      "train_runtime": 143200.507,
      "train_tokens_per_second": 4663.483
    },
    {
      "epoch": 1.4215561224489797,
      "grad_norm": 0.6691775461223044,
      "learning_rate": 1.9266069364176144e-06,
      "loss": 0.2564568996429443,
      "num_input_tokens_seen": 668122856,
      "step": 11145,
      "train_runtime": 143273.0577,
      "train_tokens_per_second": 4663.283
    },
    {
      "epoch": 1.4221938775510203,
      "grad_norm": 0.727982426741096,
      "learning_rate": 1.9226575575225002e-06,
      "loss": 0.2333740234375,
      "num_input_tokens_seen": 668422168,
      "step": 11150,
      "train_runtime": 143350.7584,
      "train_tokens_per_second": 4662.844
    },
    {
      "epoch": 1.4228316326530612,
      "grad_norm": 0.6397999916750352,
      "learning_rate": 1.918711266955448e-06,
      "loss": 0.24022703170776366,
      "num_input_tokens_seen": 668725624,
      "step": 11155,
      "train_runtime": 143412.3803,
      "train_tokens_per_second": 4662.956
    },
    {
      "epoch": 1.4234693877551021,
      "grad_norm": 0.5903023127390692,
      "learning_rate": 1.914768068676836e-06,
      "loss": 0.23648490905761718,
      "num_input_tokens_seen": 669025552,
      "step": 11160,
      "train_runtime": 143468.6606,
      "train_tokens_per_second": 4663.217
    },
    {
      "epoch": 1.4241071428571428,
      "grad_norm": 0.5430725423767777,
      "learning_rate": 1.9108279666439394e-06,
      "loss": 0.22266921997070313,
      "num_input_tokens_seen": 669328488,
      "step": 11165,
      "train_runtime": 143536.2345,
      "train_tokens_per_second": 4663.133
    },
    {
      "epoch": 1.4247448979591837,
      "grad_norm": 0.7465710670893899,
      "learning_rate": 1.9068909648109258e-06,
      "loss": 0.25256502628326416,
      "num_input_tokens_seen": 669619016,
      "step": 11170,
      "train_runtime": 143612.7017,
      "train_tokens_per_second": 4662.673
    },
    {
      "epoch": 1.4253826530612246,
      "grad_norm": 0.7000445712064963,
      "learning_rate": 1.9029570671288511e-06,
      "loss": 0.25914993286132815,
      "num_input_tokens_seen": 669913296,
      "step": 11175,
      "train_runtime": 143685.6509,
      "train_tokens_per_second": 4662.353
    },
    {
      "epoch": 1.4260204081632653,
      "grad_norm": 0.5943386482918162,
      "learning_rate": 1.899026277545658e-06,
      "loss": 0.25617499351501466,
      "num_input_tokens_seen": 670220176,
      "step": 11180,
      "train_runtime": 143733.4456,
      "train_tokens_per_second": 4662.938
    },
    {
      "epoch": 1.4266581632653061,
      "grad_norm": 0.6692861543609957,
      "learning_rate": 1.895098600006164e-06,
      "loss": 0.260732364654541,
      "num_input_tokens_seen": 670517632,
      "step": 11185,
      "train_runtime": 143791.7164,
      "train_tokens_per_second": 4663.117
    },
    {
      "epoch": 1.427295918367347,
      "grad_norm": 0.735822128187063,
      "learning_rate": 1.8911740384520717e-06,
      "loss": 0.25658307075500486,
      "num_input_tokens_seen": 670812056,
      "step": 11190,
      "train_runtime": 143866.8539,
      "train_tokens_per_second": 4662.728
    },
    {
      "epoch": 1.4279336734693877,
      "grad_norm": 0.6568459317765151,
      "learning_rate": 1.8872525968219541e-06,
      "loss": 0.23559768199920655,
      "num_input_tokens_seen": 671111464,
      "step": 11195,
      "train_runtime": 143934.053,
      "train_tokens_per_second": 4662.632
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.6042734918658409,
      "learning_rate": 1.8833342790512472e-06,
      "loss": 0.2498628616333008,
      "num_input_tokens_seen": 671417184,
      "step": 11200,
      "train_runtime": 144003.2302,
      "train_tokens_per_second": 4662.515
    },
    {
      "epoch": 1.4292091836734695,
      "grad_norm": 0.610036908356831,
      "learning_rate": 1.8794190890722575e-06,
      "loss": 0.23115887641906738,
      "num_input_tokens_seen": 671719040,
      "step": 11205,
      "train_runtime": 144075.867,
      "train_tokens_per_second": 4662.259
    },
    {
      "epoch": 1.4298469387755102,
      "grad_norm": 0.7413819535644164,
      "learning_rate": 1.875507030814156e-06,
      "loss": 0.22059884071350097,
      "num_input_tokens_seen": 672021136,
      "step": 11210,
      "train_runtime": 144134.4175,
      "train_tokens_per_second": 4662.461
    },
    {
      "epoch": 1.430484693877551,
      "grad_norm": 0.5871142382368951,
      "learning_rate": 1.871598108202961e-06,
      "loss": 0.23523523807525634,
      "num_input_tokens_seen": 672323272,
      "step": 11215,
      "train_runtime": 144196.6778,
      "train_tokens_per_second": 4662.543
    },
    {
      "epoch": 1.431122448979592,
      "grad_norm": 0.7144918772638523,
      "learning_rate": 1.8676923251615503e-06,
      "loss": 0.2446314811706543,
      "num_input_tokens_seen": 672618944,
      "step": 11220,
      "train_runtime": 144266.6283,
      "train_tokens_per_second": 4662.332
    },
    {
      "epoch": 1.4317602040816326,
      "grad_norm": 0.6881153338475912,
      "learning_rate": 1.863789685609655e-06,
      "loss": 0.2327864646911621,
      "num_input_tokens_seen": 672909344,
      "step": 11225,
      "train_runtime": 144335.4606,
      "train_tokens_per_second": 4662.121
    },
    {
      "epoch": 1.4323979591836735,
      "grad_norm": 0.6647319198471618,
      "learning_rate": 1.8598901934638402e-06,
      "loss": 0.24272985458374025,
      "num_input_tokens_seen": 673207768,
      "step": 11230,
      "train_runtime": 144393.4466,
      "train_tokens_per_second": 4662.315
    },
    {
      "epoch": 1.4330357142857144,
      "grad_norm": 0.6576414962394639,
      "learning_rate": 1.8559938526375215e-06,
      "loss": 0.2581766128540039,
      "num_input_tokens_seen": 673521112,
      "step": 11235,
      "train_runtime": 144452.7732,
      "train_tokens_per_second": 4662.57
    },
    {
      "epoch": 1.433673469387755,
      "grad_norm": 0.5675080199848578,
      "learning_rate": 1.8521006670409481e-06,
      "loss": 0.22662951946258544,
      "num_input_tokens_seen": 673817848,
      "step": 11240,
      "train_runtime": 144521.4786,
      "train_tokens_per_second": 4662.406
    },
    {
      "epoch": 1.434311224489796,
      "grad_norm": 0.6186972990996671,
      "learning_rate": 1.8482106405812046e-06,
      "loss": 0.25056777000427244,
      "num_input_tokens_seen": 674129160,
      "step": 11245,
      "train_runtime": 144594.2822,
      "train_tokens_per_second": 4662.212
    },
    {
      "epoch": 1.4349489795918369,
      "grad_norm": 0.6803632737356896,
      "learning_rate": 1.8443237771622037e-06,
      "loss": 0.2350076198577881,
      "num_input_tokens_seen": 674424928,
      "step": 11250,
      "train_runtime": 144654.8974,
      "train_tokens_per_second": 4662.303
    },
    {
      "epoch": 1.4355867346938775,
      "grad_norm": 0.6080045064271005,
      "learning_rate": 1.840440080684684e-06,
      "loss": 0.2515897750854492,
      "num_input_tokens_seen": 674722192,
      "step": 11255,
      "train_runtime": 144714.256,
      "train_tokens_per_second": 4662.445
    },
    {
      "epoch": 1.4362244897959184,
      "grad_norm": 0.6251446545007936,
      "learning_rate": 1.8365595550462068e-06,
      "loss": 0.23800618648529054,
      "num_input_tokens_seen": 675033784,
      "step": 11260,
      "train_runtime": 144776.9555,
      "train_tokens_per_second": 4662.578
    },
    {
      "epoch": 1.436862244897959,
      "grad_norm": 0.6363328109366876,
      "learning_rate": 1.8326822041411524e-06,
      "loss": 0.24979538917541505,
      "num_input_tokens_seen": 675333960,
      "step": 11265,
      "train_runtime": 144840.1081,
      "train_tokens_per_second": 4662.617
    },
    {
      "epoch": 1.4375,
      "grad_norm": 0.7408856000320213,
      "learning_rate": 1.828808031860707e-06,
      "loss": 0.2457813262939453,
      "num_input_tokens_seen": 675625264,
      "step": 11270,
      "train_runtime": 144907.527,
      "train_tokens_per_second": 4662.458
    },
    {
      "epoch": 1.438137755102041,
      "grad_norm": 0.6672055278745381,
      "learning_rate": 1.824937042092879e-06,
      "loss": 0.240432071685791,
      "num_input_tokens_seen": 675923008,
      "step": 11275,
      "train_runtime": 144975.1885,
      "train_tokens_per_second": 4662.336
    },
    {
      "epoch": 1.4387755102040816,
      "grad_norm": 0.6858424030694148,
      "learning_rate": 1.8210692387224776e-06,
      "loss": 0.23592252731323243,
      "num_input_tokens_seen": 676226016,
      "step": 11280,
      "train_runtime": 145039.8188,
      "train_tokens_per_second": 4662.347
    },
    {
      "epoch": 1.4394132653061225,
      "grad_norm": 0.5793406863901901,
      "learning_rate": 1.8172046256311087e-06,
      "loss": 0.23383541107177735,
      "num_input_tokens_seen": 676536832,
      "step": 11285,
      "train_runtime": 145098.8775,
      "train_tokens_per_second": 4662.592
    },
    {
      "epoch": 1.4400510204081631,
      "grad_norm": 0.6690683138041699,
      "learning_rate": 1.8133432066971823e-06,
      "loss": 0.24509520530700685,
      "num_input_tokens_seen": 676847736,
      "step": 11290,
      "train_runtime": 145171.9807,
      "train_tokens_per_second": 4662.385
    },
    {
      "epoch": 1.440688775510204,
      "grad_norm": 0.5989691395247146,
      "learning_rate": 1.8094849857959074e-06,
      "loss": 0.25073742866516113,
      "num_input_tokens_seen": 677149832,
      "step": 11295,
      "train_runtime": 145231.4993,
      "train_tokens_per_second": 4662.555
    },
    {
      "epoch": 1.441326530612245,
      "grad_norm": 0.619323780223799,
      "learning_rate": 1.8056299667992721e-06,
      "loss": 0.23592658042907716,
      "num_input_tokens_seen": 677444904,
      "step": 11300,
      "train_runtime": 145294.1911,
      "train_tokens_per_second": 4662.574
    },
    {
      "epoch": 1.4419642857142856,
      "grad_norm": 0.689840601159879,
      "learning_rate": 1.8017781535760581e-06,
      "loss": 0.26845948696136473,
      "num_input_tokens_seen": 677738560,
      "step": 11305,
      "train_runtime": 145348.7561,
      "train_tokens_per_second": 4662.844
    },
    {
      "epoch": 1.4426020408163265,
      "grad_norm": 0.6678270003814463,
      "learning_rate": 1.7979295499918338e-06,
      "loss": 0.25097291469573973,
      "num_input_tokens_seen": 678031816,
      "step": 11310,
      "train_runtime": 145409.6422,
      "train_tokens_per_second": 4662.908
    },
    {
      "epoch": 1.4432397959183674,
      "grad_norm": 0.6327321732567487,
      "learning_rate": 1.7940841599089364e-06,
      "loss": 0.24874565601348878,
      "num_input_tokens_seen": 678344552,
      "step": 11315,
      "train_runtime": 145477.3844,
      "train_tokens_per_second": 4662.887
    },
    {
      "epoch": 1.443877551020408,
      "grad_norm": 0.6727392456788983,
      "learning_rate": 1.790241987186485e-06,
      "loss": 0.24608402252197265,
      "num_input_tokens_seen": 678649048,
      "step": 11320,
      "train_runtime": 145537.4687,
      "train_tokens_per_second": 4663.054
    },
    {
      "epoch": 1.444515306122449,
      "grad_norm": 0.6632799743301652,
      "learning_rate": 1.7864030356803685e-06,
      "loss": 0.2536062240600586,
      "num_input_tokens_seen": 678942936,
      "step": 11325,
      "train_runtime": 145607.5696,
      "train_tokens_per_second": 4662.827
    },
    {
      "epoch": 1.4451530612244898,
      "grad_norm": 0.6491202280008166,
      "learning_rate": 1.7825673092432433e-06,
      "loss": 0.24754681587219238,
      "num_input_tokens_seen": 679245312,
      "step": 11330,
      "train_runtime": 145678.9847,
      "train_tokens_per_second": 4662.617
    },
    {
      "epoch": 1.4457908163265305,
      "grad_norm": 0.6619206219092385,
      "learning_rate": 1.7787348117245278e-06,
      "loss": 0.24240462779998778,
      "num_input_tokens_seen": 679539432,
      "step": 11335,
      "train_runtime": 145733.6449,
      "train_tokens_per_second": 4662.886
    },
    {
      "epoch": 1.4464285714285714,
      "grad_norm": 0.6689448881262019,
      "learning_rate": 1.7749055469704013e-06,
      "loss": 0.2488102436065674,
      "num_input_tokens_seen": 679830984,
      "step": 11340,
      "train_runtime": 145804.1031,
      "train_tokens_per_second": 4662.633
    },
    {
      "epoch": 1.4470663265306123,
      "grad_norm": 0.6076813000624108,
      "learning_rate": 1.771079518823799e-06,
      "loss": 0.23630599975585936,
      "num_input_tokens_seen": 680146240,
      "step": 11345,
      "train_runtime": 145858.9289,
      "train_tokens_per_second": 4663.042
    },
    {
      "epoch": 1.447704081632653,
      "grad_norm": 0.6841996950195385,
      "learning_rate": 1.7672567311244065e-06,
      "loss": 0.24865059852600097,
      "num_input_tokens_seen": 680448640,
      "step": 11350,
      "train_runtime": 145929.9343,
      "train_tokens_per_second": 4662.845
    },
    {
      "epoch": 1.4483418367346939,
      "grad_norm": 0.6220900760297762,
      "learning_rate": 1.7634371877086587e-06,
      "loss": 0.24988579750061035,
      "num_input_tokens_seen": 680757512,
      "step": 11355,
      "train_runtime": 146003.4495,
      "train_tokens_per_second": 4662.613
    },
    {
      "epoch": 1.4489795918367347,
      "grad_norm": 0.7023743281398612,
      "learning_rate": 1.7596208924097336e-06,
      "loss": 0.2530750513076782,
      "num_input_tokens_seen": 681058960,
      "step": 11360,
      "train_runtime": 146059.2871,
      "train_tokens_per_second": 4662.894
    },
    {
      "epoch": 1.4496173469387754,
      "grad_norm": 0.6872815302676676,
      "learning_rate": 1.7558078490575531e-06,
      "loss": 0.2638883113861084,
      "num_input_tokens_seen": 681353488,
      "step": 11365,
      "train_runtime": 146123.4241,
      "train_tokens_per_second": 4662.863
    },
    {
      "epoch": 1.4502551020408163,
      "grad_norm": 0.6442783314358508,
      "learning_rate": 1.7519980614787673e-06,
      "loss": 0.2459571123123169,
      "num_input_tokens_seen": 681651648,
      "step": 11370,
      "train_runtime": 146193.3064,
      "train_tokens_per_second": 4662.673
    },
    {
      "epoch": 1.4508928571428572,
      "grad_norm": 0.7182039323690438,
      "learning_rate": 1.748191533496766e-06,
      "loss": 0.23998775482177734,
      "num_input_tokens_seen": 681952928,
      "step": 11375,
      "train_runtime": 146257.6037,
      "train_tokens_per_second": 4662.684
    },
    {
      "epoch": 1.4515306122448979,
      "grad_norm": 0.6588074207114061,
      "learning_rate": 1.7443882689316693e-06,
      "loss": 0.2377242088317871,
      "num_input_tokens_seen": 682263008,
      "step": 11380,
      "train_runtime": 146320.5743,
      "train_tokens_per_second": 4662.796
    },
    {
      "epoch": 1.4521683673469388,
      "grad_norm": 0.7402664092565486,
      "learning_rate": 1.7405882716003154e-06,
      "loss": 0.23890118598937987,
      "num_input_tokens_seen": 682569416,
      "step": 11385,
      "train_runtime": 146382.9856,
      "train_tokens_per_second": 4662.901
    },
    {
      "epoch": 1.4528061224489797,
      "grad_norm": 0.6783340192705669,
      "learning_rate": 1.7367915453162653e-06,
      "loss": 0.25989651679992676,
      "num_input_tokens_seen": 682873896,
      "step": 11390,
      "train_runtime": 146446.7615,
      "train_tokens_per_second": 4662.95
    },
    {
      "epoch": 1.4534438775510203,
      "grad_norm": 0.6424306017085222,
      "learning_rate": 1.732998093889805e-06,
      "loss": 0.24322919845581054,
      "num_input_tokens_seen": 683181792,
      "step": 11395,
      "train_runtime": 146506.5491,
      "train_tokens_per_second": 4663.148
    },
    {
      "epoch": 1.4540816326530612,
      "grad_norm": 0.6450808939537568,
      "learning_rate": 1.7292079211279229e-06,
      "loss": 0.22780027389526367,
      "num_input_tokens_seen": 683480992,
      "step": 11400,
      "train_runtime": 146562.9645,
      "train_tokens_per_second": 4663.395
    },
    {
      "epoch": 1.4547193877551021,
      "grad_norm": 0.7228284999350758,
      "learning_rate": 1.7254210308343233e-06,
      "loss": 0.25200209617614744,
      "num_input_tokens_seen": 683777648,
      "step": 11405,
      "train_runtime": 146621.814,
      "train_tokens_per_second": 4663.547
    },
    {
      "epoch": 1.4553571428571428,
      "grad_norm": 0.6310647373225897,
      "learning_rate": 1.721637426809415e-06,
      "loss": 0.24065794944763183,
      "num_input_tokens_seen": 684081592,
      "step": 11410,
      "train_runtime": 146674.0932,
      "train_tokens_per_second": 4663.957
    },
    {
      "epoch": 1.4559948979591837,
      "grad_norm": 0.6400454895702979,
      "learning_rate": 1.7178571128503097e-06,
      "loss": 0.25054054260253905,
      "num_input_tokens_seen": 684365064,
      "step": 11415,
      "train_runtime": 146736.9551,
      "train_tokens_per_second": 4663.89
    },
    {
      "epoch": 1.4566326530612246,
      "grad_norm": 0.7022991364917267,
      "learning_rate": 1.7140800927508162e-06,
      "loss": 0.25025479793548583,
      "num_input_tokens_seen": 684660616,
      "step": 11420,
      "train_runtime": 146804.0507,
      "train_tokens_per_second": 4663.772
    },
    {
      "epoch": 1.4572704081632653,
      "grad_norm": 0.7418607704263092,
      "learning_rate": 1.7103063703014372e-06,
      "loss": 0.23921337127685546,
      "num_input_tokens_seen": 684961488,
      "step": 11425,
      "train_runtime": 146869.0023,
      "train_tokens_per_second": 4663.758
    },
    {
      "epoch": 1.4579081632653061,
      "grad_norm": 0.6749020113610398,
      "learning_rate": 1.706535949289368e-06,
      "loss": 0.2462780237197876,
      "num_input_tokens_seen": 685265784,
      "step": 11430,
      "train_runtime": 146926.5781,
      "train_tokens_per_second": 4664.002
    },
    {
      "epoch": 1.458545918367347,
      "grad_norm": 0.6738435636019064,
      "learning_rate": 1.7027688334984881e-06,
      "loss": 0.26296138763427734,
      "num_input_tokens_seen": 685571088,
      "step": 11435,
      "train_runtime": 146973.3196,
      "train_tokens_per_second": 4664.596
    },
    {
      "epoch": 1.4591836734693877,
      "grad_norm": 0.7166059957038663,
      "learning_rate": 1.6990050267093621e-06,
      "loss": 0.24450583457946778,
      "num_input_tokens_seen": 685872576,
      "step": 11440,
      "train_runtime": 147043.834,
      "train_tokens_per_second": 4664.409
    },
    {
      "epoch": 1.4598214285714286,
      "grad_norm": 0.6345409777464979,
      "learning_rate": 1.6952445326992327e-06,
      "loss": 0.23451249599456786,
      "num_input_tokens_seen": 686170704,
      "step": 11445,
      "train_runtime": 147107.345,
      "train_tokens_per_second": 4664.422
    },
    {
      "epoch": 1.4604591836734695,
      "grad_norm": 0.6647957595421151,
      "learning_rate": 1.6914873552420185e-06,
      "loss": 0.23938865661621095,
      "num_input_tokens_seen": 686471528,
      "step": 11450,
      "train_runtime": 147166.4848,
      "train_tokens_per_second": 4664.591
    },
    {
      "epoch": 1.4610969387755102,
      "grad_norm": 0.6064102890723383,
      "learning_rate": 1.687733498108311e-06,
      "loss": 0.2332664966583252,
      "num_input_tokens_seen": 686760360,
      "step": 11455,
      "train_runtime": 147223.8005,
      "train_tokens_per_second": 4664.737
    },
    {
      "epoch": 1.461734693877551,
      "grad_norm": 0.6355682273842634,
      "learning_rate": 1.6839829650653632e-06,
      "loss": 0.25306456089019774,
      "num_input_tokens_seen": 687070112,
      "step": 11460,
      "train_runtime": 147289.8375,
      "train_tokens_per_second": 4664.749
    },
    {
      "epoch": 1.462372448979592,
      "grad_norm": 0.5902718165267166,
      "learning_rate": 1.6802357598771014e-06,
      "loss": 0.23483102321624755,
      "num_input_tokens_seen": 687364496,
      "step": 11465,
      "train_runtime": 147352.4819,
      "train_tokens_per_second": 4664.764
    },
    {
      "epoch": 1.4630102040816326,
      "grad_norm": 0.6038581890225789,
      "learning_rate": 1.6764918863041079e-06,
      "loss": 0.24713358879089356,
      "num_input_tokens_seen": 687659152,
      "step": 11470,
      "train_runtime": 147407.6101,
      "train_tokens_per_second": 4665.018
    },
    {
      "epoch": 1.4636479591836735,
      "grad_norm": 0.6855517028015412,
      "learning_rate": 1.6727513481036167e-06,
      "loss": 0.25162179470062257,
      "num_input_tokens_seen": 687968872,
      "step": 11475,
      "train_runtime": 147464.9396,
      "train_tokens_per_second": 4665.305
    },
    {
      "epoch": 1.4642857142857144,
      "grad_norm": 0.6688300145778381,
      "learning_rate": 1.6690141490295259e-06,
      "loss": 0.22452807426452637,
      "num_input_tokens_seen": 688271208,
      "step": 11480,
      "train_runtime": 147530.6542,
      "train_tokens_per_second": 4665.276
    },
    {
      "epoch": 1.464923469387755,
      "grad_norm": 0.6924416393372076,
      "learning_rate": 1.6652802928323698e-06,
      "loss": 0.2321218490600586,
      "num_input_tokens_seen": 688573560,
      "step": 11485,
      "train_runtime": 147593.8325,
      "train_tokens_per_second": 4665.327
    },
    {
      "epoch": 1.465561224489796,
      "grad_norm": 0.6968472670839632,
      "learning_rate": 1.6615497832593365e-06,
      "loss": 0.2492440700531006,
      "num_input_tokens_seen": 688866760,
      "step": 11490,
      "train_runtime": 147663.5412,
      "train_tokens_per_second": 4665.111
    },
    {
      "epoch": 1.4661989795918369,
      "grad_norm": 0.7560208215184506,
      "learning_rate": 1.6578226240542528e-06,
      "loss": 0.25133914947509767,
      "num_input_tokens_seen": 689178576,
      "step": 11495,
      "train_runtime": 147729.2253,
      "train_tokens_per_second": 4665.147
    },
    {
      "epoch": 1.4668367346938775,
      "grad_norm": 0.665168527803295,
      "learning_rate": 1.6540988189575823e-06,
      "loss": 0.24690823554992675,
      "num_input_tokens_seen": 689463784,
      "step": 11500,
      "train_runtime": 147790.1147,
      "train_tokens_per_second": 4665.155
    },
    {
      "epoch": 1.4674744897959184,
      "grad_norm": 0.596987753677735,
      "learning_rate": 1.6503783717064247e-06,
      "loss": 0.2441627025604248,
      "num_input_tokens_seen": 689771064,
      "step": 11505,
      "train_runtime": 147862.9351,
      "train_tokens_per_second": 4664.936
    },
    {
      "epoch": 1.468112244897959,
      "grad_norm": 0.6219667258131122,
      "learning_rate": 1.6466612860345083e-06,
      "loss": 0.24189579486846924,
      "num_input_tokens_seen": 690056536,
      "step": 11510,
      "train_runtime": 147933.1245,
      "train_tokens_per_second": 4664.652
    },
    {
      "epoch": 1.46875,
      "grad_norm": 0.6721980454746388,
      "learning_rate": 1.6429475656721878e-06,
      "loss": 0.25978860855102537,
      "num_input_tokens_seen": 690365424,
      "step": 11515,
      "train_runtime": 147998.8208,
      "train_tokens_per_second": 4664.668
    },
    {
      "epoch": 1.469387755102041,
      "grad_norm": 0.6583685490789873,
      "learning_rate": 1.6392372143464407e-06,
      "loss": 0.24137365818023682,
      "num_input_tokens_seen": 690644680,
      "step": 11520,
      "train_runtime": 148073.4402,
      "train_tokens_per_second": 4664.204
    },
    {
      "epoch": 1.4700255102040816,
      "grad_norm": 0.6512613176517742,
      "learning_rate": 1.635530235780865e-06,
      "loss": 0.22860364913940429,
      "num_input_tokens_seen": 690923952,
      "step": 11525,
      "train_runtime": 148140.9507,
      "train_tokens_per_second": 4663.963
    },
    {
      "epoch": 1.4706632653061225,
      "grad_norm": 0.710859298839146,
      "learning_rate": 1.6318266336956713e-06,
      "loss": 0.22747769355773925,
      "num_input_tokens_seen": 691234968,
      "step": 11530,
      "train_runtime": 148204.3314,
      "train_tokens_per_second": 4664.067
    },
    {
      "epoch": 1.4713010204081631,
      "grad_norm": 0.6175456058852351,
      "learning_rate": 1.6281264118076833e-06,
      "loss": 0.2350527286529541,
      "num_input_tokens_seen": 691519072,
      "step": 11535,
      "train_runtime": 148256.5819,
      "train_tokens_per_second": 4664.34
    },
    {
      "epoch": 1.471938775510204,
      "grad_norm": 0.6416467457340852,
      "learning_rate": 1.6244295738303323e-06,
      "loss": 0.22203752994537354,
      "num_input_tokens_seen": 691816536,
      "step": 11540,
      "train_runtime": 148325.683,
      "train_tokens_per_second": 4664.172
    },
    {
      "epoch": 1.472576530612245,
      "grad_norm": 0.6682029999306813,
      "learning_rate": 1.6207361234736535e-06,
      "loss": 0.2506797552108765,
      "num_input_tokens_seen": 692126752,
      "step": 11545,
      "train_runtime": 148379.8864,
      "train_tokens_per_second": 4664.559
    },
    {
      "epoch": 1.4732142857142856,
      "grad_norm": 0.6975639493472159,
      "learning_rate": 1.6170460644442814e-06,
      "loss": 0.2557785987854004,
      "num_input_tokens_seen": 692426824,
      "step": 11550,
      "train_runtime": 148439.7907,
      "train_tokens_per_second": 4664.698
    },
    {
      "epoch": 1.4738520408163265,
      "grad_norm": 0.7181680193863293,
      "learning_rate": 1.6133594004454512e-06,
      "loss": 0.23441197872161865,
      "num_input_tokens_seen": 692713504,
      "step": 11555,
      "train_runtime": 148511.3149,
      "train_tokens_per_second": 4664.382
    },
    {
      "epoch": 1.4744897959183674,
      "grad_norm": 0.6459720024804818,
      "learning_rate": 1.6096761351769813e-06,
      "loss": 0.22232742309570314,
      "num_input_tokens_seen": 693002160,
      "step": 11560,
      "train_runtime": 148579.0408,
      "train_tokens_per_second": 4664.199
    },
    {
      "epoch": 1.475127551020408,
      "grad_norm": 0.678127980893922,
      "learning_rate": 1.6059962723352912e-06,
      "loss": 0.27894728183746337,
      "num_input_tokens_seen": 693315720,
      "step": 11565,
      "train_runtime": 148633.1422,
      "train_tokens_per_second": 4664.611
    },
    {
      "epoch": 1.475765306122449,
      "grad_norm": 0.6095747148965968,
      "learning_rate": 1.6023198156133796e-06,
      "loss": 0.24708163738250732,
      "num_input_tokens_seen": 693623040,
      "step": 11570,
      "train_runtime": 148687.3312,
      "train_tokens_per_second": 4664.977
    },
    {
      "epoch": 1.4764030612244898,
      "grad_norm": 0.6163226224096391,
      "learning_rate": 1.5986467687008256e-06,
      "loss": 0.23858215808868408,
      "num_input_tokens_seen": 693920448,
      "step": 11575,
      "train_runtime": 148756.2653,
      "train_tokens_per_second": 4664.815
    },
    {
      "epoch": 1.4770408163265305,
      "grad_norm": 0.6226961547586967,
      "learning_rate": 1.5949771352837873e-06,
      "loss": 0.2381221055984497,
      "num_input_tokens_seen": 694213632,
      "step": 11580,
      "train_runtime": 148818.0948,
      "train_tokens_per_second": 4664.847
    },
    {
      "epoch": 1.4776785714285714,
      "grad_norm": 0.7560406614750288,
      "learning_rate": 1.5913109190450033e-06,
      "loss": 0.25000662803649903,
      "num_input_tokens_seen": 694509360,
      "step": 11585,
      "train_runtime": 148879.4523,
      "train_tokens_per_second": 4664.911
    },
    {
      "epoch": 1.4783163265306123,
      "grad_norm": 0.580647102783899,
      "learning_rate": 1.5876481236637725e-06,
      "loss": 0.22563493251800537,
      "num_input_tokens_seen": 694824952,
      "step": 11590,
      "train_runtime": 148940.4976,
      "train_tokens_per_second": 4665.118
    },
    {
      "epoch": 1.478954081632653,
      "grad_norm": 0.714166697783697,
      "learning_rate": 1.5839887528159664e-06,
      "loss": 0.2413116455078125,
      "num_input_tokens_seen": 695137000,
      "step": 11595,
      "train_runtime": 149007.5951,
      "train_tokens_per_second": 4665.111
    },
    {
      "epoch": 1.4795918367346939,
      "grad_norm": 0.6258859637092434,
      "learning_rate": 1.580332810174024e-06,
      "loss": 0.2342731237411499,
      "num_input_tokens_seen": 695460296,
      "step": 11600,
      "train_runtime": 149061.999,
      "train_tokens_per_second": 4665.577
    },
    {
      "epoch": 1.4802295918367347,
      "grad_norm": 0.6801170050318642,
      "learning_rate": 1.5766802994069336e-06,
      "loss": 0.2347461223602295,
      "num_input_tokens_seen": 695764160,
      "step": 11605,
      "train_runtime": 149120.1798,
      "train_tokens_per_second": 4665.795
    },
    {
      "epoch": 1.4808673469387754,
      "grad_norm": 0.6262416405613398,
      "learning_rate": 1.573031224180247e-06,
      "loss": 0.2373434543609619,
      "num_input_tokens_seen": 696074920,
      "step": 11610,
      "train_runtime": 149188.6907,
      "train_tokens_per_second": 4665.735
    },
    {
      "epoch": 1.4815051020408163,
      "grad_norm": 0.6918457360931097,
      "learning_rate": 1.5693855881560654e-06,
      "loss": 0.2384523868560791,
      "num_input_tokens_seen": 696379304,
      "step": 11615,
      "train_runtime": 149263.133,
      "train_tokens_per_second": 4665.447
    },
    {
      "epoch": 1.4821428571428572,
      "grad_norm": 0.6865367278673672,
      "learning_rate": 1.5657433949930406e-06,
      "loss": 0.25263969898223876,
      "num_input_tokens_seen": 696681120,
      "step": 11620,
      "train_runtime": 149324.3801,
      "train_tokens_per_second": 4665.555
    },
    {
      "epoch": 1.4827806122448979,
      "grad_norm": 0.7296782473533788,
      "learning_rate": 1.5621046483463663e-06,
      "loss": 0.23917748928070068,
      "num_input_tokens_seen": 696991904,
      "step": 11625,
      "train_runtime": 149389.5322,
      "train_tokens_per_second": 4665.601
    },
    {
      "epoch": 1.4834183673469388,
      "grad_norm": 0.6664840276441146,
      "learning_rate": 1.5584693518677796e-06,
      "loss": 0.23252415657043457,
      "num_input_tokens_seen": 697283336,
      "step": 11630,
      "train_runtime": 149461.8494,
      "train_tokens_per_second": 4665.293
    },
    {
      "epoch": 1.4840561224489797,
      "grad_norm": 0.634565462518508,
      "learning_rate": 1.554837509205554e-06,
      "loss": 0.23175759315490724,
      "num_input_tokens_seen": 697581768,
      "step": 11635,
      "train_runtime": 149525.9994,
      "train_tokens_per_second": 4665.287
    },
    {
      "epoch": 1.4846938775510203,
      "grad_norm": 0.6221968571807043,
      "learning_rate": 1.5512091240044996e-06,
      "loss": 0.2360471487045288,
      "num_input_tokens_seen": 697879912,
      "step": 11640,
      "train_runtime": 149582.5311,
      "train_tokens_per_second": 4665.517
    },
    {
      "epoch": 1.4853316326530612,
      "grad_norm": 0.7540079611971362,
      "learning_rate": 1.5475841999059487e-06,
      "loss": 0.24431324005126953,
      "num_input_tokens_seen": 698194592,
      "step": 11645,
      "train_runtime": 149646.4544,
      "train_tokens_per_second": 4665.627
    },
    {
      "epoch": 1.4859693877551021,
      "grad_norm": 0.6705837388418601,
      "learning_rate": 1.5439627405477714e-06,
      "loss": 0.22821378707885742,
      "num_input_tokens_seen": 698513160,
      "step": 11650,
      "train_runtime": 149710.3094,
      "train_tokens_per_second": 4665.765
    },
    {
      "epoch": 1.4866071428571428,
      "grad_norm": 0.5772512077798021,
      "learning_rate": 1.5403447495643542e-06,
      "loss": 0.26444687843322756,
      "num_input_tokens_seen": 698821256,
      "step": 11655,
      "train_runtime": 149784.7413,
      "train_tokens_per_second": 4665.504
    },
    {
      "epoch": 1.4872448979591837,
      "grad_norm": 0.6237181341790158,
      "learning_rate": 1.5367302305866017e-06,
      "loss": 0.23810076713562012,
      "num_input_tokens_seen": 699098912,
      "step": 11660,
      "train_runtime": 149848.2229,
      "train_tokens_per_second": 4665.38
    },
    {
      "epoch": 1.4878826530612246,
      "grad_norm": 0.6226314180774907,
      "learning_rate": 1.5331191872419349e-06,
      "loss": 0.2813971757888794,
      "num_input_tokens_seen": 699396688,
      "step": 11665,
      "train_runtime": 149908.4462,
      "train_tokens_per_second": 4665.492
    },
    {
      "epoch": 1.4885204081632653,
      "grad_norm": 0.6437744808395093,
      "learning_rate": 1.5295116231542935e-06,
      "loss": 0.25315122604370116,
      "num_input_tokens_seen": 699700688,
      "step": 11670,
      "train_runtime": 149970.6227,
      "train_tokens_per_second": 4665.585
    },
    {
      "epoch": 1.4891581632653061,
      "grad_norm": 0.6235078179858374,
      "learning_rate": 1.5259075419441144e-06,
      "loss": 0.23633928298950196,
      "num_input_tokens_seen": 699994016,
      "step": 11675,
      "train_runtime": 150038.3311,
      "train_tokens_per_second": 4665.435
    },
    {
      "epoch": 1.489795918367347,
      "grad_norm": 0.6305730504230714,
      "learning_rate": 1.5223069472283464e-06,
      "loss": 0.24948687553405763,
      "num_input_tokens_seen": 700297120,
      "step": 11680,
      "train_runtime": 150104.4626,
      "train_tokens_per_second": 4665.398
    },
    {
      "epoch": 1.4904336734693877,
      "grad_norm": 0.6638784368909486,
      "learning_rate": 1.5187098426204411e-06,
      "loss": 0.22751235961914062,
      "num_input_tokens_seen": 700597472,
      "step": 11685,
      "train_runtime": 150160.3655,
      "train_tokens_per_second": 4665.662
    },
    {
      "epoch": 1.4910714285714286,
      "grad_norm": 0.6505087899264166,
      "learning_rate": 1.5151162317303398e-06,
      "loss": 0.24964916706085205,
      "num_input_tokens_seen": 700900528,
      "step": 11690,
      "train_runtime": 150231.9071,
      "train_tokens_per_second": 4665.457
    },
    {
      "epoch": 1.4917091836734695,
      "grad_norm": 0.7077708829756935,
      "learning_rate": 1.5115261181644842e-06,
      "loss": 0.21147603988647462,
      "num_input_tokens_seen": 701190128,
      "step": 11695,
      "train_runtime": 150294.6782,
      "train_tokens_per_second": 4665.436
    },
    {
      "epoch": 1.4923469387755102,
      "grad_norm": 0.5979843001639947,
      "learning_rate": 1.5079395055258027e-06,
      "loss": 0.2485567569732666,
      "num_input_tokens_seen": 701489056,
      "step": 11700,
      "train_runtime": 150362.9766,
      "train_tokens_per_second": 4665.304
    },
    {
      "epoch": 1.492984693877551,
      "grad_norm": 0.6409453647709838,
      "learning_rate": 1.5043563974137132e-06,
      "loss": 0.2364804267883301,
      "num_input_tokens_seen": 701786760,
      "step": 11705,
      "train_runtime": 150422.4212,
      "train_tokens_per_second": 4665.44
    },
    {
      "epoch": 1.493622448979592,
      "grad_norm": 0.641753447407374,
      "learning_rate": 1.5007767974241133e-06,
      "loss": 0.2340480089187622,
      "num_input_tokens_seen": 702077144,
      "step": 11710,
      "train_runtime": 150481.3884,
      "train_tokens_per_second": 4665.541
    },
    {
      "epoch": 1.4942602040816326,
      "grad_norm": 0.7290957319300447,
      "learning_rate": 1.497200709149383e-06,
      "loss": 0.25794596672058107,
      "num_input_tokens_seen": 702379760,
      "step": 11715,
      "train_runtime": 150545.1891,
      "train_tokens_per_second": 4665.574
    },
    {
      "epoch": 1.4948979591836735,
      "grad_norm": 0.5971458042728834,
      "learning_rate": 1.493628136178376e-06,
      "loss": 0.23512799739837648,
      "num_input_tokens_seen": 702679024,
      "step": 11720,
      "train_runtime": 150602.408,
      "train_tokens_per_second": 4665.789
    },
    {
      "epoch": 1.4955357142857144,
      "grad_norm": 0.6084064688795197,
      "learning_rate": 1.49005908209642e-06,
      "loss": 0.24218168258666992,
      "num_input_tokens_seen": 702983968,
      "step": 11725,
      "train_runtime": 150676.369,
      "train_tokens_per_second": 4665.522
    },
    {
      "epoch": 1.496173469387755,
      "grad_norm": 0.6549253711934818,
      "learning_rate": 1.4864935504853062e-06,
      "loss": 0.268617582321167,
      "num_input_tokens_seen": 703286088,
      "step": 11730,
      "train_runtime": 150721.7914,
      "train_tokens_per_second": 4666.121
    },
    {
      "epoch": 1.496811224489796,
      "grad_norm": 0.6262180880397928,
      "learning_rate": 1.4829315449232984e-06,
      "loss": 0.2677947521209717,
      "num_input_tokens_seen": 703596696,
      "step": 11735,
      "train_runtime": 150787.5204,
      "train_tokens_per_second": 4666.147
    },
    {
      "epoch": 1.4974489795918369,
      "grad_norm": 0.674671179788959,
      "learning_rate": 1.4793730689851187e-06,
      "loss": 0.2631994247436523,
      "num_input_tokens_seen": 703889592,
      "step": 11740,
      "train_runtime": 150858.4299,
      "train_tokens_per_second": 4665.895
    },
    {
      "epoch": 1.4980867346938775,
      "grad_norm": 0.62023577413827,
      "learning_rate": 1.4758181262419425e-06,
      "loss": 0.22229366302490233,
      "num_input_tokens_seen": 704193912,
      "step": 11745,
      "train_runtime": 150929.1748,
      "train_tokens_per_second": 4665.724
    },
    {
      "epoch": 1.4987244897959184,
      "grad_norm": 0.6130633296511628,
      "learning_rate": 1.4722667202614037e-06,
      "loss": 0.24085865020751954,
      "num_input_tokens_seen": 704498760,
      "step": 11750,
      "train_runtime": 150996.3203,
      "train_tokens_per_second": 4665.668
    },
    {
      "epoch": 1.499362244897959,
      "grad_norm": 0.6780347042669347,
      "learning_rate": 1.4687188546075903e-06,
      "loss": 0.2245311737060547,
      "num_input_tokens_seen": 704801800,
      "step": 11755,
      "train_runtime": 151048.7665,
      "train_tokens_per_second": 4666.055
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.6642715489613907,
      "learning_rate": 1.4651745328410289e-06,
      "loss": 0.22231788635253907,
      "num_input_tokens_seen": 705104568,
      "step": 11760,
      "train_runtime": 151101.0487,
      "train_tokens_per_second": 4666.444
    },
    {
      "epoch": 1.5006377551020407,
      "grad_norm": 0.6590290554213085,
      "learning_rate": 1.4616337585186941e-06,
      "loss": 0.2275834083557129,
      "num_input_tokens_seen": 705392808,
      "step": 11765,
      "train_runtime": 151174.6671,
      "train_tokens_per_second": 4666.078
    },
    {
      "epoch": 1.5012755102040818,
      "grad_norm": 0.6242464491086078,
      "learning_rate": 1.4580965351940045e-06,
      "loss": 0.24281435012817382,
      "num_input_tokens_seen": 705684680,
      "step": 11770,
      "train_runtime": 151240.0255,
      "train_tokens_per_second": 4665.992
    },
    {
      "epoch": 1.5019132653061225,
      "grad_norm": 0.622983279791885,
      "learning_rate": 1.454562866416806e-06,
      "loss": 0.24083242416381836,
      "num_input_tokens_seen": 705972368,
      "step": 11775,
      "train_runtime": 151312.9802,
      "train_tokens_per_second": 4665.643
    },
    {
      "epoch": 1.5025510204081631,
      "grad_norm": 0.8005802415466808,
      "learning_rate": 1.4510327557333843e-06,
      "loss": 0.26150596141815186,
      "num_input_tokens_seen": 706267752,
      "step": 11780,
      "train_runtime": 151375.5526,
      "train_tokens_per_second": 4665.666
    },
    {
      "epoch": 1.5031887755102042,
      "grad_norm": 0.785628133606859,
      "learning_rate": 1.4475062066864516e-06,
      "loss": 0.24231288433074952,
      "num_input_tokens_seen": 706568992,
      "step": 11785,
      "train_runtime": 151438.4906,
      "train_tokens_per_second": 4665.716
    },
    {
      "epoch": 1.503826530612245,
      "grad_norm": 0.7498550910124094,
      "learning_rate": 1.4439832228151457e-06,
      "loss": 0.23139348030090331,
      "num_input_tokens_seen": 706865744,
      "step": 11790,
      "train_runtime": 151509.1242,
      "train_tokens_per_second": 4665.5
    },
    {
      "epoch": 1.5044642857142856,
      "grad_norm": 0.6264347831594215,
      "learning_rate": 1.4404638076550276e-06,
      "loss": 0.24553780555725097,
      "num_input_tokens_seen": 707182696,
      "step": 11795,
      "train_runtime": 151565.6277,
      "train_tokens_per_second": 4665.851
    },
    {
      "epoch": 1.5051020408163265,
      "grad_norm": 0.6780204974233415,
      "learning_rate": 1.4369479647380758e-06,
      "loss": 0.2568678617477417,
      "num_input_tokens_seen": 707491832,
      "step": 11800,
      "train_runtime": 151617.6307,
      "train_tokens_per_second": 4666.29
    },
    {
      "epoch": 1.5057397959183674,
      "grad_norm": 0.5700173324692831,
      "learning_rate": 1.4334356975926843e-06,
      "loss": 0.2317802667617798,
      "num_input_tokens_seen": 707803072,
      "step": 11805,
      "train_runtime": 151683.0667,
      "train_tokens_per_second": 4666.329
    },
    {
      "epoch": 1.506377551020408,
      "grad_norm": 0.6842625551689768,
      "learning_rate": 1.429927009743659e-06,
      "loss": 0.23446249961853027,
      "num_input_tokens_seen": 708095776,
      "step": 11810,
      "train_runtime": 151752.5273,
      "train_tokens_per_second": 4666.122
    },
    {
      "epoch": 1.507015306122449,
      "grad_norm": 0.5955782129102939,
      "learning_rate": 1.4264219047122125e-06,
      "loss": 0.2527336597442627,
      "num_input_tokens_seen": 708407072,
      "step": 11815,
      "train_runtime": 151812.5889,
      "train_tokens_per_second": 4666.326
    },
    {
      "epoch": 1.5076530612244898,
      "grad_norm": 0.6946166566371212,
      "learning_rate": 1.422920386015963e-06,
      "loss": 0.23975543975830077,
      "num_input_tokens_seen": 708671520,
      "step": 11820,
      "train_runtime": 151878.4488,
      "train_tokens_per_second": 4666.044
    },
    {
      "epoch": 1.5082908163265305,
      "grad_norm": 0.6760653277896207,
      "learning_rate": 1.4194224571689286e-06,
      "loss": 0.2433457612991333,
      "num_input_tokens_seen": 708959944,
      "step": 11825,
      "train_runtime": 151945.957,
      "train_tokens_per_second": 4665.869
    },
    {
      "epoch": 1.5089285714285714,
      "grad_norm": 0.7344592881272071,
      "learning_rate": 1.4159281216815269e-06,
      "loss": 0.24952054023742676,
      "num_input_tokens_seen": 709254888,
      "step": 11830,
      "train_runtime": 152009.4459,
      "train_tokens_per_second": 4665.861
    },
    {
      "epoch": 1.5095663265306123,
      "grad_norm": 0.6258526373067016,
      "learning_rate": 1.412437383060563e-06,
      "loss": 0.2232728958129883,
      "num_input_tokens_seen": 709553400,
      "step": 11835,
      "train_runtime": 152069.4829,
      "train_tokens_per_second": 4665.982
    },
    {
      "epoch": 1.510204081632653,
      "grad_norm": 0.7454052567193138,
      "learning_rate": 1.4089502448092418e-06,
      "loss": 0.24971485137939453,
      "num_input_tokens_seen": 709859344,
      "step": 11840,
      "train_runtime": 152130.0077,
      "train_tokens_per_second": 4666.136
    },
    {
      "epoch": 1.5108418367346939,
      "grad_norm": 0.6347998808876008,
      "learning_rate": 1.4054667104271497e-06,
      "loss": 0.2392899990081787,
      "num_input_tokens_seen": 710163112,
      "step": 11845,
      "train_runtime": 152187.6542,
      "train_tokens_per_second": 4666.365
    },
    {
      "epoch": 1.5114795918367347,
      "grad_norm": 0.6523693559173398,
      "learning_rate": 1.4019867834102525e-06,
      "loss": 0.2507919788360596,
      "num_input_tokens_seen": 710456184,
      "step": 11850,
      "train_runtime": 152253.9669,
      "train_tokens_per_second": 4666.257
    },
    {
      "epoch": 1.5121173469387754,
      "grad_norm": 0.5805169057728986,
      "learning_rate": 1.398510467250906e-06,
      "loss": 0.24285340309143066,
      "num_input_tokens_seen": 710762584,
      "step": 11855,
      "train_runtime": 152309.8344,
      "train_tokens_per_second": 4666.557
    },
    {
      "epoch": 1.5127551020408163,
      "grad_norm": 0.7401071671180456,
      "learning_rate": 1.3950377654378312e-06,
      "loss": 0.26153364181518557,
      "num_input_tokens_seen": 711053624,
      "step": 11860,
      "train_runtime": 152378.9013,
      "train_tokens_per_second": 4666.352
    },
    {
      "epoch": 1.5133928571428572,
      "grad_norm": 0.6592369670622877,
      "learning_rate": 1.3915686814561285e-06,
      "loss": 0.26082634925842285,
      "num_input_tokens_seen": 711348184,
      "step": 11865,
      "train_runtime": 152444.1182,
      "train_tokens_per_second": 4666.288
    },
    {
      "epoch": 1.5140306122448979,
      "grad_norm": 0.6072336499780252,
      "learning_rate": 1.3881032187872666e-06,
      "loss": 0.25055556297302245,
      "num_input_tokens_seen": 711649952,
      "step": 11870,
      "train_runtime": 152510.0835,
      "train_tokens_per_second": 4666.249
    },
    {
      "epoch": 1.5146683673469388,
      "grad_norm": 0.6569088939621313,
      "learning_rate": 1.3846413809090792e-06,
      "loss": 0.25313329696655273,
      "num_input_tokens_seen": 711952072,
      "step": 11875,
      "train_runtime": 152566.0547,
      "train_tokens_per_second": 4666.517
    },
    {
      "epoch": 1.5153061224489797,
      "grad_norm": 0.6614534871650628,
      "learning_rate": 1.381183171295762e-06,
      "loss": 0.24717378616333008,
      "num_input_tokens_seen": 712256824,
      "step": 11880,
      "train_runtime": 152629.3801,
      "train_tokens_per_second": 4666.577
    },
    {
      "epoch": 1.5159438775510203,
      "grad_norm": 0.6168401468502613,
      "learning_rate": 1.3777285934178697e-06,
      "loss": 0.25480525493621825,
      "num_input_tokens_seen": 712560368,
      "step": 11885,
      "train_runtime": 152705.4023,
      "train_tokens_per_second": 4666.242
    },
    {
      "epoch": 1.5165816326530612,
      "grad_norm": 0.6099653585699996,
      "learning_rate": 1.3742776507423127e-06,
      "loss": 0.23254284858703614,
      "num_input_tokens_seen": 712868888,
      "step": 11890,
      "train_runtime": 152765.6495,
      "train_tokens_per_second": 4666.421
    },
    {
      "epoch": 1.5172193877551021,
      "grad_norm": 0.6616464288859243,
      "learning_rate": 1.3708303467323536e-06,
      "loss": 0.24589791297912597,
      "num_input_tokens_seen": 713165632,
      "step": 11895,
      "train_runtime": 152823.9928,
      "train_tokens_per_second": 4666.582
    },
    {
      "epoch": 1.5178571428571428,
      "grad_norm": 0.7249096896463343,
      "learning_rate": 1.367386684847602e-06,
      "loss": 0.24876763820648193,
      "num_input_tokens_seen": 713461784,
      "step": 11900,
      "train_runtime": 152893.0843,
      "train_tokens_per_second": 4666.41
    },
    {
      "epoch": 1.5184948979591837,
      "grad_norm": 0.5962476856616004,
      "learning_rate": 1.3639466685440133e-06,
      "loss": 0.24852046966552735,
      "num_input_tokens_seen": 713773496,
      "step": 11905,
      "train_runtime": 152942.8453,
      "train_tokens_per_second": 4666.93
    },
    {
      "epoch": 1.5191326530612246,
      "grad_norm": 0.7070420796619338,
      "learning_rate": 1.360510301273884e-06,
      "loss": 0.2318204641342163,
      "num_input_tokens_seen": 714071696,
      "step": 11910,
      "train_runtime": 152994.1435,
      "train_tokens_per_second": 4667.314
    },
    {
      "epoch": 1.5197704081632653,
      "grad_norm": 0.6092687720393368,
      "learning_rate": 1.3570775864858487e-06,
      "loss": 0.2326336145401001,
      "num_input_tokens_seen": 714373072,
      "step": 11915,
      "train_runtime": 153063.8437,
      "train_tokens_per_second": 4667.158
    },
    {
      "epoch": 1.5204081632653061,
      "grad_norm": 0.7007567256576211,
      "learning_rate": 1.3536485276248768e-06,
      "loss": 0.24536890983581544,
      "num_input_tokens_seen": 714670808,
      "step": 11920,
      "train_runtime": 153133.9611,
      "train_tokens_per_second": 4666.965
    },
    {
      "epoch": 1.521045918367347,
      "grad_norm": 0.6340967886921898,
      "learning_rate": 1.350223128132268e-06,
      "loss": 0.23796346187591552,
      "num_input_tokens_seen": 714975728,
      "step": 11925,
      "train_runtime": 153188.4068,
      "train_tokens_per_second": 4667.297
    },
    {
      "epoch": 1.5216836734693877,
      "grad_norm": 0.6477653858447132,
      "learning_rate": 1.3468013914456518e-06,
      "loss": 0.24376912117004396,
      "num_input_tokens_seen": 715272848,
      "step": 11930,
      "train_runtime": 153244.3244,
      "train_tokens_per_second": 4667.532
    },
    {
      "epoch": 1.5223214285714286,
      "grad_norm": 0.6234235892429398,
      "learning_rate": 1.3433833209989743e-06,
      "loss": 0.2379302740097046,
      "num_input_tokens_seen": 715553496,
      "step": 11935,
      "train_runtime": 153308.5292,
      "train_tokens_per_second": 4667.408
    },
    {
      "epoch": 1.5229591836734695,
      "grad_norm": 0.6581030845799587,
      "learning_rate": 1.3399689202225136e-06,
      "loss": 0.2301718473434448,
      "num_input_tokens_seen": 715847752,
      "step": 11940,
      "train_runtime": 153372.0259,
      "train_tokens_per_second": 4667.395
    },
    {
      "epoch": 1.5235969387755102,
      "grad_norm": 0.7410182128633257,
      "learning_rate": 1.3365581925428594e-06,
      "loss": 0.23722467422485352,
      "num_input_tokens_seen": 716137152,
      "step": 11945,
      "train_runtime": 153434.276,
      "train_tokens_per_second": 4667.387
    },
    {
      "epoch": 1.524234693877551,
      "grad_norm": 0.6886070210738373,
      "learning_rate": 1.3331511413829113e-06,
      "loss": 0.2482837200164795,
      "num_input_tokens_seen": 716439560,
      "step": 11950,
      "train_runtime": 153502.5991,
      "train_tokens_per_second": 4667.28
    },
    {
      "epoch": 1.524872448979592,
      "grad_norm": 0.5673444957019642,
      "learning_rate": 1.3297477701618833e-06,
      "loss": 0.21979570388793945,
      "num_input_tokens_seen": 716747032,
      "step": 11955,
      "train_runtime": 153562.373,
      "train_tokens_per_second": 4667.465
    },
    {
      "epoch": 1.5255102040816326,
      "grad_norm": 0.675068128154071,
      "learning_rate": 1.3263480822953011e-06,
      "loss": 0.2448493003845215,
      "num_input_tokens_seen": 717052576,
      "step": 11960,
      "train_runtime": 153613.8129,
      "train_tokens_per_second": 4667.891
    },
    {
      "epoch": 1.5261479591836735,
      "grad_norm": 0.6987596997751906,
      "learning_rate": 1.3229520811949836e-06,
      "loss": 0.24974460601806642,
      "num_input_tokens_seen": 717354344,
      "step": 11965,
      "train_runtime": 153677.3411,
      "train_tokens_per_second": 4667.925
    },
    {
      "epoch": 1.5267857142857144,
      "grad_norm": 0.7484144778958687,
      "learning_rate": 1.3195597702690577e-06,
      "loss": 0.24799327850341796,
      "num_input_tokens_seen": 717648840,
      "step": 11970,
      "train_runtime": 153752.3726,
      "train_tokens_per_second": 4667.563
    },
    {
      "epoch": 1.527423469387755,
      "grad_norm": 0.6071064267657477,
      "learning_rate": 1.3161711529219434e-06,
      "loss": 0.23238167762756348,
      "num_input_tokens_seen": 717949184,
      "step": 11975,
      "train_runtime": 153821.8973,
      "train_tokens_per_second": 4667.406
    },
    {
      "epoch": 1.5280612244897958,
      "grad_norm": 0.6128589850855337,
      "learning_rate": 1.3127862325543555e-06,
      "loss": 0.23907299041748048,
      "num_input_tokens_seen": 718267512,
      "step": 11980,
      "train_runtime": 153892.8073,
      "train_tokens_per_second": 4667.323
    },
    {
      "epoch": 1.5286989795918369,
      "grad_norm": 0.622447941338741,
      "learning_rate": 1.3094050125632973e-06,
      "loss": 0.23790667057037354,
      "num_input_tokens_seen": 718562104,
      "step": 11985,
      "train_runtime": 153950.5398,
      "train_tokens_per_second": 4667.487
    },
    {
      "epoch": 1.5293367346938775,
      "grad_norm": 0.6253634222220107,
      "learning_rate": 1.30602749634206e-06,
      "loss": 0.24044299125671387,
      "num_input_tokens_seen": 718840408,
      "step": 11990,
      "train_runtime": 154015.8282,
      "train_tokens_per_second": 4667.315
    },
    {
      "epoch": 1.5299744897959182,
      "grad_norm": 0.6783952443597888,
      "learning_rate": 1.302653687280216e-06,
      "loss": 0.2323903799057007,
      "num_input_tokens_seen": 719130280,
      "step": 11995,
      "train_runtime": 154084.7779,
      "train_tokens_per_second": 4667.108
    },
    {
      "epoch": 1.5306122448979593,
      "grad_norm": 0.6945063615258878,
      "learning_rate": 1.299283588763619e-06,
      "loss": 0.24817018508911132,
      "num_input_tokens_seen": 719432752,
      "step": 12000,
      "train_runtime": 154135.898,
      "train_tokens_per_second": 4667.522
    },
    {
      "epoch": 1.53125,
      "grad_norm": 0.6329196402660104,
      "learning_rate": 1.2959172041743985e-06,
      "loss": 0.2507436990737915,
      "num_input_tokens_seen": 719749568,
      "step": 12005,
      "train_runtime": 154197.5307,
      "train_tokens_per_second": 4667.711
    },
    {
      "epoch": 1.5318877551020407,
      "grad_norm": 0.621708965357833,
      "learning_rate": 1.2925545368909558e-06,
      "loss": 0.2255568504333496,
      "num_input_tokens_seen": 720050496,
      "step": 12010,
      "train_runtime": 154250.7911,
      "train_tokens_per_second": 4668.051
    },
    {
      "epoch": 1.5325255102040818,
      "grad_norm": 0.75523776564148,
      "learning_rate": 1.289195590287965e-06,
      "loss": 0.24075818061828613,
      "num_input_tokens_seen": 720343552,
      "step": 12015,
      "train_runtime": 154308.762,
      "train_tokens_per_second": 4668.196
    },
    {
      "epoch": 1.5331632653061225,
      "grad_norm": 0.6733882703579561,
      "learning_rate": 1.285840367736358e-06,
      "loss": 0.24710659980773925,
      "num_input_tokens_seen": 720650448,
      "step": 12020,
      "train_runtime": 154379.0719,
      "train_tokens_per_second": 4668.058
    },
    {
      "epoch": 1.5338010204081631,
      "grad_norm": 0.6910433010762052,
      "learning_rate": 1.2824888726033391e-06,
      "loss": 0.243064022064209,
      "num_input_tokens_seen": 720955344,
      "step": 12025,
      "train_runtime": 154430.8445,
      "train_tokens_per_second": 4668.467
    },
    {
      "epoch": 1.5344387755102042,
      "grad_norm": 0.6246426595842839,
      "learning_rate": 1.2791411082523692e-06,
      "loss": 0.23680500984191893,
      "num_input_tokens_seen": 721246104,
      "step": 12030,
      "train_runtime": 154495.0162,
      "train_tokens_per_second": 4668.41
    },
    {
      "epoch": 1.535076530612245,
      "grad_norm": 0.7113607226927038,
      "learning_rate": 1.2757970780431594e-06,
      "loss": 0.243087100982666,
      "num_input_tokens_seen": 721546096,
      "step": 12035,
      "train_runtime": 154562.9343,
      "train_tokens_per_second": 4668.3
    },
    {
      "epoch": 1.5357142857142856,
      "grad_norm": 0.7032655313783374,
      "learning_rate": 1.2724567853316788e-06,
      "loss": 0.22990317344665528,
      "num_input_tokens_seen": 721837336,
      "step": 12040,
      "train_runtime": 154631.697,
      "train_tokens_per_second": 4668.107
    },
    {
      "epoch": 1.5363520408163265,
      "grad_norm": 0.6344855608090577,
      "learning_rate": 1.2691202334701474e-06,
      "loss": 0.25099179744720457,
      "num_input_tokens_seen": 722138240,
      "step": 12045,
      "train_runtime": 154702.7814,
      "train_tokens_per_second": 4667.907
    },
    {
      "epoch": 1.5369897959183674,
      "grad_norm": 0.583662349867842,
      "learning_rate": 1.2657874258070247e-06,
      "loss": 0.23311772346496581,
      "num_input_tokens_seen": 722452088,
      "step": 12050,
      "train_runtime": 154754.0773,
      "train_tokens_per_second": 4668.388
    },
    {
      "epoch": 1.537627551020408,
      "grad_norm": 0.7045862347703993,
      "learning_rate": 1.2624583656870153e-06,
      "loss": 0.24171760082244872,
      "num_input_tokens_seen": 722758192,
      "step": 12055,
      "train_runtime": 154820.0006,
      "train_tokens_per_second": 4668.377
    },
    {
      "epoch": 1.538265306122449,
      "grad_norm": 0.6739785249759156,
      "learning_rate": 1.2591330564510683e-06,
      "loss": 0.22491805553436278,
      "num_input_tokens_seen": 723054440,
      "step": 12060,
      "train_runtime": 154882.878,
      "train_tokens_per_second": 4668.395
    },
    {
      "epoch": 1.5389030612244898,
      "grad_norm": 0.7843959663478336,
      "learning_rate": 1.2558115014363592e-06,
      "loss": 0.2372133255004883,
      "num_input_tokens_seen": 723359152,
      "step": 12065,
      "train_runtime": 154943.582,
      "train_tokens_per_second": 4668.533
    },
    {
      "epoch": 1.5395408163265305,
      "grad_norm": 0.8164871100890743,
      "learning_rate": 1.2524937039763025e-06,
      "loss": 0.2527186870574951,
      "num_input_tokens_seen": 723654448,
      "step": 12070,
      "train_runtime": 155002.0966,
      "train_tokens_per_second": 4668.675
    },
    {
      "epoch": 1.5401785714285714,
      "grad_norm": 0.7210224573481199,
      "learning_rate": 1.2491796674005385e-06,
      "loss": 0.25359201431274414,
      "num_input_tokens_seen": 723963864,
      "step": 12075,
      "train_runtime": 155073.6699,
      "train_tokens_per_second": 4668.516
    },
    {
      "epoch": 1.5408163265306123,
      "grad_norm": 0.8022499591976578,
      "learning_rate": 1.2458693950349349e-06,
      "loss": 0.246889328956604,
      "num_input_tokens_seen": 724255576,
      "step": 12080,
      "train_runtime": 155143.0431,
      "train_tokens_per_second": 4668.308
    },
    {
      "epoch": 1.541454081632653,
      "grad_norm": 0.6603952877473191,
      "learning_rate": 1.2425628902015818e-06,
      "loss": 0.24393868446350098,
      "num_input_tokens_seen": 724549344,
      "step": 12085,
      "train_runtime": 155211.7265,
      "train_tokens_per_second": 4668.135
    },
    {
      "epoch": 1.5420918367346939,
      "grad_norm": 0.680804934740307,
      "learning_rate": 1.2392601562187867e-06,
      "loss": 0.24800515174865723,
      "num_input_tokens_seen": 724854896,
      "step": 12090,
      "train_runtime": 155282.8795,
      "train_tokens_per_second": 4667.964
    },
    {
      "epoch": 1.5427295918367347,
      "grad_norm": 0.6345715341836401,
      "learning_rate": 1.2359611964010747e-06,
      "loss": 0.2576450347900391,
      "num_input_tokens_seen": 725156064,
      "step": 12095,
      "train_runtime": 155345.7552,
      "train_tokens_per_second": 4668.013
    },
    {
      "epoch": 1.5433673469387754,
      "grad_norm": 0.6113492139638761,
      "learning_rate": 1.2326660140591835e-06,
      "loss": 0.23038859367370607,
      "num_input_tokens_seen": 725461232,
      "step": 12100,
      "train_runtime": 155419.126,
      "train_tokens_per_second": 4667.773
    },
    {
      "epoch": 1.5440051020408163,
      "grad_norm": 0.6236819131279752,
      "learning_rate": 1.2293746125000538e-06,
      "loss": 0.2277146816253662,
      "num_input_tokens_seen": 725754896,
      "step": 12105,
      "train_runtime": 155490.0221,
      "train_tokens_per_second": 4667.534
    },
    {
      "epoch": 1.5446428571428572,
      "grad_norm": 0.6670732652943816,
      "learning_rate": 1.2260869950268422e-06,
      "loss": 0.24714210033416747,
      "num_input_tokens_seen": 726057584,
      "step": 12110,
      "train_runtime": 155562.6185,
      "train_tokens_per_second": 4667.301
    },
    {
      "epoch": 1.5452806122448979,
      "grad_norm": 0.7632437554992797,
      "learning_rate": 1.2228031649389021e-06,
      "loss": 0.236268949508667,
      "num_input_tokens_seen": 726335360,
      "step": 12115,
      "train_runtime": 155619.6084,
      "train_tokens_per_second": 4667.377
    },
    {
      "epoch": 1.5459183673469388,
      "grad_norm": 0.7236167943503262,
      "learning_rate": 1.2195231255317834e-06,
      "loss": 0.24866657257080077,
      "num_input_tokens_seen": 726626968,
      "step": 12120,
      "train_runtime": 155685.8264,
      "train_tokens_per_second": 4667.265
    },
    {
      "epoch": 1.5465561224489797,
      "grad_norm": 0.8505834324878082,
      "learning_rate": 1.2162468800972344e-06,
      "loss": 0.23465731143951415,
      "num_input_tokens_seen": 726907752,
      "step": 12125,
      "train_runtime": 155747.1417,
      "train_tokens_per_second": 4667.23
    },
    {
      "epoch": 1.5471938775510203,
      "grad_norm": 0.6702233970780133,
      "learning_rate": 1.2129744319232013e-06,
      "loss": 0.22164678573608398,
      "num_input_tokens_seen": 727196712,
      "step": 12130,
      "train_runtime": 155806.6165,
      "train_tokens_per_second": 4667.303
    },
    {
      "epoch": 1.5478316326530612,
      "grad_norm": 0.6471140669180683,
      "learning_rate": 1.2097057842938093e-06,
      "loss": 0.23233861923217775,
      "num_input_tokens_seen": 727482832,
      "step": 12135,
      "train_runtime": 155875.9898,
      "train_tokens_per_second": 4667.062
    },
    {
      "epoch": 1.5484693877551021,
      "grad_norm": 0.6157611642938653,
      "learning_rate": 1.206440940489375e-06,
      "loss": 0.23581852912902831,
      "num_input_tokens_seen": 727782296,
      "step": 12140,
      "train_runtime": 155935.2833,
      "train_tokens_per_second": 4667.207
    },
    {
      "epoch": 1.5491071428571428,
      "grad_norm": 0.5963500844337734,
      "learning_rate": 1.203179903786401e-06,
      "loss": 0.22914860248565674,
      "num_input_tokens_seen": 728088904,
      "step": 12145,
      "train_runtime": 156004.1634,
      "train_tokens_per_second": 4667.112
    },
    {
      "epoch": 1.5497448979591837,
      "grad_norm": 0.7361686577537355,
      "learning_rate": 1.1999226774575618e-06,
      "loss": 0.21748099327087403,
      "num_input_tokens_seen": 728382016,
      "step": 12150,
      "train_runtime": 156068.6042,
      "train_tokens_per_second": 4667.063
    },
    {
      "epoch": 1.5503826530612246,
      "grad_norm": 0.6996395183231016,
      "learning_rate": 1.1966692647717127e-06,
      "loss": 0.24199068546295166,
      "num_input_tokens_seen": 728678440,
      "step": 12155,
      "train_runtime": 156132.4753,
      "train_tokens_per_second": 4667.052
    },
    {
      "epoch": 1.5510204081632653,
      "grad_norm": 0.6397591614429556,
      "learning_rate": 1.1934196689938804e-06,
      "loss": 0.23775906562805177,
      "num_input_tokens_seen": 728986480,
      "step": 12160,
      "train_runtime": 156199.2921,
      "train_tokens_per_second": 4667.028
    },
    {
      "epoch": 1.5516581632653061,
      "grad_norm": 0.7352359065152636,
      "learning_rate": 1.1901738933852608e-06,
      "loss": 0.2632540464401245,
      "num_input_tokens_seen": 729287568,
      "step": 12165,
      "train_runtime": 156256.6385,
      "train_tokens_per_second": 4667.242
    },
    {
      "epoch": 1.552295918367347,
      "grad_norm": 0.6203159734089616,
      "learning_rate": 1.1869319412032171e-06,
      "loss": 0.24134318828582763,
      "num_input_tokens_seen": 729580632,
      "step": 12170,
      "train_runtime": 156327.0294,
      "train_tokens_per_second": 4667.015
    },
    {
      "epoch": 1.5529336734693877,
      "grad_norm": 0.6144417798093549,
      "learning_rate": 1.1836938157012746e-06,
      "loss": 0.24182264804840087,
      "num_input_tokens_seen": 729885248,
      "step": 12175,
      "train_runtime": 156384.9372,
      "train_tokens_per_second": 4667.235
    },
    {
      "epoch": 1.5535714285714286,
      "grad_norm": 0.6286938807568226,
      "learning_rate": 1.180459520129118e-06,
      "loss": 0.22440452575683595,
      "num_input_tokens_seen": 730181648,
      "step": 12180,
      "train_runtime": 156447.3014,
      "train_tokens_per_second": 4667.269
    },
    {
      "epoch": 1.5542091836734695,
      "grad_norm": 0.6422738199085359,
      "learning_rate": 1.1772290577325895e-06,
      "loss": 0.24348549842834472,
      "num_input_tokens_seen": 730503360,
      "step": 12185,
      "train_runtime": 156502.6241,
      "train_tokens_per_second": 4667.675
    },
    {
      "epoch": 1.5548469387755102,
      "grad_norm": 0.6515386793597964,
      "learning_rate": 1.174002431753683e-06,
      "loss": 0.2396906852722168,
      "num_input_tokens_seen": 730801680,
      "step": 12190,
      "train_runtime": 156574.663,
      "train_tokens_per_second": 4667.433
    },
    {
      "epoch": 1.555484693877551,
      "grad_norm": 0.617281142350979,
      "learning_rate": 1.1707796454305437e-06,
      "loss": 0.21960153579711914,
      "num_input_tokens_seen": 731088968,
      "step": 12195,
      "train_runtime": 156637.9622,
      "train_tokens_per_second": 4667.38
    },
    {
      "epoch": 1.556122448979592,
      "grad_norm": 0.7140282480812199,
      "learning_rate": 1.167560701997462e-06,
      "loss": 0.24591443538665772,
      "num_input_tokens_seen": 731381120,
      "step": 12200,
      "train_runtime": 156700.3172,
      "train_tokens_per_second": 4667.388
    },
    {
      "epoch": 1.5567602040816326,
      "grad_norm": 0.7404723800350034,
      "learning_rate": 1.1643456046848745e-06,
      "loss": 0.24412984848022462,
      "num_input_tokens_seen": 731677024,
      "step": 12205,
      "train_runtime": 156757.7499,
      "train_tokens_per_second": 4667.565
    },
    {
      "epoch": 1.5573979591836735,
      "grad_norm": 0.6294267377024202,
      "learning_rate": 1.161134356719351e-06,
      "loss": 0.2619341850280762,
      "num_input_tokens_seen": 731960800,
      "step": 12210,
      "train_runtime": 156826.9864,
      "train_tokens_per_second": 4667.314
    },
    {
      "epoch": 1.5580357142857144,
      "grad_norm": 0.5994403997217085,
      "learning_rate": 1.1579269613236094e-06,
      "loss": 0.26063690185546873,
      "num_input_tokens_seen": 732276544,
      "step": 12215,
      "train_runtime": 156886.9605,
      "train_tokens_per_second": 4667.542
    },
    {
      "epoch": 1.558673469387755,
      "grad_norm": 0.6014416982113758,
      "learning_rate": 1.1547234217164904e-06,
      "loss": 0.23040192127227782,
      "num_input_tokens_seen": 732590184,
      "step": 12220,
      "train_runtime": 156964.2101,
      "train_tokens_per_second": 4667.243
    },
    {
      "epoch": 1.5593112244897958,
      "grad_norm": 0.6018550376268367,
      "learning_rate": 1.1515237411129698e-06,
      "loss": 0.22739155292510987,
      "num_input_tokens_seen": 732902512,
      "step": 12225,
      "train_runtime": 157031.2963,
      "train_tokens_per_second": 4667.239
    },
    {
      "epoch": 1.5599489795918369,
      "grad_norm": 0.6531853561303372,
      "learning_rate": 1.148327922724155e-06,
      "loss": 0.25479941368103026,
      "num_input_tokens_seen": 733196336,
      "step": 12230,
      "train_runtime": 157090.1437,
      "train_tokens_per_second": 4667.361
    },
    {
      "epoch": 1.5605867346938775,
      "grad_norm": 0.6792004519096303,
      "learning_rate": 1.1451359697572684e-06,
      "loss": 0.24005823135375975,
      "num_input_tokens_seen": 733489424,
      "step": 12235,
      "train_runtime": 157141.875,
      "train_tokens_per_second": 4667.689
    },
    {
      "epoch": 1.5612244897959182,
      "grad_norm": 0.6756138286546107,
      "learning_rate": 1.141947885415659e-06,
      "loss": 0.22763824462890625,
      "num_input_tokens_seen": 733796688,
      "step": 12240,
      "train_runtime": 157197.1705,
      "train_tokens_per_second": 4668.002
    },
    {
      "epoch": 1.5618622448979593,
      "grad_norm": 0.7494574512586077,
      "learning_rate": 1.1387636728987927e-06,
      "loss": 0.2540790796279907,
      "num_input_tokens_seen": 734080432,
      "step": 12245,
      "train_runtime": 157262.8572,
      "train_tokens_per_second": 4667.856
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.6972769493447613,
      "learning_rate": 1.135583335402249e-06,
      "loss": 0.2700526714324951,
      "num_input_tokens_seen": 734376320,
      "step": 12250,
      "train_runtime": 157331.3394,
      "train_tokens_per_second": 4667.705
    },
    {
      "epoch": 1.5631377551020407,
      "grad_norm": 0.6890260436067339,
      "learning_rate": 1.1324068761177176e-06,
      "loss": 0.25571448802948,
      "num_input_tokens_seen": 734673928,
      "step": 12255,
      "train_runtime": 157403.2433,
      "train_tokens_per_second": 4667.464
    },
    {
      "epoch": 1.5637755102040818,
      "grad_norm": 0.5984765002329604,
      "learning_rate": 1.129234298232999e-06,
      "loss": 0.22999925613403321,
      "num_input_tokens_seen": 734977064,
      "step": 12260,
      "train_runtime": 157474.077,
      "train_tokens_per_second": 4667.289
    },
    {
      "epoch": 1.5644132653061225,
      "grad_norm": 0.6601949929536787,
      "learning_rate": 1.1260656049319957e-06,
      "loss": 0.2365842342376709,
      "num_input_tokens_seen": 735279168,
      "step": 12265,
      "train_runtime": 157548.6826,
      "train_tokens_per_second": 4666.997
    },
    {
      "epoch": 1.5650510204081631,
      "grad_norm": 0.6805812610372765,
      "learning_rate": 1.1229007993947122e-06,
      "loss": 0.23756906986236573,
      "num_input_tokens_seen": 735583336,
      "step": 12270,
      "train_runtime": 157613.418,
      "train_tokens_per_second": 4667.01
    },
    {
      "epoch": 1.5656887755102042,
      "grad_norm": 0.6290302208737203,
      "learning_rate": 1.119739884797253e-06,
      "loss": 0.24585089683532715,
      "num_input_tokens_seen": 735896568,
      "step": 12275,
      "train_runtime": 157671.3166,
      "train_tokens_per_second": 4667.282
    },
    {
      "epoch": 1.566326530612245,
      "grad_norm": 0.6463052759249044,
      "learning_rate": 1.116582864311816e-06,
      "loss": 0.22634570598602294,
      "num_input_tokens_seen": 736199848,
      "step": 12280,
      "train_runtime": 157742.2436,
      "train_tokens_per_second": 4667.106
    },
    {
      "epoch": 1.5669642857142856,
      "grad_norm": 0.5805619461811257,
      "learning_rate": 1.1134297411066925e-06,
      "loss": 0.2247623920440674,
      "num_input_tokens_seen": 736505336,
      "step": 12285,
      "train_runtime": 157799.8095,
      "train_tokens_per_second": 4667.34
    },
    {
      "epoch": 1.5676020408163265,
      "grad_norm": 0.7141466447797546,
      "learning_rate": 1.1102805183462605e-06,
      "loss": 0.25339384078979493,
      "num_input_tokens_seen": 736797016,
      "step": 12290,
      "train_runtime": 157867.3211,
      "train_tokens_per_second": 4667.191
    },
    {
      "epoch": 1.5682397959183674,
      "grad_norm": 0.5839728980221631,
      "learning_rate": 1.1071351991909863e-06,
      "loss": 0.2568344593048096,
      "num_input_tokens_seen": 737111864,
      "step": 12295,
      "train_runtime": 157937.2893,
      "train_tokens_per_second": 4667.117
    },
    {
      "epoch": 1.568877551020408,
      "grad_norm": 0.6939502207995728,
      "learning_rate": 1.1039937867974166e-06,
      "loss": 0.2604703426361084,
      "num_input_tokens_seen": 737416048,
      "step": 12300,
      "train_runtime": 157991.5562,
      "train_tokens_per_second": 4667.44
    },
    {
      "epoch": 1.569515306122449,
      "grad_norm": 0.6999492214773526,
      "learning_rate": 1.1008562843181796e-06,
      "loss": 0.24059996604919434,
      "num_input_tokens_seen": 737723824,
      "step": 12305,
      "train_runtime": 158046.4973,
      "train_tokens_per_second": 4667.764
    },
    {
      "epoch": 1.5701530612244898,
      "grad_norm": 0.7321332951134449,
      "learning_rate": 1.0977226949019743e-06,
      "loss": 0.2498480796813965,
      "num_input_tokens_seen": 738027632,
      "step": 12310,
      "train_runtime": 158112.2063,
      "train_tokens_per_second": 4667.746
    },
    {
      "epoch": 1.5707908163265305,
      "grad_norm": 0.6633475427756443,
      "learning_rate": 1.09459302169358e-06,
      "loss": 0.2533766746520996,
      "num_input_tokens_seen": 738333472,
      "step": 12315,
      "train_runtime": 158180.3748,
      "train_tokens_per_second": 4667.668
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.6280044273693439,
      "learning_rate": 1.0914672678338434e-06,
      "loss": 0.23186039924621582,
      "num_input_tokens_seen": 738639544,
      "step": 12320,
      "train_runtime": 158248.8884,
      "train_tokens_per_second": 4667.581
    },
    {
      "epoch": 1.5720663265306123,
      "grad_norm": 0.6535015556023339,
      "learning_rate": 1.088345436459673e-06,
      "loss": 0.2551730155944824,
      "num_input_tokens_seen": 738947504,
      "step": 12325,
      "train_runtime": 158312.341,
      "train_tokens_per_second": 4667.656
    },
    {
      "epoch": 1.572704081632653,
      "grad_norm": 0.6392517071683543,
      "learning_rate": 1.0852275307040454e-06,
      "loss": 0.25040478706359864,
      "num_input_tokens_seen": 739256992,
      "step": 12330,
      "train_runtime": 158380.0934,
      "train_tokens_per_second": 4667.613
    },
    {
      "epoch": 1.5733418367346939,
      "grad_norm": 0.6530792561870343,
      "learning_rate": 1.082113553696001e-06,
      "loss": 0.25493450164794923,
      "num_input_tokens_seen": 739559392,
      "step": 12335,
      "train_runtime": 158444.7723,
      "train_tokens_per_second": 4667.616
    },
    {
      "epoch": 1.5739795918367347,
      "grad_norm": 0.6030977562922666,
      "learning_rate": 1.0790035085606282e-06,
      "loss": 0.2312842607498169,
      "num_input_tokens_seen": 739846688,
      "step": 12340,
      "train_runtime": 158497.1012,
      "train_tokens_per_second": 4667.888
    },
    {
      "epoch": 1.5746173469387754,
      "grad_norm": 0.6480304860769777,
      "learning_rate": 1.0758973984190764e-06,
      "loss": 0.24834418296813965,
      "num_input_tokens_seen": 740146568,
      "step": 12345,
      "train_runtime": 158570.4446,
      "train_tokens_per_second": 4667.62
    },
    {
      "epoch": 1.5752551020408163,
      "grad_norm": 0.6416794020452617,
      "learning_rate": 1.072795226388544e-06,
      "loss": 0.24527735710144044,
      "num_input_tokens_seen": 740451296,
      "step": 12350,
      "train_runtime": 158637.3538,
      "train_tokens_per_second": 4667.572
    },
    {
      "epoch": 1.5758928571428572,
      "grad_norm": 0.6841422613359235,
      "learning_rate": 1.0696969955822772e-06,
      "loss": 0.2373568296432495,
      "num_input_tokens_seen": 740740000,
      "step": 12355,
      "train_runtime": 158698.8702,
      "train_tokens_per_second": 4667.582
    },
    {
      "epoch": 1.5765306122448979,
      "grad_norm": 0.6797464373561524,
      "learning_rate": 1.0666027091095672e-06,
      "loss": 0.24493880271911622,
      "num_input_tokens_seen": 741042248,
      "step": 12360,
      "train_runtime": 158764.9238,
      "train_tokens_per_second": 4667.544
    },
    {
      "epoch": 1.5771683673469388,
      "grad_norm": 0.6444618107327897,
      "learning_rate": 1.063512370075746e-06,
      "loss": 0.24353444576263428,
      "num_input_tokens_seen": 741337520,
      "step": 12365,
      "train_runtime": 158835.3912,
      "train_tokens_per_second": 4667.332
    },
    {
      "epoch": 1.5778061224489797,
      "grad_norm": 0.6527816475387099,
      "learning_rate": 1.060425981582186e-06,
      "loss": 0.22241063117980958,
      "num_input_tokens_seen": 741637616,
      "step": 12370,
      "train_runtime": 158896.1148,
      "train_tokens_per_second": 4667.437
    },
    {
      "epoch": 1.5784438775510203,
      "grad_norm": 0.6355874400538328,
      "learning_rate": 1.0573435467262922e-06,
      "loss": 0.24539518356323242,
      "num_input_tokens_seen": 741932880,
      "step": 12375,
      "train_runtime": 158971.348,
      "train_tokens_per_second": 4667.086
    },
    {
      "epoch": 1.5790816326530612,
      "grad_norm": 0.5993159408239529,
      "learning_rate": 1.0542650686015039e-06,
      "loss": 0.21699769496917726,
      "num_input_tokens_seen": 742243584,
      "step": 12380,
      "train_runtime": 159033.4402,
      "train_tokens_per_second": 4667.217
    },
    {
      "epoch": 1.5797193877551021,
      "grad_norm": 0.7044253186693795,
      "learning_rate": 1.0511905502972885e-06,
      "loss": 0.24109046459197997,
      "num_input_tokens_seen": 742534568,
      "step": 12385,
      "train_runtime": 159105.6225,
      "train_tokens_per_second": 4666.929
    },
    {
      "epoch": 1.5803571428571428,
      "grad_norm": 0.6439837302875744,
      "learning_rate": 1.0481199948991416e-06,
      "loss": 0.2449514389038086,
      "num_input_tokens_seen": 742848256,
      "step": 12390,
      "train_runtime": 159172.1846,
      "train_tokens_per_second": 4666.948
    },
    {
      "epoch": 1.5809948979591837,
      "grad_norm": 0.6974978330015498,
      "learning_rate": 1.0450534054885765e-06,
      "loss": 0.21677944660186768,
      "num_input_tokens_seen": 743148272,
      "step": 12395,
      "train_runtime": 159234.9335,
      "train_tokens_per_second": 4666.993
    },
    {
      "epoch": 1.5816326530612246,
      "grad_norm": 0.6875081242778867,
      "learning_rate": 1.0419907851431321e-06,
      "loss": 0.2608932971954346,
      "num_input_tokens_seen": 743452104,
      "step": 12400,
      "train_runtime": 159299.8493,
      "train_tokens_per_second": 4666.998
    },
    {
      "epoch": 1.5822704081632653,
      "grad_norm": 0.6433749779925141,
      "learning_rate": 1.0389321369363636e-06,
      "loss": 0.23439176082611085,
      "num_input_tokens_seen": 743764856,
      "step": 12405,
      "train_runtime": 159363.5741,
      "train_tokens_per_second": 4667.094
    },
    {
      "epoch": 1.5829081632653061,
      "grad_norm": 0.6439855842632521,
      "learning_rate": 1.0358774639378344e-06,
      "loss": 0.23441028594970703,
      "num_input_tokens_seen": 744061392,
      "step": 12410,
      "train_runtime": 159427.616,
      "train_tokens_per_second": 4667.08
    },
    {
      "epoch": 1.583545918367347,
      "grad_norm": 0.6487079206891848,
      "learning_rate": 1.0328267692131227e-06,
      "loss": 0.23293354511260986,
      "num_input_tokens_seen": 744372808,
      "step": 12415,
      "train_runtime": 159484.9136,
      "train_tokens_per_second": 4667.356
    },
    {
      "epoch": 1.5841836734693877,
      "grad_norm": 0.6146677300144434,
      "learning_rate": 1.0297800558238175e-06,
      "loss": 0.21918458938598634,
      "num_input_tokens_seen": 744670928,
      "step": 12420,
      "train_runtime": 159546.6699,
      "train_tokens_per_second": 4667.418
    },
    {
      "epoch": 1.5848214285714286,
      "grad_norm": 0.6916442609320224,
      "learning_rate": 1.026737326827505e-06,
      "loss": 0.25938949584960935,
      "num_input_tokens_seen": 744986504,
      "step": 12425,
      "train_runtime": 159616.3739,
      "train_tokens_per_second": 4667.356
    },
    {
      "epoch": 1.5854591836734695,
      "grad_norm": 0.6367232002134973,
      "learning_rate": 1.0236985852777748e-06,
      "loss": 0.23069686889648439,
      "num_input_tokens_seen": 745272504,
      "step": 12430,
      "train_runtime": 159685.0408,
      "train_tokens_per_second": 4667.14
    },
    {
      "epoch": 1.5860969387755102,
      "grad_norm": 0.6880768873752653,
      "learning_rate": 1.020663834224222e-06,
      "loss": 0.23309216499328614,
      "num_input_tokens_seen": 745568368,
      "step": 12435,
      "train_runtime": 159757.1664,
      "train_tokens_per_second": 4666.885
    },
    {
      "epoch": 1.586734693877551,
      "grad_norm": 0.6494159907274198,
      "learning_rate": 1.017633076712426e-06,
      "loss": 0.2540991544723511,
      "num_input_tokens_seen": 745869248,
      "step": 12440,
      "train_runtime": 159826.122,
      "train_tokens_per_second": 4666.754
    },
    {
      "epoch": 1.587372448979592,
      "grad_norm": 0.6629497075583926,
      "learning_rate": 1.0146063157839653e-06,
      "loss": 0.22958106994628907,
      "num_input_tokens_seen": 746172312,
      "step": 12445,
      "train_runtime": 159885.4601,
      "train_tokens_per_second": 4666.918
    },
    {
      "epoch": 1.5880102040816326,
      "grad_norm": 0.7702819194299826,
      "learning_rate": 1.0115835544764063e-06,
      "loss": 0.2517064571380615,
      "num_input_tokens_seen": 746470200,
      "step": 12450,
      "train_runtime": 159951.4637,
      "train_tokens_per_second": 4666.854
    },
    {
      "epoch": 1.5886479591836735,
      "grad_norm": 0.694287201386589,
      "learning_rate": 1.0085647958233008e-06,
      "loss": 0.24434814453125,
      "num_input_tokens_seen": 746764984,
      "step": 12455,
      "train_runtime": 160004.2979,
      "train_tokens_per_second": 4667.156
    },
    {
      "epoch": 1.5892857142857144,
      "grad_norm": 0.613071630625283,
      "learning_rate": 1.0055500428541836e-06,
      "loss": 0.22049248218536377,
      "num_input_tokens_seen": 747064784,
      "step": 12460,
      "train_runtime": 160065.5375,
      "train_tokens_per_second": 4667.243
    },
    {
      "epoch": 1.589923469387755,
      "grad_norm": 0.5905878829379532,
      "learning_rate": 1.0025392985945703e-06,
      "loss": 0.2363584518432617,
      "num_input_tokens_seen": 747372304,
      "step": 12465,
      "train_runtime": 160120.181,
      "train_tokens_per_second": 4667.571
    },
    {
      "epoch": 1.5905612244897958,
      "grad_norm": 0.6543739192723753,
      "learning_rate": 9.995325660659533e-07,
      "loss": 0.23653178215026854,
      "num_input_tokens_seen": 747660040,
      "step": 12470,
      "train_runtime": 160181.4819,
      "train_tokens_per_second": 4667.581
    },
    {
      "epoch": 1.5911989795918369,
      "grad_norm": 0.580553517738533,
      "learning_rate": 9.965298482857993e-07,
      "loss": 0.24920930862426757,
      "num_input_tokens_seen": 747965592,
      "step": 12475,
      "train_runtime": 160249.2731,
      "train_tokens_per_second": 4667.513
    },
    {
      "epoch": 1.5918367346938775,
      "grad_norm": 0.7006306259163959,
      "learning_rate": 9.935311482675413e-07,
      "loss": 0.23427369594573974,
      "num_input_tokens_seen": 748239896,
      "step": 12480,
      "train_runtime": 160307.4589,
      "train_tokens_per_second": 4667.53
    },
    {
      "epoch": 1.5924744897959182,
      "grad_norm": 0.6803649036060584,
      "learning_rate": 9.905364690205888e-07,
      "loss": 0.23743538856506347,
      "num_input_tokens_seen": 748525640,
      "step": 12485,
      "train_runtime": 160372.5366,
      "train_tokens_per_second": 4667.418
    },
    {
      "epoch": 1.5931122448979593,
      "grad_norm": 0.6021010522647829,
      "learning_rate": 9.875458135503113e-07,
      "loss": 0.21800599098205567,
      "num_input_tokens_seen": 748837360,
      "step": 12490,
      "train_runtime": 160433.398,
      "train_tokens_per_second": 4667.59
    },
    {
      "epoch": 1.59375,
      "grad_norm": 0.6465303084231173,
      "learning_rate": 9.845591848580376e-07,
      "loss": 0.24465422630310057,
      "num_input_tokens_seen": 749125880,
      "step": 12495,
      "train_runtime": 160491.8689,
      "train_tokens_per_second": 4667.687
    },
    {
      "epoch": 1.5943877551020407,
      "grad_norm": 0.613797587978045,
      "learning_rate": 9.815765859410586e-07,
      "loss": 0.23654673099517823,
      "num_input_tokens_seen": 749431832,
      "step": 12500,
      "train_runtime": 160552.6619,
      "train_tokens_per_second": 4667.826
    },
    {
      "epoch": 1.5950255102040818,
      "grad_norm": 0.6863986151342331,
      "learning_rate": 9.785980197926242e-07,
      "loss": 0.2438525676727295,
      "num_input_tokens_seen": 749731576,
      "step": 12505,
      "train_runtime": 160614.7497,
      "train_tokens_per_second": 4667.887
    },
    {
      "epoch": 1.5956632653061225,
      "grad_norm": 0.7793642516046543,
      "learning_rate": 9.7562348940193e-07,
      "loss": 0.22856268882751465,
      "num_input_tokens_seen": 750026320,
      "step": 12510,
      "train_runtime": 160676.5839,
      "train_tokens_per_second": 4667.925
    },
    {
      "epoch": 1.5963010204081631,
      "grad_norm": 0.6918633876196689,
      "learning_rate": 9.726529977541243e-07,
      "loss": 0.2514913082122803,
      "num_input_tokens_seen": 750315040,
      "step": 12515,
      "train_runtime": 160731.9811,
      "train_tokens_per_second": 4668.113
    },
    {
      "epoch": 1.5969387755102042,
      "grad_norm": 0.6298207620096105,
      "learning_rate": 9.696865478303075e-07,
      "loss": 0.25145344734191893,
      "num_input_tokens_seen": 750618512,
      "step": 12520,
      "train_runtime": 160801.8338,
      "train_tokens_per_second": 4667.972
    },
    {
      "epoch": 1.597576530612245,
      "grad_norm": 0.6782354727972346,
      "learning_rate": 9.667241426075152e-07,
      "loss": 0.2843494176864624,
      "num_input_tokens_seen": 750934152,
      "step": 12525,
      "train_runtime": 160855.0452,
      "train_tokens_per_second": 4668.39
    },
    {
      "epoch": 1.5982142857142856,
      "grad_norm": 0.6599116171683902,
      "learning_rate": 9.637657850587289e-07,
      "loss": 0.24234216213226317,
      "num_input_tokens_seen": 751230872,
      "step": 12530,
      "train_runtime": 160923.4965,
      "train_tokens_per_second": 4668.249
    },
    {
      "epoch": 1.5988520408163265,
      "grad_norm": 0.671000036187092,
      "learning_rate": 9.608114781528671e-07,
      "loss": 0.2262669563293457,
      "num_input_tokens_seen": 751520384,
      "step": 12535,
      "train_runtime": 160981.1021,
      "train_tokens_per_second": 4668.376
    },
    {
      "epoch": 1.5994897959183674,
      "grad_norm": 0.6439374254598299,
      "learning_rate": 9.578612248547836e-07,
      "loss": 0.23630757331848146,
      "num_input_tokens_seen": 751811720,
      "step": 12540,
      "train_runtime": 161058.3484,
      "train_tokens_per_second": 4667.946
    },
    {
      "epoch": 1.600127551020408,
      "grad_norm": 0.6694706643664425,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.21743435859680177,
      "num_input_tokens_seen": 752118136,
      "step": 12545,
      "train_runtime": 161129.8272,
      "train_tokens_per_second": 4667.777
    },
    {
      "epoch": 1.600765306122449,
      "grad_norm": 0.6859891226469327,
      "learning_rate": 9.519728909210202e-07,
      "loss": 0.2244157075881958,
      "num_input_tokens_seen": 752412704,
      "step": 12550,
      "train_runtime": 161188.7732,
      "train_tokens_per_second": 4667.898
    },
    {
      "epoch": 1.6014030612244898,
      "grad_norm": 0.6449266735319855,
      "learning_rate": 9.49034816194695e-07,
      "loss": 0.22257812023162843,
      "num_input_tokens_seen": 752713704,
      "step": 12555,
      "train_runtime": 161252.8927,
      "train_tokens_per_second": 4667.908
    },
    {
      "epoch": 1.6020408163265305,
      "grad_norm": 0.7676452639078382,
      "learning_rate": 9.461008068948502e-07,
      "loss": 0.24138007164001465,
      "num_input_tokens_seen": 753003032,
      "step": 12560,
      "train_runtime": 161317.0235,
      "train_tokens_per_second": 4667.846
    },
    {
      "epoch": 1.6026785714285714,
      "grad_norm": 0.6130795704749278,
      "learning_rate": 9.431708659659694e-07,
      "loss": 0.22036666870117189,
      "num_input_tokens_seen": 753310144,
      "step": 12565,
      "train_runtime": 161380.829,
      "train_tokens_per_second": 4667.904
    },
    {
      "epoch": 1.6033163265306123,
      "grad_norm": 0.6408949066338637,
      "learning_rate": 9.402449963484528e-07,
      "loss": 0.2612826585769653,
      "num_input_tokens_seen": 753603880,
      "step": 12570,
      "train_runtime": 161456.0607,
      "train_tokens_per_second": 4667.548
    },
    {
      "epoch": 1.603954081632653,
      "grad_norm": 0.699654142064855,
      "learning_rate": 9.373232009786154e-07,
      "loss": 0.24999663829803467,
      "num_input_tokens_seen": 753892464,
      "step": 12575,
      "train_runtime": 161522.4346,
      "train_tokens_per_second": 4667.416
    },
    {
      "epoch": 1.6045918367346939,
      "grad_norm": 0.6384457950020472,
      "learning_rate": 9.344054827886839e-07,
      "loss": 0.2503360748291016,
      "num_input_tokens_seen": 754190928,
      "step": 12580,
      "train_runtime": 161587.9504,
      "train_tokens_per_second": 4667.371
    },
    {
      "epoch": 1.6052295918367347,
      "grad_norm": 0.737978309490727,
      "learning_rate": 9.314918447067877e-07,
      "loss": 0.21660540103912354,
      "num_input_tokens_seen": 754472440,
      "step": 12585,
      "train_runtime": 161660.6712,
      "train_tokens_per_second": 4667.013
    },
    {
      "epoch": 1.6058673469387754,
      "grad_norm": 0.6791704799144782,
      "learning_rate": 9.285822896569724e-07,
      "loss": 0.22802515029907228,
      "num_input_tokens_seen": 754761096,
      "step": 12590,
      "train_runtime": 161730.4204,
      "train_tokens_per_second": 4666.785
    },
    {
      "epoch": 1.6065051020408163,
      "grad_norm": 0.6422862658942546,
      "learning_rate": 9.256768205591754e-07,
      "loss": 0.2276700496673584,
      "num_input_tokens_seen": 755067464,
      "step": 12595,
      "train_runtime": 161791.3075,
      "train_tokens_per_second": 4666.922
    },
    {
      "epoch": 1.6071428571428572,
      "grad_norm": 0.5970500161006491,
      "learning_rate": 9.227754403292377e-07,
      "loss": 0.2209826946258545,
      "num_input_tokens_seen": 755364080,
      "step": 12600,
      "train_runtime": 161861.3277,
      "train_tokens_per_second": 4666.736
    },
    {
      "epoch": 1.6077806122448979,
      "grad_norm": 0.6395583620104077,
      "learning_rate": 9.198781518789007e-07,
      "loss": 0.25316319465637205,
      "num_input_tokens_seen": 755661280,
      "step": 12605,
      "train_runtime": 161918.6041,
      "train_tokens_per_second": 4666.921
    },
    {
      "epoch": 1.6084183673469388,
      "grad_norm": 0.6418346734876251,
      "learning_rate": 9.169849581157925e-07,
      "loss": 0.25020828247070315,
      "num_input_tokens_seen": 755980408,
      "step": 12610,
      "train_runtime": 161987.8425,
      "train_tokens_per_second": 4666.896
    },
    {
      "epoch": 1.6090561224489797,
      "grad_norm": 0.5610877522772107,
      "learning_rate": 9.140958619434353e-07,
      "loss": 0.24742977619171141,
      "num_input_tokens_seen": 756282560,
      "step": 12615,
      "train_runtime": 162051.1772,
      "train_tokens_per_second": 4666.937
    },
    {
      "epoch": 1.6096938775510203,
      "grad_norm": 0.6935628664099879,
      "learning_rate": 9.112108662612401e-07,
      "loss": 0.24512836933135987,
      "num_input_tokens_seen": 756588056,
      "step": 12620,
      "train_runtime": 162111.8387,
      "train_tokens_per_second": 4667.075
    },
    {
      "epoch": 1.6103316326530612,
      "grad_norm": 0.5612648974736708,
      "learning_rate": 9.083299739645007e-07,
      "loss": 0.22108681201934816,
      "num_input_tokens_seen": 756894296,
      "step": 12625,
      "train_runtime": 162183.7017,
      "train_tokens_per_second": 4666.895
    },
    {
      "epoch": 1.6109693877551021,
      "grad_norm": 0.7867473407636176,
      "learning_rate": 9.054531879443945e-07,
      "loss": 0.23173933029174804,
      "num_input_tokens_seen": 757199688,
      "step": 12630,
      "train_runtime": 162244.7123,
      "train_tokens_per_second": 4667.022
    },
    {
      "epoch": 1.6116071428571428,
      "grad_norm": 0.689728429677776,
      "learning_rate": 9.025805110879765e-07,
      "loss": 0.25670275688171384,
      "num_input_tokens_seen": 757499624,
      "step": 12635,
      "train_runtime": 162309.6087,
      "train_tokens_per_second": 4667.004
    },
    {
      "epoch": 1.6122448979591837,
      "grad_norm": 0.6108552373822652,
      "learning_rate": 8.997119462781784e-07,
      "loss": 0.24854421615600586,
      "num_input_tokens_seen": 757804672,
      "step": 12640,
      "train_runtime": 162369.0132,
      "train_tokens_per_second": 4667.175
    },
    {
      "epoch": 1.6128826530612246,
      "grad_norm": 0.6791550048614775,
      "learning_rate": 8.968474963938062e-07,
      "loss": 0.2311779499053955,
      "num_input_tokens_seen": 758105208,
      "step": 12645,
      "train_runtime": 162434.0353,
      "train_tokens_per_second": 4667.157
    },
    {
      "epoch": 1.6135204081632653,
      "grad_norm": 0.6599731643148613,
      "learning_rate": 8.939871643095344e-07,
      "loss": 0.25216753482818605,
      "num_input_tokens_seen": 758406120,
      "step": 12650,
      "train_runtime": 162487.0774,
      "train_tokens_per_second": 4667.486
    },
    {
      "epoch": 1.6141581632653061,
      "grad_norm": 0.6583367280408391,
      "learning_rate": 8.911309528959072e-07,
      "loss": 0.24410347938537597,
      "num_input_tokens_seen": 758701624,
      "step": 12655,
      "train_runtime": 162547.4265,
      "train_tokens_per_second": 4667.571
    },
    {
      "epoch": 1.614795918367347,
      "grad_norm": 0.6178294445564312,
      "learning_rate": 8.882788650193308e-07,
      "loss": 0.2473817825317383,
      "num_input_tokens_seen": 759014720,
      "step": 12660,
      "train_runtime": 162611.2003,
      "train_tokens_per_second": 4667.666
    },
    {
      "epoch": 1.6154336734693877,
      "grad_norm": 0.6493910949400037,
      "learning_rate": 8.854309035420772e-07,
      "loss": 0.21530449390411377,
      "num_input_tokens_seen": 759297760,
      "step": 12665,
      "train_runtime": 162686.0367,
      "train_tokens_per_second": 4667.258
    },
    {
      "epoch": 1.6160714285714286,
      "grad_norm": 0.6400351711700686,
      "learning_rate": 8.8258707132227e-07,
      "loss": 0.2471524477005005,
      "num_input_tokens_seen": 759591568,
      "step": 12670,
      "train_runtime": 162755.0653,
      "train_tokens_per_second": 4667.084
    },
    {
      "epoch": 1.6167091836734695,
      "grad_norm": 0.6115359996465839,
      "learning_rate": 8.797473712138976e-07,
      "loss": 0.23382058143615722,
      "num_input_tokens_seen": 759902808,
      "step": 12675,
      "train_runtime": 162810.921,
      "train_tokens_per_second": 4667.395
    },
    {
      "epoch": 1.6173469387755102,
      "grad_norm": 0.6887977770934273,
      "learning_rate": 8.769118060667969e-07,
      "loss": 0.24280138015747071,
      "num_input_tokens_seen": 760202816,
      "step": 12680,
      "train_runtime": 162878.1636,
      "train_tokens_per_second": 4667.31
    },
    {
      "epoch": 1.617984693877551,
      "grad_norm": 0.6529361686821481,
      "learning_rate": 8.740803787266522e-07,
      "loss": 0.23381223678588867,
      "num_input_tokens_seen": 760505048,
      "step": 12685,
      "train_runtime": 162936.5705,
      "train_tokens_per_second": 4667.491
    },
    {
      "epoch": 1.618622448979592,
      "grad_norm": 0.65762465735013,
      "learning_rate": 8.712530920350026e-07,
      "loss": 0.2271873712539673,
      "num_input_tokens_seen": 760808016,
      "step": 12690,
      "train_runtime": 163004.7719,
      "train_tokens_per_second": 4667.397
    },
    {
      "epoch": 1.6192602040816326,
      "grad_norm": 0.6924109243905973,
      "learning_rate": 8.684299488292275e-07,
      "loss": 0.22521586418151857,
      "num_input_tokens_seen": 761094160,
      "step": 12695,
      "train_runtime": 163071.8349,
      "train_tokens_per_second": 4667.232
    },
    {
      "epoch": 1.6198979591836735,
      "grad_norm": 0.6763067913069459,
      "learning_rate": 8.656109519425454e-07,
      "loss": 0.21990556716918946,
      "num_input_tokens_seen": 761391048,
      "step": 12700,
      "train_runtime": 163141.7484,
      "train_tokens_per_second": 4667.052
    },
    {
      "epoch": 1.6205357142857144,
      "grad_norm": 0.6926440670904408,
      "learning_rate": 8.627961042040183e-07,
      "loss": 0.22145428657531738,
      "num_input_tokens_seen": 761693320,
      "step": 12705,
      "train_runtime": 163195.8345,
      "train_tokens_per_second": 4667.358
    },
    {
      "epoch": 1.621173469387755,
      "grad_norm": 0.60466273211022,
      "learning_rate": 8.599854084385422e-07,
      "loss": 0.22990632057189941,
      "num_input_tokens_seen": 761997136,
      "step": 12710,
      "train_runtime": 163258.7861,
      "train_tokens_per_second": 4667.419
    },
    {
      "epoch": 1.6218112244897958,
      "grad_norm": 0.7343372763610768,
      "learning_rate": 8.571788674668469e-07,
      "loss": 0.2373216152191162,
      "num_input_tokens_seen": 762285056,
      "step": 12715,
      "train_runtime": 163329.7229,
      "train_tokens_per_second": 4667.155
    },
    {
      "epoch": 1.6224489795918369,
      "grad_norm": 0.6287937244240275,
      "learning_rate": 8.543764841054919e-07,
      "loss": 0.2283334493637085,
      "num_input_tokens_seen": 762592192,
      "step": 12720,
      "train_runtime": 163383.1502,
      "train_tokens_per_second": 4667.508
    },
    {
      "epoch": 1.6230867346938775,
      "grad_norm": 0.6521525005005768,
      "learning_rate": 8.515782611668649e-07,
      "loss": 0.23148722648620607,
      "num_input_tokens_seen": 762890992,
      "step": 12725,
      "train_runtime": 163449.8333,
      "train_tokens_per_second": 4667.432
    },
    {
      "epoch": 1.6237244897959182,
      "grad_norm": 0.6600663449520306,
      "learning_rate": 8.487842014591779e-07,
      "loss": 0.21741807460784912,
      "num_input_tokens_seen": 763193856,
      "step": 12730,
      "train_runtime": 163515.5507,
      "train_tokens_per_second": 4667.408
    },
    {
      "epoch": 1.6243622448979593,
      "grad_norm": 0.7789719798647102,
      "learning_rate": 8.459943077864651e-07,
      "loss": 0.2391869068145752,
      "num_input_tokens_seen": 763504336,
      "step": 12735,
      "train_runtime": 163588.0196,
      "train_tokens_per_second": 4667.239
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.6134353333305429,
      "learning_rate": 8.432085829485792e-07,
      "loss": 0.2195131778717041,
      "num_input_tokens_seen": 763806240,
      "step": 12740,
      "train_runtime": 163647.8414,
      "train_tokens_per_second": 4667.377
    },
    {
      "epoch": 1.6256377551020407,
      "grad_norm": 0.6538388554272606,
      "learning_rate": 8.404270297411904e-07,
      "loss": 0.2050943374633789,
      "num_input_tokens_seen": 764107216,
      "step": 12745,
      "train_runtime": 163710.3535,
      "train_tokens_per_second": 4667.434
    },
    {
      "epoch": 1.6262755102040818,
      "grad_norm": 0.5831986692183562,
      "learning_rate": 8.376496509557808e-07,
      "loss": 0.23582563400268555,
      "num_input_tokens_seen": 764417856,
      "step": 12750,
      "train_runtime": 163780.9757,
      "train_tokens_per_second": 4667.318
    },
    {
      "epoch": 1.6269132653061225,
      "grad_norm": 0.7391994258303932,
      "learning_rate": 8.348764493796447e-07,
      "loss": 0.2493993043899536,
      "num_input_tokens_seen": 764723472,
      "step": 12755,
      "train_runtime": 163838.2995,
      "train_tokens_per_second": 4667.55
    },
    {
      "epoch": 1.6275510204081631,
      "grad_norm": 0.5893725270395765,
      "learning_rate": 8.321074277958835e-07,
      "loss": 0.2396076202392578,
      "num_input_tokens_seen": 765025456,
      "step": 12760,
      "train_runtime": 163904.8398,
      "train_tokens_per_second": 4667.498
    },
    {
      "epoch": 1.6281887755102042,
      "grad_norm": 0.683388506903838,
      "learning_rate": 8.293425889834045e-07,
      "loss": 0.2349012851715088,
      "num_input_tokens_seen": 765324376,
      "step": 12765,
      "train_runtime": 163967.0832,
      "train_tokens_per_second": 4667.549
    },
    {
      "epoch": 1.628826530612245,
      "grad_norm": 0.5994200421359469,
      "learning_rate": 8.265819357169124e-07,
      "loss": 0.22822265625,
      "num_input_tokens_seen": 765624400,
      "step": 12770,
      "train_runtime": 164036.2669,
      "train_tokens_per_second": 4667.409
    },
    {
      "epoch": 1.6294642857142856,
      "grad_norm": 0.6790832287384436,
      "learning_rate": 8.238254707669208e-07,
      "loss": 0.2320777177810669,
      "num_input_tokens_seen": 765922240,
      "step": 12775,
      "train_runtime": 164097.3783,
      "train_tokens_per_second": 4667.486
    },
    {
      "epoch": 1.6301020408163265,
      "grad_norm": 0.5237539020463818,
      "learning_rate": 8.210731968997331e-07,
      "loss": 0.22750141620635986,
      "num_input_tokens_seen": 766229384,
      "step": 12780,
      "train_runtime": 164166.5126,
      "train_tokens_per_second": 4667.391
    },
    {
      "epoch": 1.6307397959183674,
      "grad_norm": 0.803909351774539,
      "learning_rate": 8.183251168774476e-07,
      "loss": 0.2525877237319946,
      "num_input_tokens_seen": 766534376,
      "step": 12785,
      "train_runtime": 164225.4142,
      "train_tokens_per_second": 4667.575
    },
    {
      "epoch": 1.631377551020408,
      "grad_norm": 0.6228117481387875,
      "learning_rate": 8.155812334579532e-07,
      "loss": 0.24632534980773926,
      "num_input_tokens_seen": 766829832,
      "step": 12790,
      "train_runtime": 164288.5263,
      "train_tokens_per_second": 4667.58
    },
    {
      "epoch": 1.632015306122449,
      "grad_norm": 0.6468751335422971,
      "learning_rate": 8.128415493949326e-07,
      "loss": 0.24460504055023194,
      "num_input_tokens_seen": 767132728,
      "step": 12795,
      "train_runtime": 164364.9839,
      "train_tokens_per_second": 4667.252
    },
    {
      "epoch": 1.6326530612244898,
      "grad_norm": 0.7048209442808318,
      "learning_rate": 8.101060674378463e-07,
      "loss": 0.24550023078918456,
      "num_input_tokens_seen": 767417392,
      "step": 12800,
      "train_runtime": 164428.0565,
      "train_tokens_per_second": 4667.193
    },
    {
      "epoch": 1.6332908163265305,
      "grad_norm": 0.7051579119075166,
      "learning_rate": 8.073747903319418e-07,
      "loss": 0.22775611877441407,
      "num_input_tokens_seen": 767705848,
      "step": 12805,
      "train_runtime": 164500.9506,
      "train_tokens_per_second": 4666.878
    },
    {
      "epoch": 1.6339285714285714,
      "grad_norm": 0.648719003052251,
      "learning_rate": 8.046477208182501e-07,
      "loss": 0.23567018508911133,
      "num_input_tokens_seen": 768013696,
      "step": 12810,
      "train_runtime": 164571.091,
      "train_tokens_per_second": 4666.759
    },
    {
      "epoch": 1.6345663265306123,
      "grad_norm": 0.6773509046491577,
      "learning_rate": 8.019248616335717e-07,
      "loss": 0.23838484287261963,
      "num_input_tokens_seen": 768327288,
      "step": 12815,
      "train_runtime": 164641.8254,
      "train_tokens_per_second": 4666.659
    },
    {
      "epoch": 1.635204081632653,
      "grad_norm": 0.5646156572146046,
      "learning_rate": 7.992062155104879e-07,
      "loss": 0.2176501512527466,
      "num_input_tokens_seen": 768616768,
      "step": 12820,
      "train_runtime": 164706.7403,
      "train_tokens_per_second": 4666.577
    },
    {
      "epoch": 1.6358418367346939,
      "grad_norm": 0.6954283907400376,
      "learning_rate": 7.964917851773496e-07,
      "loss": 0.24968400001525878,
      "num_input_tokens_seen": 768920584,
      "step": 12825,
      "train_runtime": 164776.3047,
      "train_tokens_per_second": 4666.451
    },
    {
      "epoch": 1.6364795918367347,
      "grad_norm": 0.6214926377318728,
      "learning_rate": 7.937815733582771e-07,
      "loss": 0.23613603115081788,
      "num_input_tokens_seen": 769214152,
      "step": 12830,
      "train_runtime": 164840.5597,
      "train_tokens_per_second": 4666.413
    },
    {
      "epoch": 1.6371173469387754,
      "grad_norm": 0.7560759819320956,
      "learning_rate": 7.910755827731576e-07,
      "loss": 0.25022363662719727,
      "num_input_tokens_seen": 769499728,
      "step": 12835,
      "train_runtime": 164901.9149,
      "train_tokens_per_second": 4666.409
    },
    {
      "epoch": 1.6377551020408163,
      "grad_norm": 0.6149173034823581,
      "learning_rate": 7.883738161376414e-07,
      "loss": 0.23003995418548584,
      "num_input_tokens_seen": 769809576,
      "step": 12840,
      "train_runtime": 164967.6093,
      "train_tokens_per_second": 4666.429
    },
    {
      "epoch": 1.6383928571428572,
      "grad_norm": 0.7002756775623914,
      "learning_rate": 7.856762761631398e-07,
      "loss": 0.22579755783081054,
      "num_input_tokens_seen": 770103216,
      "step": 12845,
      "train_runtime": 165037.0297,
      "train_tokens_per_second": 4666.245
    },
    {
      "epoch": 1.6390306122448979,
      "grad_norm": 0.602518033828995,
      "learning_rate": 7.829829655568239e-07,
      "loss": 0.221382212638855,
      "num_input_tokens_seen": 770399288,
      "step": 12850,
      "train_runtime": 165100.3255,
      "train_tokens_per_second": 4666.249
    },
    {
      "epoch": 1.6396683673469388,
      "grad_norm": 0.6454536317219705,
      "learning_rate": 7.802938870216159e-07,
      "loss": 0.2331010103225708,
      "num_input_tokens_seen": 770705752,
      "step": 12855,
      "train_runtime": 165165.3793,
      "train_tokens_per_second": 4666.267
    },
    {
      "epoch": 1.6403061224489797,
      "grad_norm": 0.6313468269887681,
      "learning_rate": 7.776090432561967e-07,
      "loss": 0.2524754524230957,
      "num_input_tokens_seen": 771004912,
      "step": 12860,
      "train_runtime": 165239.7967,
      "train_tokens_per_second": 4665.976
    },
    {
      "epoch": 1.6409438775510203,
      "grad_norm": 0.6439996542817811,
      "learning_rate": 7.749284369549954e-07,
      "loss": 0.2454533576965332,
      "num_input_tokens_seen": 771297416,
      "step": 12865,
      "train_runtime": 165307.6819,
      "train_tokens_per_second": 4665.829
    },
    {
      "epoch": 1.6415816326530612,
      "grad_norm": 0.7231457912328102,
      "learning_rate": 7.722520708081843e-07,
      "loss": 0.23851680755615234,
      "num_input_tokens_seen": 771602160,
      "step": 12870,
      "train_runtime": 165358.9515,
      "train_tokens_per_second": 4666.226
    },
    {
      "epoch": 1.6422193877551021,
      "grad_norm": 0.6945461006719696,
      "learning_rate": 7.695799475016846e-07,
      "loss": 0.2278822898864746,
      "num_input_tokens_seen": 771900656,
      "step": 12875,
      "train_runtime": 165409.9536,
      "train_tokens_per_second": 4666.591
    },
    {
      "epoch": 1.6428571428571428,
      "grad_norm": 0.6459446946379738,
      "learning_rate": 7.669120697171617e-07,
      "loss": 0.23678188323974608,
      "num_input_tokens_seen": 772199328,
      "step": 12880,
      "train_runtime": 165475.9357,
      "train_tokens_per_second": 4666.535
    },
    {
      "epoch": 1.6434948979591837,
      "grad_norm": 0.7076929676685086,
      "learning_rate": 7.642484401320138e-07,
      "loss": 0.26769237518310546,
      "num_input_tokens_seen": 772488824,
      "step": 12885,
      "train_runtime": 165540.8839,
      "train_tokens_per_second": 4666.453
    },
    {
      "epoch": 1.6441326530612246,
      "grad_norm": 0.6801416091689853,
      "learning_rate": 7.615890614193788e-07,
      "loss": 0.2113205909729004,
      "num_input_tokens_seen": 772777152,
      "step": 12890,
      "train_runtime": 165604.5606,
      "train_tokens_per_second": 4666.4
    },
    {
      "epoch": 1.6447704081632653,
      "grad_norm": 0.6286940841455992,
      "learning_rate": 7.589339362481334e-07,
      "loss": 0.26149330139160154,
      "num_input_tokens_seen": 773086576,
      "step": 12895,
      "train_runtime": 165665.4672,
      "train_tokens_per_second": 4666.552
    },
    {
      "epoch": 1.6454081632653061,
      "grad_norm": 0.643186966516067,
      "learning_rate": 7.562830672828775e-07,
      "loss": 0.24932973384857177,
      "num_input_tokens_seen": 773394056,
      "step": 12900,
      "train_runtime": 165722.7588,
      "train_tokens_per_second": 4666.794
    },
    {
      "epoch": 1.646045918367347,
      "grad_norm": 0.768531296265755,
      "learning_rate": 7.536364571839439e-07,
      "loss": 0.24275400638580322,
      "num_input_tokens_seen": 773684880,
      "step": 12905,
      "train_runtime": 165780.6721,
      "train_tokens_per_second": 4666.918
    },
    {
      "epoch": 1.6466836734693877,
      "grad_norm": 0.5946491667331976,
      "learning_rate": 7.509941086073907e-07,
      "loss": 0.24175541400909423,
      "num_input_tokens_seen": 773966264,
      "step": 12910,
      "train_runtime": 165850.1885,
      "train_tokens_per_second": 4666.659
    },
    {
      "epoch": 1.6473214285714286,
      "grad_norm": 0.6616771306782862,
      "learning_rate": 7.483560242049998e-07,
      "loss": 0.2291048526763916,
      "num_input_tokens_seen": 774255256,
      "step": 12915,
      "train_runtime": 165913.8996,
      "train_tokens_per_second": 4666.609
    },
    {
      "epoch": 1.6479591836734695,
      "grad_norm": 0.637810008406539,
      "learning_rate": 7.457222066242725e-07,
      "loss": 0.2123131275177002,
      "num_input_tokens_seen": 774563400,
      "step": 12920,
      "train_runtime": 165970.0259,
      "train_tokens_per_second": 4666.887
    },
    {
      "epoch": 1.6485969387755102,
      "grad_norm": 0.5833980269359663,
      "learning_rate": 7.430926585084292e-07,
      "loss": 0.24314284324645996,
      "num_input_tokens_seen": 774865184,
      "step": 12925,
      "train_runtime": 166046.0083,
      "train_tokens_per_second": 4666.569
    },
    {
      "epoch": 1.649234693877551,
      "grad_norm": 0.6269677811841762,
      "learning_rate": 7.404673824964048e-07,
      "loss": 0.25867276191711425,
      "num_input_tokens_seen": 775151648,
      "step": 12930,
      "train_runtime": 166111.804,
      "train_tokens_per_second": 4666.445
    },
    {
      "epoch": 1.649872448979592,
      "grad_norm": 0.7009629658692085,
      "learning_rate": 7.37846381222847e-07,
      "loss": 0.23241810798645018,
      "num_input_tokens_seen": 775451920,
      "step": 12935,
      "train_runtime": 166161.0996,
      "train_tokens_per_second": 4666.868
    },
    {
      "epoch": 1.6505102040816326,
      "grad_norm": 0.6850529082515331,
      "learning_rate": 7.352296573181139e-07,
      "loss": 0.24730679988861085,
      "num_input_tokens_seen": 775764736,
      "step": 12940,
      "train_runtime": 166217.2646,
      "train_tokens_per_second": 4667.173
    },
    {
      "epoch": 1.6511479591836735,
      "grad_norm": 0.6800982510258878,
      "learning_rate": 7.326172134082704e-07,
      "loss": 0.23886303901672362,
      "num_input_tokens_seen": 776069736,
      "step": 12945,
      "train_runtime": 166278.8623,
      "train_tokens_per_second": 4667.278
    },
    {
      "epoch": 1.6517857142857144,
      "grad_norm": 0.6689264040173888,
      "learning_rate": 7.300090521150882e-07,
      "loss": 0.2485065460205078,
      "num_input_tokens_seen": 776357896,
      "step": 12950,
      "train_runtime": 166339.9362,
      "train_tokens_per_second": 4667.297
    },
    {
      "epoch": 1.652423469387755,
      "grad_norm": 0.681451726147646,
      "learning_rate": 7.274051760560364e-07,
      "loss": 0.22038323879241944,
      "num_input_tokens_seen": 776649896,
      "step": 12955,
      "train_runtime": 166401.8881,
      "train_tokens_per_second": 4667.314
    },
    {
      "epoch": 1.6530612244897958,
      "grad_norm": 0.7070238248522148,
      "learning_rate": 7.248055878442861e-07,
      "loss": 0.24848322868347167,
      "num_input_tokens_seen": 776949464,
      "step": 12960,
      "train_runtime": 166470.1233,
      "train_tokens_per_second": 4667.201
    },
    {
      "epoch": 1.6536989795918369,
      "grad_norm": 0.6734788082036008,
      "learning_rate": 7.222102900887102e-07,
      "loss": 0.24950251579284669,
      "num_input_tokens_seen": 777237288,
      "step": 12965,
      "train_runtime": 166533.3783,
      "train_tokens_per_second": 4667.156
    },
    {
      "epoch": 1.6543367346938775,
      "grad_norm": 0.6336921668653426,
      "learning_rate": 7.196192853938672e-07,
      "loss": 0.24173741340637206,
      "num_input_tokens_seen": 777544344,
      "step": 12970,
      "train_runtime": 166599.3719,
      "train_tokens_per_second": 4667.151
    },
    {
      "epoch": 1.6549744897959182,
      "grad_norm": 0.6225871386378725,
      "learning_rate": 7.170325763600106e-07,
      "loss": 0.2383871555328369,
      "num_input_tokens_seen": 777845928,
      "step": 12975,
      "train_runtime": 166671.7908,
      "train_tokens_per_second": 4666.932
    },
    {
      "epoch": 1.6556122448979593,
      "grad_norm": 0.7231042620966402,
      "learning_rate": 7.144501655830882e-07,
      "loss": 0.24619700908660888,
      "num_input_tokens_seen": 778148472,
      "step": 12980,
      "train_runtime": 166732.7476,
      "train_tokens_per_second": 4667.04
    },
    {
      "epoch": 1.65625,
      "grad_norm": 0.6940467107240171,
      "learning_rate": 7.118720556547259e-07,
      "loss": 0.24862949848175048,
      "num_input_tokens_seen": 778453920,
      "step": 12985,
      "train_runtime": 166799.4558,
      "train_tokens_per_second": 4667.005
    },
    {
      "epoch": 1.6568877551020407,
      "grad_norm": 0.5750729482589093,
      "learning_rate": 7.092982491622374e-07,
      "loss": 0.21929986476898194,
      "num_input_tokens_seen": 778750288,
      "step": 12990,
      "train_runtime": 166860.6477,
      "train_tokens_per_second": 4667.07
    },
    {
      "epoch": 1.6575255102040818,
      "grad_norm": 0.6489435981619424,
      "learning_rate": 7.067287486886181e-07,
      "loss": 0.23588500022888184,
      "num_input_tokens_seen": 779035720,
      "step": 12995,
      "train_runtime": 166922.4854,
      "train_tokens_per_second": 4667.051
    },
    {
      "epoch": 1.6581632653061225,
      "grad_norm": 0.7335545461464273,
      "learning_rate": 7.04163556812541e-07,
      "loss": 0.2200075149536133,
      "num_input_tokens_seen": 779323160,
      "step": 13000,
      "train_runtime": 166985.5123,
      "train_tokens_per_second": 4667.011
    },
    {
      "epoch": 1.6588010204081631,
      "grad_norm": 0.7174608148247873,
      "learning_rate": 7.016026761083556e-07,
      "loss": 0.2337871789932251,
      "num_input_tokens_seen": 779634368,
      "step": 13005,
      "train_runtime": 167053.8145,
      "train_tokens_per_second": 4666.965
    },
    {
      "epoch": 1.6594387755102042,
      "grad_norm": 0.6471454219871797,
      "learning_rate": 6.990461091460832e-07,
      "loss": 0.21842832565307618,
      "num_input_tokens_seen": 779949304,
      "step": 13010,
      "train_runtime": 167118.1744,
      "train_tokens_per_second": 4667.053
    },
    {
      "epoch": 1.660076530612245,
      "grad_norm": 0.6536762279535385,
      "learning_rate": 6.964938584914188e-07,
      "loss": 0.2667720317840576,
      "num_input_tokens_seen": 780251160,
      "step": 13015,
      "train_runtime": 167172.9351,
      "train_tokens_per_second": 4667.329
    },
    {
      "epoch": 1.6607142857142856,
      "grad_norm": 0.6513513980210826,
      "learning_rate": 6.939459267057236e-07,
      "loss": 0.23891825675964357,
      "num_input_tokens_seen": 780542248,
      "step": 13020,
      "train_runtime": 167238.6688,
      "train_tokens_per_second": 4667.235
    },
    {
      "epoch": 1.6613520408163265,
      "grad_norm": 0.6690796184761739,
      "learning_rate": 6.914023163460248e-07,
      "loss": 0.23524801731109618,
      "num_input_tokens_seen": 780835848,
      "step": 13025,
      "train_runtime": 167297.1518,
      "train_tokens_per_second": 4667.359
    },
    {
      "epoch": 1.6619897959183674,
      "grad_norm": 0.6494022476204548,
      "learning_rate": 6.888630299650134e-07,
      "loss": 0.23164083957672119,
      "num_input_tokens_seen": 781129632,
      "step": 13030,
      "train_runtime": 167368.3896,
      "train_tokens_per_second": 4667.128
    },
    {
      "epoch": 1.662627551020408,
      "grad_norm": 0.6726015289964822,
      "learning_rate": 6.863280701110409e-07,
      "loss": 0.2392362594604492,
      "num_input_tokens_seen": 781440616,
      "step": 13035,
      "train_runtime": 167441.0063,
      "train_tokens_per_second": 4666.961
    },
    {
      "epoch": 1.663265306122449,
      "grad_norm": 0.5515137395417501,
      "learning_rate": 6.837974393281171e-07,
      "loss": 0.23983891010284425,
      "num_input_tokens_seen": 781753168,
      "step": 13040,
      "train_runtime": 167500.6804,
      "train_tokens_per_second": 4667.164
    },
    {
      "epoch": 1.6639030612244898,
      "grad_norm": 0.6252944736562958,
      "learning_rate": 6.812711401559036e-07,
      "loss": 0.24457283020019532,
      "num_input_tokens_seen": 782054760,
      "step": 13045,
      "train_runtime": 167569.8892,
      "train_tokens_per_second": 4667.036
    },
    {
      "epoch": 1.6645408163265305,
      "grad_norm": 0.7101997830680842,
      "learning_rate": 6.787491751297215e-07,
      "loss": 0.2504986047744751,
      "num_input_tokens_seen": 782357888,
      "step": 13050,
      "train_runtime": 167628.6852,
      "train_tokens_per_second": 4667.208
    },
    {
      "epoch": 1.6651785714285714,
      "grad_norm": 0.6509633575539316,
      "learning_rate": 6.762315467805391e-07,
      "loss": 0.23212313652038574,
      "num_input_tokens_seen": 782670360,
      "step": 13055,
      "train_runtime": 167702.6376,
      "train_tokens_per_second": 4667.013
    },
    {
      "epoch": 1.6658163265306123,
      "grad_norm": 0.6520969667067783,
      "learning_rate": 6.737182576349682e-07,
      "loss": 0.2592411279678345,
      "num_input_tokens_seen": 782974768,
      "step": 13060,
      "train_runtime": 167771.9509,
      "train_tokens_per_second": 4666.899
    },
    {
      "epoch": 1.666454081632653,
      "grad_norm": 0.7351402048932675,
      "learning_rate": 6.712093102152739e-07,
      "loss": 0.21734251976013183,
      "num_input_tokens_seen": 783266480,
      "step": 13065,
      "train_runtime": 167829.777,
      "train_tokens_per_second": 4667.029
    },
    {
      "epoch": 1.6670918367346939,
      "grad_norm": 0.5915552133165983,
      "learning_rate": 6.687047070393599e-07,
      "loss": 0.2443648099899292,
      "num_input_tokens_seen": 783559480,
      "step": 13070,
      "train_runtime": 167880.3444,
      "train_tokens_per_second": 4667.369
    },
    {
      "epoch": 1.6677295918367347,
      "grad_norm": 0.6550876999474847,
      "learning_rate": 6.662044506207682e-07,
      "loss": 0.2329019069671631,
      "num_input_tokens_seen": 783856904,
      "step": 13075,
      "train_runtime": 167950.7256,
      "train_tokens_per_second": 4667.184
    },
    {
      "epoch": 1.6683673469387754,
      "grad_norm": 0.7762900714349238,
      "learning_rate": 6.637085434686818e-07,
      "loss": 0.24540672302246094,
      "num_input_tokens_seen": 784148840,
      "step": 13080,
      "train_runtime": 168014.4768,
      "train_tokens_per_second": 4667.15
    },
    {
      "epoch": 1.6690051020408163,
      "grad_norm": 0.5831630072297834,
      "learning_rate": 6.612169880879182e-07,
      "loss": 0.23792333602905275,
      "num_input_tokens_seen": 784434088,
      "step": 13085,
      "train_runtime": 168077.5178,
      "train_tokens_per_second": 4667.097
    },
    {
      "epoch": 1.6696428571428572,
      "grad_norm": 0.6974623502410974,
      "learning_rate": 6.587297869789267e-07,
      "loss": 0.25593783855438235,
      "num_input_tokens_seen": 784751688,
      "step": 13090,
      "train_runtime": 168129.4993,
      "train_tokens_per_second": 4667.543
    },
    {
      "epoch": 1.6702806122448979,
      "grad_norm": 0.5647333080925127,
      "learning_rate": 6.562469426377876e-07,
      "loss": 0.2372671604156494,
      "num_input_tokens_seen": 785058632,
      "step": 13095,
      "train_runtime": 168195.8955,
      "train_tokens_per_second": 4667.526
    },
    {
      "epoch": 1.6709183673469388,
      "grad_norm": 0.6270225616945366,
      "learning_rate": 6.537684575562092e-07,
      "loss": 0.23312408924102784,
      "num_input_tokens_seen": 785358784,
      "step": 13100,
      "train_runtime": 168258.5469,
      "train_tokens_per_second": 4667.571
    },
    {
      "epoch": 1.6715561224489797,
      "grad_norm": 0.6756205304063072,
      "learning_rate": 6.512943342215234e-07,
      "loss": 0.24810218811035156,
      "num_input_tokens_seen": 785664184,
      "step": 13105,
      "train_runtime": 168314.0596,
      "train_tokens_per_second": 4667.846
    },
    {
      "epoch": 1.6721938775510203,
      "grad_norm": 0.6922093905073984,
      "learning_rate": 6.48824575116686e-07,
      "loss": 0.2378976821899414,
      "num_input_tokens_seen": 785968104,
      "step": 13110,
      "train_runtime": 168379.0689,
      "train_tokens_per_second": 4667.849
    },
    {
      "epoch": 1.6728316326530612,
      "grad_norm": 0.6205064925837434,
      "learning_rate": 6.463591827202731e-07,
      "loss": 0.2181920289993286,
      "num_input_tokens_seen": 786276736,
      "step": 13115,
      "train_runtime": 168442.4663,
      "train_tokens_per_second": 4667.925
    },
    {
      "epoch": 1.6734693877551021,
      "grad_norm": 0.6322769862399636,
      "learning_rate": 6.43898159506478e-07,
      "loss": 0.2371192455291748,
      "num_input_tokens_seen": 786586048,
      "step": 13120,
      "train_runtime": 168492.2687,
      "train_tokens_per_second": 4668.381
    },
    {
      "epoch": 1.6741071428571428,
      "grad_norm": 0.6650323471277796,
      "learning_rate": 6.414415079451092e-07,
      "loss": 0.23830742835998536,
      "num_input_tokens_seen": 786901176,
      "step": 13125,
      "train_runtime": 168561.7027,
      "train_tokens_per_second": 4668.327
    },
    {
      "epoch": 1.6747448979591837,
      "grad_norm": 0.7764672572292787,
      "learning_rate": 6.389892305015882e-07,
      "loss": 0.2065678834915161,
      "num_input_tokens_seen": 787215728,
      "step": 13130,
      "train_runtime": 168625.5521,
      "train_tokens_per_second": 4668.425
    },
    {
      "epoch": 1.6753826530612246,
      "grad_norm": 0.6564512817800019,
      "learning_rate": 6.365413296369466e-07,
      "loss": 0.22998642921447754,
      "num_input_tokens_seen": 787527752,
      "step": 13135,
      "train_runtime": 168683.2805,
      "train_tokens_per_second": 4668.677
    },
    {
      "epoch": 1.6760204081632653,
      "grad_norm": 0.6018613749799228,
      "learning_rate": 6.340978078078253e-07,
      "loss": 0.23836636543273926,
      "num_input_tokens_seen": 787831184,
      "step": 13140,
      "train_runtime": 168754.1292,
      "train_tokens_per_second": 4668.515
    },
    {
      "epoch": 1.6766581632653061,
      "grad_norm": 0.6919813918443481,
      "learning_rate": 6.316586674664654e-07,
      "loss": 0.2525269269943237,
      "num_input_tokens_seen": 788136568,
      "step": 13145,
      "train_runtime": 168821.1967,
      "train_tokens_per_second": 4668.469
    },
    {
      "epoch": 1.677295918367347,
      "grad_norm": 0.6528782752201642,
      "learning_rate": 6.292239110607185e-07,
      "loss": 0.21977272033691406,
      "num_input_tokens_seen": 788439488,
      "step": 13150,
      "train_runtime": 168879.1023,
      "train_tokens_per_second": 4668.662
    },
    {
      "epoch": 1.6779336734693877,
      "grad_norm": 0.6089059934202922,
      "learning_rate": 6.26793541034032e-07,
      "loss": 0.23867287635803222,
      "num_input_tokens_seen": 788744384,
      "step": 13155,
      "train_runtime": 168940.4936,
      "train_tokens_per_second": 4668.77
    },
    {
      "epoch": 1.6785714285714286,
      "grad_norm": 0.6574053513640027,
      "learning_rate": 6.243675598254501e-07,
      "loss": 0.24979965686798095,
      "num_input_tokens_seen": 789060160,
      "step": 13160,
      "train_runtime": 168986.667,
      "train_tokens_per_second": 4669.363
    },
    {
      "epoch": 1.6792091836734695,
      "grad_norm": 0.7242613578315252,
      "learning_rate": 6.21945969869614e-07,
      "loss": 0.22568295001983643,
      "num_input_tokens_seen": 789333192,
      "step": 13165,
      "train_runtime": 169048.3113,
      "train_tokens_per_second": 4669.276
    },
    {
      "epoch": 1.6798469387755102,
      "grad_norm": 0.6199896905911603,
      "learning_rate": 6.195287735967626e-07,
      "loss": 0.2524087429046631,
      "num_input_tokens_seen": 789645128,
      "step": 13170,
      "train_runtime": 169110.1758,
      "train_tokens_per_second": 4669.412
    },
    {
      "epoch": 1.680484693877551,
      "grad_norm": 0.6120028452633463,
      "learning_rate": 6.171159734327164e-07,
      "loss": 0.2139357805252075,
      "num_input_tokens_seen": 789949752,
      "step": 13175,
      "train_runtime": 169181.7725,
      "train_tokens_per_second": 4669.237
    },
    {
      "epoch": 1.681122448979592,
      "grad_norm": 0.6758272900434804,
      "learning_rate": 6.147075717988904e-07,
      "loss": 0.2241701602935791,
      "num_input_tokens_seen": 790244912,
      "step": 13180,
      "train_runtime": 169251.2962,
      "train_tokens_per_second": 4669.063
    },
    {
      "epoch": 1.6817602040816326,
      "grad_norm": 0.6275304724163975,
      "learning_rate": 6.12303571112286e-07,
      "loss": 0.22757821083068847,
      "num_input_tokens_seen": 790546744,
      "step": 13185,
      "train_runtime": 169314.7937,
      "train_tokens_per_second": 4669.094
    },
    {
      "epoch": 1.6823979591836735,
      "grad_norm": 0.6270764021857508,
      "learning_rate": 6.099039737854834e-07,
      "loss": 0.25056018829345705,
      "num_input_tokens_seen": 790863352,
      "step": 13190,
      "train_runtime": 169370.6591,
      "train_tokens_per_second": 4669.424
    },
    {
      "epoch": 1.6830357142857144,
      "grad_norm": 0.6809793155677599,
      "learning_rate": 6.075087822266462e-07,
      "loss": 0.20741257667541504,
      "num_input_tokens_seen": 791171280,
      "step": 13195,
      "train_runtime": 169436.5744,
      "train_tokens_per_second": 4669.424
    },
    {
      "epoch": 1.683673469387755,
      "grad_norm": 0.5878364576150888,
      "learning_rate": 6.051179988395172e-07,
      "loss": 0.22054729461669922,
      "num_input_tokens_seen": 791486320,
      "step": 13200,
      "train_runtime": 169492.293,
      "train_tokens_per_second": 4669.748
    },
    {
      "epoch": 1.6843112244897958,
      "grad_norm": 0.7203585835063011,
      "learning_rate": 6.027316260234146e-07,
      "loss": 0.27373924255371096,
      "num_input_tokens_seen": 791793976,
      "step": 13205,
      "train_runtime": 169559.8932,
      "train_tokens_per_second": 4669.701
    },
    {
      "epoch": 1.6849489795918369,
      "grad_norm": 0.5865106613971609,
      "learning_rate": 6.003496661732294e-07,
      "loss": 0.23949275016784669,
      "num_input_tokens_seen": 792089672,
      "step": 13210,
      "train_runtime": 169625.1798,
      "train_tokens_per_second": 4669.647
    },
    {
      "epoch": 1.6855867346938775,
      "grad_norm": 0.7162563935397777,
      "learning_rate": 5.979721216794248e-07,
      "loss": 0.2525262594223022,
      "num_input_tokens_seen": 792382456,
      "step": 13215,
      "train_runtime": 169692.1822,
      "train_tokens_per_second": 4669.528
    },
    {
      "epoch": 1.6862244897959182,
      "grad_norm": 0.6748019054294158,
      "learning_rate": 5.955989949280339e-07,
      "loss": 0.23312003612518312,
      "num_input_tokens_seen": 792686240,
      "step": 13220,
      "train_runtime": 169752.6983,
      "train_tokens_per_second": 4669.653
    },
    {
      "epoch": 1.6868622448979593,
      "grad_norm": 0.6002767804793778,
      "learning_rate": 5.932302883006546e-07,
      "loss": 0.22830729484558104,
      "num_input_tokens_seen": 792990384,
      "step": 13225,
      "train_runtime": 169807.1179,
      "train_tokens_per_second": 4669.948
    },
    {
      "epoch": 1.6875,
      "grad_norm": 0.6021912915515264,
      "learning_rate": 5.908660041744473e-07,
      "loss": 0.22178337574005128,
      "num_input_tokens_seen": 793290664,
      "step": 13230,
      "train_runtime": 169870.3054,
      "train_tokens_per_second": 4669.978
    },
    {
      "epoch": 1.6881377551020407,
      "grad_norm": 0.6938996846206934,
      "learning_rate": 5.88506144922139e-07,
      "loss": 0.2509466171264648,
      "num_input_tokens_seen": 793592912,
      "step": 13235,
      "train_runtime": 169944.4774,
      "train_tokens_per_second": 4669.719
    },
    {
      "epoch": 1.6887755102040818,
      "grad_norm": 0.6052751439246761,
      "learning_rate": 5.861507129120136e-07,
      "loss": 0.2257427453994751,
      "num_input_tokens_seen": 793890616,
      "step": 13240,
      "train_runtime": 170005.1722,
      "train_tokens_per_second": 4669.803
    },
    {
      "epoch": 1.6894132653061225,
      "grad_norm": 0.7134737879671132,
      "learning_rate": 5.83799710507909e-07,
      "loss": 0.25400538444519044,
      "num_input_tokens_seen": 794185328,
      "step": 13245,
      "train_runtime": 170065.4423,
      "train_tokens_per_second": 4669.881
    },
    {
      "epoch": 1.6900510204081631,
      "grad_norm": 0.7258888105327589,
      "learning_rate": 5.814531400692208e-07,
      "loss": 0.26824064254760743,
      "num_input_tokens_seen": 794496184,
      "step": 13250,
      "train_runtime": 170124.3445,
      "train_tokens_per_second": 4670.091
    },
    {
      "epoch": 1.6906887755102042,
      "grad_norm": 0.6331302847597774,
      "learning_rate": 5.791110039508991e-07,
      "loss": 0.2345498561859131,
      "num_input_tokens_seen": 794793464,
      "step": 13255,
      "train_runtime": 170177.5113,
      "train_tokens_per_second": 4670.379
    },
    {
      "epoch": 1.691326530612245,
      "grad_norm": 0.6529794287832457,
      "learning_rate": 5.767733045034384e-07,
      "loss": 0.23070354461669923,
      "num_input_tokens_seen": 795101568,
      "step": 13260,
      "train_runtime": 170228.2802,
      "train_tokens_per_second": 4670.796
    },
    {
      "epoch": 1.6919642857142856,
      "grad_norm": 0.7162889014250071,
      "learning_rate": 5.744400440728826e-07,
      "loss": 0.24982943534851074,
      "num_input_tokens_seen": 795394008,
      "step": 13265,
      "train_runtime": 170287.2087,
      "train_tokens_per_second": 4670.897
    },
    {
      "epoch": 1.6926020408163265,
      "grad_norm": 0.7169809815043828,
      "learning_rate": 5.721112250008254e-07,
      "loss": 0.24927678108215331,
      "num_input_tokens_seen": 795698152,
      "step": 13270,
      "train_runtime": 170335.1968,
      "train_tokens_per_second": 4671.367
    },
    {
      "epoch": 1.6932397959183674,
      "grad_norm": 0.7229019398440883,
      "learning_rate": 5.697868496243958e-07,
      "loss": 0.2426917314529419,
      "num_input_tokens_seen": 795986528,
      "step": 13275,
      "train_runtime": 170403.9763,
      "train_tokens_per_second": 4671.173
    },
    {
      "epoch": 1.693877551020408,
      "grad_norm": 0.6823064047188397,
      "learning_rate": 5.674669202762684e-07,
      "loss": 0.2601162433624268,
      "num_input_tokens_seen": 796272320,
      "step": 13280,
      "train_runtime": 170469.4471,
      "train_tokens_per_second": 4671.056
    },
    {
      "epoch": 1.694515306122449,
      "grad_norm": 0.6795799080739466,
      "learning_rate": 5.651514392846535e-07,
      "loss": 0.22376790046691894,
      "num_input_tokens_seen": 796555656,
      "step": 13285,
      "train_runtime": 170540.7743,
      "train_tokens_per_second": 4670.764
    },
    {
      "epoch": 1.6951530612244898,
      "grad_norm": 0.6295307107669424,
      "learning_rate": 5.628404089732986e-07,
      "loss": 0.22962615489959717,
      "num_input_tokens_seen": 796856728,
      "step": 13290,
      "train_runtime": 170597.2119,
      "train_tokens_per_second": 4670.983
    },
    {
      "epoch": 1.6957908163265305,
      "grad_norm": 0.6399541549108114,
      "learning_rate": 5.605338316614839e-07,
      "loss": 0.2222541332244873,
      "num_input_tokens_seen": 797155856,
      "step": 13295,
      "train_runtime": 170665.2359,
      "train_tokens_per_second": 4670.874
    },
    {
      "epoch": 1.6964285714285714,
      "grad_norm": 0.6903583723354867,
      "learning_rate": 5.582317096640205e-07,
      "loss": 0.24221839904785156,
      "num_input_tokens_seen": 797460584,
      "step": 13300,
      "train_runtime": 170726.1599,
      "train_tokens_per_second": 4670.992
    },
    {
      "epoch": 1.6970663265306123,
      "grad_norm": 0.6184520869805074,
      "learning_rate": 5.559340452912488e-07,
      "loss": 0.23153388500213623,
      "num_input_tokens_seen": 797751920,
      "step": 13305,
      "train_runtime": 170790.754,
      "train_tokens_per_second": 4670.932
    },
    {
      "epoch": 1.697704081632653,
      "grad_norm": 0.7135832499278726,
      "learning_rate": 5.536408408490357e-07,
      "loss": 0.22246937751770018,
      "num_input_tokens_seen": 798043360,
      "step": 13310,
      "train_runtime": 170860.2489,
      "train_tokens_per_second": 4670.737
    },
    {
      "epoch": 1.6983418367346939,
      "grad_norm": 0.7078778236042537,
      "learning_rate": 5.513520986387721e-07,
      "loss": 0.24500691890716553,
      "num_input_tokens_seen": 798332080,
      "step": 13315,
      "train_runtime": 170927.4031,
      "train_tokens_per_second": 4670.592
    },
    {
      "epoch": 1.6989795918367347,
      "grad_norm": 0.7227787438942411,
      "learning_rate": 5.490678209573702e-07,
      "loss": 0.2364062786102295,
      "num_input_tokens_seen": 798607976,
      "step": 13320,
      "train_runtime": 170993.9651,
      "train_tokens_per_second": 4670.387
    },
    {
      "epoch": 1.6996173469387754,
      "grad_norm": 0.6375791863318517,
      "learning_rate": 5.467880100972633e-07,
      "loss": 0.2318039655685425,
      "num_input_tokens_seen": 798914928,
      "step": 13325,
      "train_runtime": 171066.7726,
      "train_tokens_per_second": 4670.193
    },
    {
      "epoch": 1.7002551020408163,
      "grad_norm": 0.7024225587032696,
      "learning_rate": 5.445126683463991e-07,
      "loss": 0.225300931930542,
      "num_input_tokens_seen": 799201912,
      "step": 13330,
      "train_runtime": 171136.0756,
      "train_tokens_per_second": 4669.979
    },
    {
      "epoch": 1.7008928571428572,
      "grad_norm": 0.6947061308185476,
      "learning_rate": 5.422417979882416e-07,
      "loss": 0.2390727996826172,
      "num_input_tokens_seen": 799500912,
      "step": 13335,
      "train_runtime": 171195.6903,
      "train_tokens_per_second": 4670.1
    },
    {
      "epoch": 1.7015306122448979,
      "grad_norm": 0.6744260978311201,
      "learning_rate": 5.399754013017711e-07,
      "loss": 0.24486470222473145,
      "num_input_tokens_seen": 799802056,
      "step": 13340,
      "train_runtime": 171273.9825,
      "train_tokens_per_second": 4669.723
    },
    {
      "epoch": 1.7021683673469388,
      "grad_norm": 0.6429457060308061,
      "learning_rate": 5.377134805614714e-07,
      "loss": 0.24208824634552,
      "num_input_tokens_seen": 800105448,
      "step": 13345,
      "train_runtime": 171335.2325,
      "train_tokens_per_second": 4669.824
    },
    {
      "epoch": 1.7028061224489797,
      "grad_norm": 0.6426895549519162,
      "learning_rate": 5.354560380373391e-07,
      "loss": 0.24773249626159669,
      "num_input_tokens_seen": 800402784,
      "step": 13350,
      "train_runtime": 171404.2368,
      "train_tokens_per_second": 4669.679
    },
    {
      "epoch": 1.7034438775510203,
      "grad_norm": 0.5991718681845893,
      "learning_rate": 5.33203075994877e-07,
      "loss": 0.21396217346191407,
      "num_input_tokens_seen": 800698176,
      "step": 13355,
      "train_runtime": 171463.7506,
      "train_tokens_per_second": 4669.781
    },
    {
      "epoch": 1.7040816326530612,
      "grad_norm": 0.6201191314297776,
      "learning_rate": 5.30954596695088e-07,
      "loss": 0.22633423805236816,
      "num_input_tokens_seen": 800984856,
      "step": 13360,
      "train_runtime": 171531.9766,
      "train_tokens_per_second": 4669.595
    },
    {
      "epoch": 1.7047193877551021,
      "grad_norm": 0.6999381622909089,
      "learning_rate": 5.287106023944782e-07,
      "loss": 0.23737623691558837,
      "num_input_tokens_seen": 801276848,
      "step": 13365,
      "train_runtime": 171595.5522,
      "train_tokens_per_second": 4669.567
    },
    {
      "epoch": 1.7053571428571428,
      "grad_norm": 0.6318852729096377,
      "learning_rate": 5.264710953450536e-07,
      "loss": 0.2482755422592163,
      "num_input_tokens_seen": 801596640,
      "step": 13370,
      "train_runtime": 171651.1702,
      "train_tokens_per_second": 4669.917
    },
    {
      "epoch": 1.7059948979591837,
      "grad_norm": 0.6961074086784965,
      "learning_rate": 5.242360777943151e-07,
      "loss": 0.23192543983459474,
      "num_input_tokens_seen": 801897464,
      "step": 13375,
      "train_runtime": 171724.8242,
      "train_tokens_per_second": 4669.665
    },
    {
      "epoch": 1.7066326530612246,
      "grad_norm": 0.5684368095853556,
      "learning_rate": 5.220055519852596e-07,
      "loss": 0.22409448623657227,
      "num_input_tokens_seen": 802200952,
      "step": 13380,
      "train_runtime": 171785.9245,
      "train_tokens_per_second": 4669.771
    },
    {
      "epoch": 1.7072704081632653,
      "grad_norm": 0.6515430866792863,
      "learning_rate": 5.197795201563744e-07,
      "loss": 0.2705721139907837,
      "num_input_tokens_seen": 802506256,
      "step": 13385,
      "train_runtime": 171849.8353,
      "train_tokens_per_second": 4669.811
    },
    {
      "epoch": 1.7079081632653061,
      "grad_norm": 0.6668819348988576,
      "learning_rate": 5.17557984541639e-07,
      "loss": 0.2662747621536255,
      "num_input_tokens_seen": 802805520,
      "step": 13390,
      "train_runtime": 171911.4233,
      "train_tokens_per_second": 4669.879
    },
    {
      "epoch": 1.708545918367347,
      "grad_norm": 0.6743605237110354,
      "learning_rate": 5.153409473705196e-07,
      "loss": 0.2485663652420044,
      "num_input_tokens_seen": 803103280,
      "step": 13395,
      "train_runtime": 171982.8737,
      "train_tokens_per_second": 4669.67
    },
    {
      "epoch": 1.7091836734693877,
      "grad_norm": 0.6440519264044283,
      "learning_rate": 5.131284108679669e-07,
      "loss": 0.2337883949279785,
      "num_input_tokens_seen": 803410592,
      "step": 13400,
      "train_runtime": 172056.6255,
      "train_tokens_per_second": 4669.455
    },
    {
      "epoch": 1.7098214285714286,
      "grad_norm": 0.6136284043578738,
      "learning_rate": 5.109203772544164e-07,
      "loss": 0.2468946933746338,
      "num_input_tokens_seen": 803694072,
      "step": 13405,
      "train_runtime": 172114.153,
      "train_tokens_per_second": 4669.541
    },
    {
      "epoch": 1.7104591836734695,
      "grad_norm": 0.6631202438302672,
      "learning_rate": 5.087168487457838e-07,
      "loss": 0.24713754653930664,
      "num_input_tokens_seen": 803988080,
      "step": 13410,
      "train_runtime": 172181.2472,
      "train_tokens_per_second": 4669.429
    },
    {
      "epoch": 1.7110969387755102,
      "grad_norm": 0.6270691642392973,
      "learning_rate": 5.065178275534649e-07,
      "loss": 0.24309072494506836,
      "num_input_tokens_seen": 804281616,
      "step": 13415,
      "train_runtime": 172243.0338,
      "train_tokens_per_second": 4669.458
    },
    {
      "epoch": 1.711734693877551,
      "grad_norm": 0.6303950881365012,
      "learning_rate": 5.043233158843286e-07,
      "loss": 0.21336078643798828,
      "num_input_tokens_seen": 804588296,
      "step": 13420,
      "train_runtime": 172306.2337,
      "train_tokens_per_second": 4669.525
    },
    {
      "epoch": 1.712372448979592,
      "grad_norm": 0.6823793728366182,
      "learning_rate": 5.021333159407232e-07,
      "loss": 0.21080243587493896,
      "num_input_tokens_seen": 804894048,
      "step": 13425,
      "train_runtime": 172376.5821,
      "train_tokens_per_second": 4669.393
    },
    {
      "epoch": 1.7130102040816326,
      "grad_norm": 0.6965415710532609,
      "learning_rate": 4.999478299204663e-07,
      "loss": 0.22796895503997802,
      "num_input_tokens_seen": 805181152,
      "step": 13430,
      "train_runtime": 172452.0614,
      "train_tokens_per_second": 4669.014
    },
    {
      "epoch": 1.7136479591836735,
      "grad_norm": 0.6604348579422982,
      "learning_rate": 4.977668600168428e-07,
      "loss": 0.23527843952178956,
      "num_input_tokens_seen": 805494080,
      "step": 13435,
      "train_runtime": 172501.7404,
      "train_tokens_per_second": 4669.484
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.676402075964292,
      "learning_rate": 4.95590408418613e-07,
      "loss": 0.24013857841491698,
      "num_input_tokens_seen": 805803936,
      "step": 13440,
      "train_runtime": 172567.1657,
      "train_tokens_per_second": 4669.509
    },
    {
      "epoch": 1.714923469387755,
      "grad_norm": 0.6911538717885582,
      "learning_rate": 4.934184773099943e-07,
      "loss": 0.22080633640289307,
      "num_input_tokens_seen": 806106240,
      "step": 13445,
      "train_runtime": 172632.8709,
      "train_tokens_per_second": 4669.483
    },
    {
      "epoch": 1.7155612244897958,
      "grad_norm": 0.6713450261422883,
      "learning_rate": 4.912510688706729e-07,
      "loss": 0.23345427513122557,
      "num_input_tokens_seen": 806415144,
      "step": 13450,
      "train_runtime": 172693.6075,
      "train_tokens_per_second": 4669.629
    },
    {
      "epoch": 1.7161989795918369,
      "grad_norm": 0.6415476000447592,
      "learning_rate": 4.89088185275794e-07,
      "loss": 0.2405320405960083,
      "num_input_tokens_seen": 806721576,
      "step": 13455,
      "train_runtime": 172753.864,
      "train_tokens_per_second": 4669.774
    },
    {
      "epoch": 1.7168367346938775,
      "grad_norm": 0.6394336288283835,
      "learning_rate": 4.869298286959629e-07,
      "loss": 0.24434318542480468,
      "num_input_tokens_seen": 807016592,
      "step": 13460,
      "train_runtime": 172814.3827,
      "train_tokens_per_second": 4669.846
    },
    {
      "epoch": 1.7174744897959182,
      "grad_norm": 0.7250715253066639,
      "learning_rate": 4.847760012972402e-07,
      "loss": 0.24424867630004882,
      "num_input_tokens_seen": 807311592,
      "step": 13465,
      "train_runtime": 172872.9505,
      "train_tokens_per_second": 4669.971
    },
    {
      "epoch": 1.7181122448979593,
      "grad_norm": 0.6743289428080376,
      "learning_rate": 4.826267052411432e-07,
      "loss": 0.24404535293579102,
      "num_input_tokens_seen": 807612352,
      "step": 13470,
      "train_runtime": 172930.3198,
      "train_tokens_per_second": 4670.161
    },
    {
      "epoch": 1.71875,
      "grad_norm": 0.7351586720837548,
      "learning_rate": 4.804819426846402e-07,
      "loss": 0.22565245628356934,
      "num_input_tokens_seen": 807897064,
      "step": 13475,
      "train_runtime": 172999.1838,
      "train_tokens_per_second": 4669.947
    },
    {
      "epoch": 1.7193877551020407,
      "grad_norm": 0.6678568236586356,
      "learning_rate": 4.783417157801496e-07,
      "loss": 0.21748955249786378,
      "num_input_tokens_seen": 808193664,
      "step": 13480,
      "train_runtime": 173064.7835,
      "train_tokens_per_second": 4669.891
    },
    {
      "epoch": 1.7200255102040818,
      "grad_norm": 0.6671307263692728,
      "learning_rate": 4.7620602667553973e-07,
      "loss": 0.2572862386703491,
      "num_input_tokens_seen": 808500880,
      "step": 13485,
      "train_runtime": 173120.8158,
      "train_tokens_per_second": 4670.154
    },
    {
      "epoch": 1.7206632653061225,
      "grad_norm": 0.6409595628558943,
      "learning_rate": 4.7407487751412297e-07,
      "loss": 0.22354741096496583,
      "num_input_tokens_seen": 808807400,
      "step": 13490,
      "train_runtime": 173185.3061,
      "train_tokens_per_second": 4670.185
    },
    {
      "epoch": 1.7213010204081631,
      "grad_norm": 0.6579699415119424,
      "learning_rate": 4.719482704346567e-07,
      "loss": 0.23170833587646483,
      "num_input_tokens_seen": 809101600,
      "step": 13495,
      "train_runtime": 173257.6534,
      "train_tokens_per_second": 4669.933
    },
    {
      "epoch": 1.7219387755102042,
      "grad_norm": 0.6774820763490025,
      "learning_rate": 4.6982620757133947e-07,
      "loss": 0.23648290634155272,
      "num_input_tokens_seen": 809390464,
      "step": 13500,
      "train_runtime": 173325.8921,
      "train_tokens_per_second": 4669.761
    },
    {
      "epoch": 1.722576530612245,
      "grad_norm": 0.5914618604032441,
      "learning_rate": 4.677086910538092e-07,
      "loss": 0.2374885082244873,
      "num_input_tokens_seen": 809689432,
      "step": 13505,
      "train_runtime": 173381.4046,
      "train_tokens_per_second": 4669.99
    },
    {
      "epoch": 1.7232142857142856,
      "grad_norm": 0.6726778635616372,
      "learning_rate": 4.6559572300714227e-07,
      "loss": 0.24542546272277832,
      "num_input_tokens_seen": 809969104,
      "step": 13510,
      "train_runtime": 173455.2436,
      "train_tokens_per_second": 4669.614
    },
    {
      "epoch": 1.7238520408163265,
      "grad_norm": 0.6190729925694712,
      "learning_rate": 4.6348730555185017e-07,
      "loss": 0.24277071952819823,
      "num_input_tokens_seen": 810286584,
      "step": 13515,
      "train_runtime": 173520.539,
      "train_tokens_per_second": 4669.687
    },
    {
      "epoch": 1.7244897959183674,
      "grad_norm": 0.617446990395065,
      "learning_rate": 4.6138344080387477e-07,
      "loss": 0.23009822368621827,
      "num_input_tokens_seen": 810602320,
      "step": 13520,
      "train_runtime": 173580.8412,
      "train_tokens_per_second": 4669.884
    },
    {
      "epoch": 1.725127551020408,
      "grad_norm": 0.5909057085469849,
      "learning_rate": 4.5928413087459325e-07,
      "loss": 0.23138022422790527,
      "num_input_tokens_seen": 810900880,
      "step": 13525,
      "train_runtime": 173646.1816,
      "train_tokens_per_second": 4669.846
    },
    {
      "epoch": 1.725765306122449,
      "grad_norm": 0.6990994921236008,
      "learning_rate": 4.571893778708103e-07,
      "loss": 0.24251241683959962,
      "num_input_tokens_seen": 811204872,
      "step": 13530,
      "train_runtime": 173714.0285,
      "train_tokens_per_second": 4669.772
    },
    {
      "epoch": 1.7264030612244898,
      "grad_norm": 0.691710106606899,
      "learning_rate": 4.550991838947555e-07,
      "loss": 0.2421931028366089,
      "num_input_tokens_seen": 811514968,
      "step": 13535,
      "train_runtime": 173769.9635,
      "train_tokens_per_second": 4670.053
    },
    {
      "epoch": 1.7270408163265305,
      "grad_norm": 0.6412155657066987,
      "learning_rate": 4.5301355104408364e-07,
      "loss": 0.2364500045776367,
      "num_input_tokens_seen": 811819304,
      "step": 13540,
      "train_runtime": 173834.298,
      "train_tokens_per_second": 4670.076
    },
    {
      "epoch": 1.7276785714285714,
      "grad_norm": 0.6589027866184222,
      "learning_rate": 4.509324814118754e-07,
      "loss": 0.23405904769897462,
      "num_input_tokens_seen": 812110048,
      "step": 13545,
      "train_runtime": 173896.7703,
      "train_tokens_per_second": 4670.07
    },
    {
      "epoch": 1.7283163265306123,
      "grad_norm": 0.6636270040091218,
      "learning_rate": 4.48855977086628e-07,
      "loss": 0.24407567977905273,
      "num_input_tokens_seen": 812418688,
      "step": 13550,
      "train_runtime": 173956.2348,
      "train_tokens_per_second": 4670.248
    },
    {
      "epoch": 1.728954081632653,
      "grad_norm": 0.658411946742043,
      "learning_rate": 4.4678404015225774e-07,
      "loss": 0.2599534749984741,
      "num_input_tokens_seen": 812723312,
      "step": 13555,
      "train_runtime": 174012.9141,
      "train_tokens_per_second": 4670.477
    },
    {
      "epoch": 1.7295918367346939,
      "grad_norm": 0.6139143066309292,
      "learning_rate": 4.447166726881008e-07,
      "loss": 0.23822329044342042,
      "num_input_tokens_seen": 813035912,
      "step": 13560,
      "train_runtime": 174083.1745,
      "train_tokens_per_second": 4670.388
    },
    {
      "epoch": 1.7302295918367347,
      "grad_norm": 0.6621552613210114,
      "learning_rate": 4.4265387676890203e-07,
      "loss": 0.24881305694580078,
      "num_input_tokens_seen": 813329208,
      "step": 13565,
      "train_runtime": 174154.5532,
      "train_tokens_per_second": 4670.158
    },
    {
      "epoch": 1.7308673469387754,
      "grad_norm": 0.626801875804379,
      "learning_rate": 4.4059565446482264e-07,
      "loss": 0.21699411869049073,
      "num_input_tokens_seen": 813632984,
      "step": 13570,
      "train_runtime": 174207.556,
      "train_tokens_per_second": 4670.48
    },
    {
      "epoch": 1.7315051020408163,
      "grad_norm": 0.8243709857496591,
      "learning_rate": 4.385420078414321e-07,
      "loss": 0.2210158586502075,
      "num_input_tokens_seen": 813920824,
      "step": 13575,
      "train_runtime": 174276.7198,
      "train_tokens_per_second": 4670.279
    },
    {
      "epoch": 1.7321428571428572,
      "grad_norm": 0.6718667392684093,
      "learning_rate": 4.364929389597078e-07,
      "loss": 0.24734091758728027,
      "num_input_tokens_seen": 814211728,
      "step": 13580,
      "train_runtime": 174346.4516,
      "train_tokens_per_second": 4670.079
    },
    {
      "epoch": 1.7327806122448979,
      "grad_norm": 0.7430646024505064,
      "learning_rate": 4.344484498760343e-07,
      "loss": 0.2424243211746216,
      "num_input_tokens_seen": 814535280,
      "step": 13585,
      "train_runtime": 174388.9031,
      "train_tokens_per_second": 4670.798
    },
    {
      "epoch": 1.7334183673469388,
      "grad_norm": 0.6676563012269507,
      "learning_rate": 4.324085426421981e-07,
      "loss": 0.25947110652923583,
      "num_input_tokens_seen": 814835792,
      "step": 13590,
      "train_runtime": 174457.9295,
      "train_tokens_per_second": 4670.672
    },
    {
      "epoch": 1.7340561224489797,
      "grad_norm": 0.7575342624104773,
      "learning_rate": 4.3037321930539e-07,
      "loss": 0.2515279769897461,
      "num_input_tokens_seen": 815155832,
      "step": 13595,
      "train_runtime": 174521.0114,
      "train_tokens_per_second": 4670.818
    },
    {
      "epoch": 1.7346938775510203,
      "grad_norm": 0.7319250653273267,
      "learning_rate": 4.283424819081988e-07,
      "loss": 0.24448864459991454,
      "num_input_tokens_seen": 815456800,
      "step": 13600,
      "train_runtime": 174588.2544,
      "train_tokens_per_second": 4670.743
    },
    {
      "epoch": 1.7353316326530612,
      "grad_norm": 0.6386486022394067,
      "learning_rate": 4.2631633248860937e-07,
      "loss": 0.22787261009216309,
      "num_input_tokens_seen": 815744312,
      "step": 13605,
      "train_runtime": 174664.7613,
      "train_tokens_per_second": 4670.343
    },
    {
      "epoch": 1.7359693877551021,
      "grad_norm": 0.6561912731197608,
      "learning_rate": 4.2429477308000676e-07,
      "loss": 0.24429092407226563,
      "num_input_tokens_seen": 816042632,
      "step": 13610,
      "train_runtime": 174726.5898,
      "train_tokens_per_second": 4670.398
    },
    {
      "epoch": 1.7366071428571428,
      "grad_norm": 0.7504185585120464,
      "learning_rate": 4.222778057111665e-07,
      "loss": 0.2465668201446533,
      "num_input_tokens_seen": 816330096,
      "step": 13615,
      "train_runtime": 174781.9564,
      "train_tokens_per_second": 4670.563
    },
    {
      "epoch": 1.7372448979591837,
      "grad_norm": 0.6378267197499972,
      "learning_rate": 4.202654324062544e-07,
      "loss": 0.23370888233184814,
      "num_input_tokens_seen": 816617648,
      "step": 13620,
      "train_runtime": 174851.6027,
      "train_tokens_per_second": 4670.347
    },
    {
      "epoch": 1.7378826530612246,
      "grad_norm": 0.664002130605753,
      "learning_rate": 4.182576551848283e-07,
      "loss": 0.2455160140991211,
      "num_input_tokens_seen": 816918360,
      "step": 13625,
      "train_runtime": 174911.2085,
      "train_tokens_per_second": 4670.475
    },
    {
      "epoch": 1.7385204081632653,
      "grad_norm": 0.61713932962022,
      "learning_rate": 4.1625447606183533e-07,
      "loss": 0.24306926727294922,
      "num_input_tokens_seen": 817224032,
      "step": 13630,
      "train_runtime": 174969.0144,
      "train_tokens_per_second": 4670.679
    },
    {
      "epoch": 1.7391581632653061,
      "grad_norm": 0.6178408609649475,
      "learning_rate": 4.142558970476024e-07,
      "loss": 0.22744331359863282,
      "num_input_tokens_seen": 817531176,
      "step": 13635,
      "train_runtime": 175035.7027,
      "train_tokens_per_second": 4670.654
    },
    {
      "epoch": 1.739795918367347,
      "grad_norm": 0.6692420345619271,
      "learning_rate": 4.1226192014784393e-07,
      "loss": 0.2176975965499878,
      "num_input_tokens_seen": 817805400,
      "step": 13640,
      "train_runtime": 175109.2489,
      "train_tokens_per_second": 4670.258
    },
    {
      "epoch": 1.7404336734693877,
      "grad_norm": 0.6272228440769623,
      "learning_rate": 4.102725473636571e-07,
      "loss": 0.25208315849304197,
      "num_input_tokens_seen": 818110408,
      "step": 13645,
      "train_runtime": 175173.8421,
      "train_tokens_per_second": 4670.277
    },
    {
      "epoch": 1.7410714285714286,
      "grad_norm": 0.6756531091745722,
      "learning_rate": 4.0828778069151363e-07,
      "loss": 0.24423251152038575,
      "num_input_tokens_seen": 818422504,
      "step": 13650,
      "train_runtime": 175231.7334,
      "train_tokens_per_second": 4670.515
    },
    {
      "epoch": 1.7417091836734695,
      "grad_norm": 0.7468077834189175,
      "learning_rate": 4.063076221232676e-07,
      "loss": 0.25256857872009275,
      "num_input_tokens_seen": 818719216,
      "step": 13655,
      "train_runtime": 175295.187,
      "train_tokens_per_second": 4670.517
    },
    {
      "epoch": 1.7423469387755102,
      "grad_norm": 0.6249208640493598,
      "learning_rate": 4.043320736461448e-07,
      "loss": 0.25737662315368653,
      "num_input_tokens_seen": 819022224,
      "step": 13660,
      "train_runtime": 175365.3005,
      "train_tokens_per_second": 4670.378
    },
    {
      "epoch": 1.742984693877551,
      "grad_norm": 0.6040912449091471,
      "learning_rate": 4.0236113724274716e-07,
      "loss": 0.24058666229248046,
      "num_input_tokens_seen": 819315192,
      "step": 13665,
      "train_runtime": 175434.7397,
      "train_tokens_per_second": 4670.199
    },
    {
      "epoch": 1.743622448979592,
      "grad_norm": 0.7164460139205696,
      "learning_rate": 4.0039481489104636e-07,
      "loss": 0.2532172203063965,
      "num_input_tokens_seen": 819627288,
      "step": 13670,
      "train_runtime": 175494.5877,
      "train_tokens_per_second": 4670.385
    },
    {
      "epoch": 1.7442602040816326,
      "grad_norm": 0.7226232309413686,
      "learning_rate": 3.98433108564385e-07,
      "loss": 0.2506558418273926,
      "num_input_tokens_seen": 819937352,
      "step": 13675,
      "train_runtime": 175555.4148,
      "train_tokens_per_second": 4670.533
    },
    {
      "epoch": 1.7448979591836735,
      "grad_norm": 0.7186342580135948,
      "learning_rate": 3.964760202314716e-07,
      "loss": 0.256072473526001,
      "num_input_tokens_seen": 820240232,
      "step": 13680,
      "train_runtime": 175632.523,
      "train_tokens_per_second": 4670.207
    },
    {
      "epoch": 1.7455357142857144,
      "grad_norm": 0.6704285693346819,
      "learning_rate": 3.9452355185638226e-07,
      "loss": 0.25684480667114257,
      "num_input_tokens_seen": 820547016,
      "step": 13685,
      "train_runtime": 175699.6846,
      "train_tokens_per_second": 4670.168
    },
    {
      "epoch": 1.746173469387755,
      "grad_norm": 0.6074480140719041,
      "learning_rate": 3.9257570539855237e-07,
      "loss": 0.23230316638946533,
      "num_input_tokens_seen": 820846064,
      "step": 13690,
      "train_runtime": 175772.8875,
      "train_tokens_per_second": 4669.924
    },
    {
      "epoch": 1.7468112244897958,
      "grad_norm": 0.6260795731147497,
      "learning_rate": 3.9063248281278555e-07,
      "loss": 0.2343238115310669,
      "num_input_tokens_seen": 821136984,
      "step": 13695,
      "train_runtime": 175842.1621,
      "train_tokens_per_second": 4669.739
    },
    {
      "epoch": 1.7474489795918369,
      "grad_norm": 0.6832923954708081,
      "learning_rate": 3.8869388604924007e-07,
      "loss": 0.2325591564178467,
      "num_input_tokens_seen": 821444248,
      "step": 13700,
      "train_runtime": 175904.4337,
      "train_tokens_per_second": 4669.833
    },
    {
      "epoch": 1.7480867346938775,
      "grad_norm": 0.6382970626342359,
      "learning_rate": 3.8675991705343197e-07,
      "loss": 0.24201679229736328,
      "num_input_tokens_seen": 821748976,
      "step": 13705,
      "train_runtime": 175969.7855,
      "train_tokens_per_second": 4669.83
    },
    {
      "epoch": 1.7487244897959182,
      "grad_norm": 0.587902063077867,
      "learning_rate": 3.848305777662342e-07,
      "loss": 0.2240581512451172,
      "num_input_tokens_seen": 822049696,
      "step": 13710,
      "train_runtime": 176041.5452,
      "train_tokens_per_second": 4669.635
    },
    {
      "epoch": 1.7493622448979593,
      "grad_norm": 0.6280674600029451,
      "learning_rate": 3.829058701238764e-07,
      "loss": 0.23671512603759765,
      "num_input_tokens_seen": 822348152,
      "step": 13715,
      "train_runtime": 176103.2596,
      "train_tokens_per_second": 4669.693
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.690888298586804,
      "learning_rate": 3.8098579605793385e-07,
      "loss": 0.22222754955291749,
      "num_input_tokens_seen": 822648608,
      "step": 13720,
      "train_runtime": 176160.0031,
      "train_tokens_per_second": 4669.894
    },
    {
      "epoch": 1.7506377551020407,
      "grad_norm": 0.5888112153098812,
      "learning_rate": 3.790703574953353e-07,
      "loss": 0.21944701671600342,
      "num_input_tokens_seen": 822950904,
      "step": 13725,
      "train_runtime": 176223.4126,
      "train_tokens_per_second": 4669.929
    },
    {
      "epoch": 1.7512755102040818,
      "grad_norm": 0.6151321877697761,
      "learning_rate": 3.771595563583602e-07,
      "loss": 0.23566222190856934,
      "num_input_tokens_seen": 823259904,
      "step": 13730,
      "train_runtime": 176288.164,
      "train_tokens_per_second": 4669.967
    },
    {
      "epoch": 1.7519132653061225,
      "grad_norm": 0.6655558670013475,
      "learning_rate": 3.752533945646275e-07,
      "loss": 0.25287594795227053,
      "num_input_tokens_seen": 823565512,
      "step": 13735,
      "train_runtime": 176358.7721,
      "train_tokens_per_second": 4669.83
    },
    {
      "epoch": 1.7525510204081631,
      "grad_norm": 0.6508335845939857,
      "learning_rate": 3.733518740271047e-07,
      "loss": 0.2242781400680542,
      "num_input_tokens_seen": 823859624,
      "step": 13740,
      "train_runtime": 176428.4289,
      "train_tokens_per_second": 4669.653
    },
    {
      "epoch": 1.7531887755102042,
      "grad_norm": 0.6079860747794961,
      "learning_rate": 3.7145499665410147e-07,
      "loss": 0.23161578178405762,
      "num_input_tokens_seen": 824173752,
      "step": 13745,
      "train_runtime": 176495.3122,
      "train_tokens_per_second": 4669.664
    },
    {
      "epoch": 1.753826530612245,
      "grad_norm": 0.6504089562154073,
      "learning_rate": 3.695627643492655e-07,
      "loss": 0.24319536685943605,
      "num_input_tokens_seen": 824475376,
      "step": 13750,
      "train_runtime": 176557.0608,
      "train_tokens_per_second": 4669.739
    },
    {
      "epoch": 1.7544642857142856,
      "grad_norm": 0.6482203781267267,
      "learning_rate": 3.6767517901158455e-07,
      "loss": 0.23995392322540282,
      "num_input_tokens_seen": 824767584,
      "step": 13755,
      "train_runtime": 176626.9062,
      "train_tokens_per_second": 4669.547
    },
    {
      "epoch": 1.7551020408163265,
      "grad_norm": 0.6017165332799732,
      "learning_rate": 3.6579224253538213e-07,
      "loss": 0.2299355983734131,
      "num_input_tokens_seen": 825079160,
      "step": 13760,
      "train_runtime": 176691.7999,
      "train_tokens_per_second": 4669.595
    },
    {
      "epoch": 1.7557397959183674,
      "grad_norm": 0.5771062782445211,
      "learning_rate": 3.639139568103173e-07,
      "loss": 0.2594435453414917,
      "num_input_tokens_seen": 825395496,
      "step": 13765,
      "train_runtime": 176754.7416,
      "train_tokens_per_second": 4669.722
    },
    {
      "epoch": 1.756377551020408,
      "grad_norm": 0.6260698337351052,
      "learning_rate": 3.620403237213799e-07,
      "loss": 0.22344212532043456,
      "num_input_tokens_seen": 825696584,
      "step": 13770,
      "train_runtime": 176816.8389,
      "train_tokens_per_second": 4669.785
    },
    {
      "epoch": 1.757015306122449,
      "grad_norm": 0.8303140631169927,
      "learning_rate": 3.6017134514889196e-07,
      "loss": 0.22419793605804444,
      "num_input_tokens_seen": 825994344,
      "step": 13775,
      "train_runtime": 176876.8366,
      "train_tokens_per_second": 4669.884
    },
    {
      "epoch": 1.7576530612244898,
      "grad_norm": 0.6002592186980982,
      "learning_rate": 3.583070229685043e-07,
      "loss": 0.24105172157287597,
      "num_input_tokens_seen": 826291616,
      "step": 13780,
      "train_runtime": 176950.403,
      "train_tokens_per_second": 4669.623
    },
    {
      "epoch": 1.7582908163265305,
      "grad_norm": 0.615381677018483,
      "learning_rate": 3.564473590511941e-07,
      "loss": 0.2350989818572998,
      "num_input_tokens_seen": 826588040,
      "step": 13785,
      "train_runtime": 177000.2875,
      "train_tokens_per_second": 4669.981
    },
    {
      "epoch": 1.7589285714285714,
      "grad_norm": 0.6272890183125568,
      "learning_rate": 3.5459235526326464e-07,
      "loss": 0.21449654102325438,
      "num_input_tokens_seen": 826881872,
      "step": 13790,
      "train_runtime": 177063.6363,
      "train_tokens_per_second": 4669.97
    },
    {
      "epoch": 1.7595663265306123,
      "grad_norm": 0.5824579104221645,
      "learning_rate": 3.527420134663401e-07,
      "loss": 0.23701989650726318,
      "num_input_tokens_seen": 827181464,
      "step": 13795,
      "train_runtime": 177124.007,
      "train_tokens_per_second": 4670.07
    },
    {
      "epoch": 1.760204081632653,
      "grad_norm": 0.653381271184272,
      "learning_rate": 3.5089633551736946e-07,
      "loss": 0.24289774894714355,
      "num_input_tokens_seen": 827479144,
      "step": 13800,
      "train_runtime": 177177.2529,
      "train_tokens_per_second": 4670.346
    },
    {
      "epoch": 1.7608418367346939,
      "grad_norm": 0.6148624342498099,
      "learning_rate": 3.4905532326861946e-07,
      "loss": 0.23552067279815675,
      "num_input_tokens_seen": 827784936,
      "step": 13805,
      "train_runtime": 177233.5487,
      "train_tokens_per_second": 4670.588
    },
    {
      "epoch": 1.7614795918367347,
      "grad_norm": 0.5299110104184739,
      "learning_rate": 3.4721897856767275e-07,
      "loss": 0.21198663711547852,
      "num_input_tokens_seen": 828096552,
      "step": 13810,
      "train_runtime": 177287.588,
      "train_tokens_per_second": 4670.922
    },
    {
      "epoch": 1.7621173469387754,
      "grad_norm": 0.6683036540754396,
      "learning_rate": 3.4538730325743174e-07,
      "loss": 0.2555697917938232,
      "num_input_tokens_seen": 828385320,
      "step": 13815,
      "train_runtime": 177353.237,
      "train_tokens_per_second": 4670.822
    },
    {
      "epoch": 1.7627551020408163,
      "grad_norm": 0.685457457591933,
      "learning_rate": 3.4356029917610833e-07,
      "loss": 0.23040683269500734,
      "num_input_tokens_seen": 828679568,
      "step": 13820,
      "train_runtime": 177427.6049,
      "train_tokens_per_second": 4670.522
    },
    {
      "epoch": 1.7633928571428572,
      "grad_norm": 0.6940270902195277,
      "learning_rate": 3.417379681572297e-07,
      "loss": 0.25145397186279295,
      "num_input_tokens_seen": 828985408,
      "step": 13825,
      "train_runtime": 177501.0429,
      "train_tokens_per_second": 4670.313
    },
    {
      "epoch": 1.7640306122448979,
      "grad_norm": 0.7088989504768888,
      "learning_rate": 3.3992031202963293e-07,
      "loss": 0.23242881298065185,
      "num_input_tokens_seen": 829281192,
      "step": 13830,
      "train_runtime": 177561.6522,
      "train_tokens_per_second": 4670.385
    },
    {
      "epoch": 1.7646683673469388,
      "grad_norm": 0.6179773270375348,
      "learning_rate": 3.3810733261746167e-07,
      "loss": 0.2483144760131836,
      "num_input_tokens_seen": 829588008,
      "step": 13835,
      "train_runtime": 177620.6994,
      "train_tokens_per_second": 4670.559
    },
    {
      "epoch": 1.7653061224489797,
      "grad_norm": 0.6353823748269662,
      "learning_rate": 3.362990317401688e-07,
      "loss": 0.23761091232299805,
      "num_input_tokens_seen": 829874728,
      "step": 13840,
      "train_runtime": 177686.5023,
      "train_tokens_per_second": 4670.443
    },
    {
      "epoch": 1.7659438775510203,
      "grad_norm": 0.6134309097723778,
      "learning_rate": 3.344954112125087e-07,
      "loss": 0.24033880233764648,
      "num_input_tokens_seen": 830169128,
      "step": 13845,
      "train_runtime": 177760.5155,
      "train_tokens_per_second": 4670.155
    },
    {
      "epoch": 1.7665816326530612,
      "grad_norm": 0.6842351383304307,
      "learning_rate": 3.3269647284454253e-07,
      "loss": 0.23505787849426268,
      "num_input_tokens_seen": 830469976,
      "step": 13850,
      "train_runtime": 177832.828,
      "train_tokens_per_second": 4669.948
    },
    {
      "epoch": 1.7672193877551021,
      "grad_norm": 0.6362356413450317,
      "learning_rate": 3.309022184416288e-07,
      "loss": 0.2307974576950073,
      "num_input_tokens_seen": 830765392,
      "step": 13855,
      "train_runtime": 177886.9152,
      "train_tokens_per_second": 4670.188
    },
    {
      "epoch": 1.7678571428571428,
      "grad_norm": 0.7063557852619481,
      "learning_rate": 3.2911264980442834e-07,
      "loss": 0.24306538105010986,
      "num_input_tokens_seen": 831071568,
      "step": 13860,
      "train_runtime": 177949.932,
      "train_tokens_per_second": 4670.255
    },
    {
      "epoch": 1.7684948979591837,
      "grad_norm": 0.661390891296154,
      "learning_rate": 3.273277687288978e-07,
      "loss": 0.23237895965576172,
      "num_input_tokens_seen": 831369760,
      "step": 13865,
      "train_runtime": 178016.5951,
      "train_tokens_per_second": 4670.181
    },
    {
      "epoch": 1.7691326530612246,
      "grad_norm": 0.6898088178147898,
      "learning_rate": 3.255475770062899e-07,
      "loss": 0.26753573417663573,
      "num_input_tokens_seen": 831662672,
      "step": 13870,
      "train_runtime": 178085.7383,
      "train_tokens_per_second": 4670.013
    },
    {
      "epoch": 1.7697704081632653,
      "grad_norm": 0.6394911350645628,
      "learning_rate": 3.237720764231517e-07,
      "loss": 0.2302546501159668,
      "num_input_tokens_seen": 831963664,
      "step": 13875,
      "train_runtime": 178144.9731,
      "train_tokens_per_second": 4670.15
    },
    {
      "epoch": 1.7704081632653061,
      "grad_norm": 0.6939548195274975,
      "learning_rate": 3.220012687613222e-07,
      "loss": 0.25703885555267336,
      "num_input_tokens_seen": 832276272,
      "step": 13880,
      "train_runtime": 178205.6694,
      "train_tokens_per_second": 4670.313
    },
    {
      "epoch": 1.771045918367347,
      "grad_norm": 0.5628021167165467,
      "learning_rate": 3.2023515579793e-07,
      "loss": 0.24952988624572753,
      "num_input_tokens_seen": 832580640,
      "step": 13885,
      "train_runtime": 178260.5732,
      "train_tokens_per_second": 4670.582
    },
    {
      "epoch": 1.7716836734693877,
      "grad_norm": 0.8021529602282467,
      "learning_rate": 3.1847373930539326e-07,
      "loss": 0.2512929677963257,
      "num_input_tokens_seen": 832880288,
      "step": 13890,
      "train_runtime": 178326.1198,
      "train_tokens_per_second": 4670.546
    },
    {
      "epoch": 1.7723214285714286,
      "grad_norm": 0.7091388799731951,
      "learning_rate": 3.167170210514153e-07,
      "loss": 0.237908935546875,
      "num_input_tokens_seen": 833177360,
      "step": 13895,
      "train_runtime": 178381.3188,
      "train_tokens_per_second": 4670.766
    },
    {
      "epoch": 1.7729591836734695,
      "grad_norm": 0.6933618943436181,
      "learning_rate": 3.14965002798987e-07,
      "loss": 0.25426063537597654,
      "num_input_tokens_seen": 833481240,
      "step": 13900,
      "train_runtime": 178448.9426,
      "train_tokens_per_second": 4670.699
    },
    {
      "epoch": 1.7735969387755102,
      "grad_norm": 0.6825957096513657,
      "learning_rate": 3.1321768630638073e-07,
      "loss": 0.24481732845306398,
      "num_input_tokens_seen": 833789864,
      "step": 13905,
      "train_runtime": 178518.8869,
      "train_tokens_per_second": 4670.597
    },
    {
      "epoch": 1.774234693877551,
      "grad_norm": 0.7054173966164635,
      "learning_rate": 3.1147507332715044e-07,
      "loss": 0.2531277656555176,
      "num_input_tokens_seen": 834104624,
      "step": 13910,
      "train_runtime": 178586.8554,
      "train_tokens_per_second": 4670.582
    },
    {
      "epoch": 1.774872448979592,
      "grad_norm": 0.6597772456241627,
      "learning_rate": 3.097371656101289e-07,
      "loss": 0.2587900638580322,
      "num_input_tokens_seen": 834408072,
      "step": 13915,
      "train_runtime": 178654.1547,
      "train_tokens_per_second": 4670.522
    },
    {
      "epoch": 1.7755102040816326,
      "grad_norm": 0.6327697397211675,
      "learning_rate": 3.080039648994304e-07,
      "loss": 0.2404341220855713,
      "num_input_tokens_seen": 834706232,
      "step": 13920,
      "train_runtime": 178729.9973,
      "train_tokens_per_second": 4670.208
    },
    {
      "epoch": 1.7761479591836735,
      "grad_norm": 0.7293595138469602,
      "learning_rate": 3.062754729344408e-07,
      "loss": 0.25003910064697266,
      "num_input_tokens_seen": 835009240,
      "step": 13925,
      "train_runtime": 178791.3031,
      "train_tokens_per_second": 4670.301
    },
    {
      "epoch": 1.7767857142857144,
      "grad_norm": 0.6663773758034158,
      "learning_rate": 3.0455169144982257e-07,
      "loss": 0.24301655292510987,
      "num_input_tokens_seen": 835304568,
      "step": 13930,
      "train_runtime": 178858.2634,
      "train_tokens_per_second": 4670.204
    },
    {
      "epoch": 1.777423469387755,
      "grad_norm": 0.6164569733038848,
      "learning_rate": 3.028326221755118e-07,
      "loss": 0.23880252838134766,
      "num_input_tokens_seen": 835604696,
      "step": 13935,
      "train_runtime": 178913.5245,
      "train_tokens_per_second": 4670.439
    },
    {
      "epoch": 1.7780612244897958,
      "grad_norm": 0.6310932598969143,
      "learning_rate": 3.0111826683671407e-07,
      "loss": 0.25446553230285646,
      "num_input_tokens_seen": 835900296,
      "step": 13940,
      "train_runtime": 178984.3912,
      "train_tokens_per_second": 4670.241
    },
    {
      "epoch": 1.7786989795918369,
      "grad_norm": 0.6639849717106197,
      "learning_rate": 2.9940862715390483e-07,
      "loss": 0.24060540199279784,
      "num_input_tokens_seen": 836194792,
      "step": 13945,
      "train_runtime": 179051.944,
      "train_tokens_per_second": 4670.124
    },
    {
      "epoch": 1.7793367346938775,
      "grad_norm": 0.6559897532506478,
      "learning_rate": 2.9770370484282665e-07,
      "loss": 0.23808369636535645,
      "num_input_tokens_seen": 836496792,
      "step": 13950,
      "train_runtime": 179121.316,
      "train_tokens_per_second": 4670.001
    },
    {
      "epoch": 1.7799744897959182,
      "grad_norm": 0.6383602673957005,
      "learning_rate": 2.960035016144885e-07,
      "loss": 0.22525038719177246,
      "num_input_tokens_seen": 836782688,
      "step": 13955,
      "train_runtime": 179198.633,
      "train_tokens_per_second": 4669.582
    },
    {
      "epoch": 1.7806122448979593,
      "grad_norm": 0.7000522726804709,
      "learning_rate": 2.943080191751629e-07,
      "loss": 0.25488481521606443,
      "num_input_tokens_seen": 837094584,
      "step": 13960,
      "train_runtime": 179263.3355,
      "train_tokens_per_second": 4669.636
    },
    {
      "epoch": 1.78125,
      "grad_norm": 0.6994596296350564,
      "learning_rate": 2.9261725922638485e-07,
      "loss": 0.2449193477630615,
      "num_input_tokens_seen": 837392928,
      "step": 13965,
      "train_runtime": 179328.2157,
      "train_tokens_per_second": 4669.611
    },
    {
      "epoch": 1.7818877551020407,
      "grad_norm": 0.6513016169940382,
      "learning_rate": 2.9093122346494974e-07,
      "loss": 0.25704920291900635,
      "num_input_tokens_seen": 837679544,
      "step": 13970,
      "train_runtime": 179384.5775,
      "train_tokens_per_second": 4669.741
    },
    {
      "epoch": 1.7825255102040818,
      "grad_norm": 0.6833897369640713,
      "learning_rate": 2.8924991358291354e-07,
      "loss": 0.24814114570617676,
      "num_input_tokens_seen": 837972976,
      "step": 13975,
      "train_runtime": 179459.6037,
      "train_tokens_per_second": 4669.424
    },
    {
      "epoch": 1.7831632653061225,
      "grad_norm": 0.6194541374721899,
      "learning_rate": 2.875733312675855e-07,
      "loss": 0.23839044570922852,
      "num_input_tokens_seen": 838284000,
      "step": 13980,
      "train_runtime": 179531.3583,
      "train_tokens_per_second": 4669.29
    },
    {
      "epoch": 1.7838010204081631,
      "grad_norm": 0.6398052840816332,
      "learning_rate": 2.8590147820153513e-07,
      "loss": 0.2350918769836426,
      "num_input_tokens_seen": 838588496,
      "step": 13985,
      "train_runtime": 179586.2281,
      "train_tokens_per_second": 4669.559
    },
    {
      "epoch": 1.7844387755102042,
      "grad_norm": 0.6096802222597947,
      "learning_rate": 2.8423435606258485e-07,
      "loss": 0.24754149913787843,
      "num_input_tokens_seen": 838896680,
      "step": 13990,
      "train_runtime": 179654.0069,
      "train_tokens_per_second": 4669.513
    },
    {
      "epoch": 1.785076530612245,
      "grad_norm": 0.5870268496149098,
      "learning_rate": 2.8257196652380523e-07,
      "loss": 0.23757634162902833,
      "num_input_tokens_seen": 839199288,
      "step": 13995,
      "train_runtime": 179713.1109,
      "train_tokens_per_second": 4669.661
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 0.5922973135813985,
      "learning_rate": 2.809143112535212e-07,
      "loss": 0.20630648136138915,
      "num_input_tokens_seen": 839485064,
      "step": 14000,
      "train_runtime": 179784.4954,
      "train_tokens_per_second": 4669.396
    },
    {
      "epoch": 1.7863520408163265,
      "grad_norm": 0.6555882493386753,
      "learning_rate": 2.7926139191530697e-07,
      "loss": 0.23436861038208007,
      "num_input_tokens_seen": 839784112,
      "step": 14005,
      "train_runtime": 179852.9178,
      "train_tokens_per_second": 4669.283
    },
    {
      "epoch": 1.7869897959183674,
      "grad_norm": 0.7213819487431138,
      "learning_rate": 2.7761321016798183e-07,
      "loss": 0.2294372797012329,
      "num_input_tokens_seen": 840085248,
      "step": 14010,
      "train_runtime": 179919.0506,
      "train_tokens_per_second": 4669.24
    },
    {
      "epoch": 1.787627551020408,
      "grad_norm": 0.5780708327960133,
      "learning_rate": 2.7596976766560977e-07,
      "loss": 0.23864274024963378,
      "num_input_tokens_seen": 840389216,
      "step": 14015,
      "train_runtime": 179981.9428,
      "train_tokens_per_second": 4669.297
    },
    {
      "epoch": 1.788265306122449,
      "grad_norm": 0.6267420865230794,
      "learning_rate": 2.743310660575027e-07,
      "loss": 0.2206709861755371,
      "num_input_tokens_seen": 840684392,
      "step": 14020,
      "train_runtime": 180057.9611,
      "train_tokens_per_second": 4668.965
    },
    {
      "epoch": 1.7889030612244898,
      "grad_norm": 0.6955593023546806,
      "learning_rate": 2.726971069882101e-07,
      "loss": 0.23168370723724366,
      "num_input_tokens_seen": 840974992,
      "step": 14025,
      "train_runtime": 180132.6187,
      "train_tokens_per_second": 4668.644
    },
    {
      "epoch": 1.7895408163265305,
      "grad_norm": 0.5605018184777858,
      "learning_rate": 2.7106789209752513e-07,
      "loss": 0.224169921875,
      "num_input_tokens_seen": 841289592,
      "step": 14030,
      "train_runtime": 180180.4933,
      "train_tokens_per_second": 4669.149
    },
    {
      "epoch": 1.7901785714285714,
      "grad_norm": 0.6648559866948053,
      "learning_rate": 2.6944342302047785e-07,
      "loss": 0.2617345809936523,
      "num_input_tokens_seen": 841593808,
      "step": 14035,
      "train_runtime": 180239.2903,
      "train_tokens_per_second": 4669.314
    },
    {
      "epoch": 1.7908163265306123,
      "grad_norm": 0.6231065297804759,
      "learning_rate": 2.678237013873375e-07,
      "loss": 0.25885226726531985,
      "num_input_tokens_seen": 841895528,
      "step": 14040,
      "train_runtime": 180306.8164,
      "train_tokens_per_second": 4669.238
    },
    {
      "epoch": 1.791454081632653,
      "grad_norm": 0.6317211141544594,
      "learning_rate": 2.66208728823607e-07,
      "loss": 0.22434628009796143,
      "num_input_tokens_seen": 842191312,
      "step": 14045,
      "train_runtime": 180377.3203,
      "train_tokens_per_second": 4669.053
    },
    {
      "epoch": 1.7920918367346939,
      "grad_norm": 0.6571417556533412,
      "learning_rate": 2.6459850695002454e-07,
      "loss": 0.24795680046081542,
      "num_input_tokens_seen": 842496784,
      "step": 14050,
      "train_runtime": 180431.2072,
      "train_tokens_per_second": 4669.352
    },
    {
      "epoch": 1.7927295918367347,
      "grad_norm": 0.6598004061013982,
      "learning_rate": 2.629930373825595e-07,
      "loss": 0.22257847785949708,
      "num_input_tokens_seen": 842798288,
      "step": 14055,
      "train_runtime": 180488.3174,
      "train_tokens_per_second": 4669.545
    },
    {
      "epoch": 1.7933673469387754,
      "grad_norm": 0.6637345292647696,
      "learning_rate": 2.6139232173241434e-07,
      "loss": 0.2368372917175293,
      "num_input_tokens_seen": 843094680,
      "step": 14060,
      "train_runtime": 180550.121,
      "train_tokens_per_second": 4669.588
    },
    {
      "epoch": 1.7940051020408163,
      "grad_norm": 0.7038406559286605,
      "learning_rate": 2.5979636160601673e-07,
      "loss": 0.24609622955322266,
      "num_input_tokens_seen": 843407144,
      "step": 14065,
      "train_runtime": 180614.613,
      "train_tokens_per_second": 4669.651
    },
    {
      "epoch": 1.7946428571428572,
      "grad_norm": 0.6291640765271659,
      "learning_rate": 2.5820515860502604e-07,
      "loss": 0.2567103862762451,
      "num_input_tokens_seen": 843714272,
      "step": 14070,
      "train_runtime": 180676.5227,
      "train_tokens_per_second": 4669.75
    },
    {
      "epoch": 1.7952806122448979,
      "grad_norm": 0.6097260626734088,
      "learning_rate": 2.5661871432632545e-07,
      "loss": 0.2330172300338745,
      "num_input_tokens_seen": 844010616,
      "step": 14075,
      "train_runtime": 180740.3,
      "train_tokens_per_second": 4669.742
    },
    {
      "epoch": 1.7959183673469388,
      "grad_norm": 0.6801791912877365,
      "learning_rate": 2.550370303620214e-07,
      "loss": 0.2431706428527832,
      "num_input_tokens_seen": 844313000,
      "step": 14080,
      "train_runtime": 180805.2674,
      "train_tokens_per_second": 4669.737
    },
    {
      "epoch": 1.7965561224489797,
      "grad_norm": 0.5757820793193561,
      "learning_rate": 2.534601082994437e-07,
      "loss": 0.232086181640625,
      "num_input_tokens_seen": 844609600,
      "step": 14085,
      "train_runtime": 180881.9251,
      "train_tokens_per_second": 4669.397
    },
    {
      "epoch": 1.7971938775510203,
      "grad_norm": 0.6941251008293757,
      "learning_rate": 2.5188794972114684e-07,
      "loss": 0.24894967079162597,
      "num_input_tokens_seen": 844903104,
      "step": 14090,
      "train_runtime": 180941.835,
      "train_tokens_per_second": 4669.474
    },
    {
      "epoch": 1.7978316326530612,
      "grad_norm": 0.6892429730086918,
      "learning_rate": 2.503205562048994e-07,
      "loss": 0.23433713912963866,
      "num_input_tokens_seen": 845203632,
      "step": 14095,
      "train_runtime": 181000.9271,
      "train_tokens_per_second": 4669.609
    },
    {
      "epoch": 1.7984693877551021,
      "grad_norm": 0.648582833181612,
      "learning_rate": 2.487579293236897e-07,
      "loss": 0.24241786003112792,
      "num_input_tokens_seen": 845505776,
      "step": 14100,
      "train_runtime": 181065.5296,
      "train_tokens_per_second": 4669.612
    },
    {
      "epoch": 1.7991071428571428,
      "grad_norm": 0.6444873623587022,
      "learning_rate": 2.4720007064572506e-07,
      "loss": 0.24049158096313478,
      "num_input_tokens_seen": 845817520,
      "step": 14105,
      "train_runtime": 181136.9664,
      "train_tokens_per_second": 4669.491
    },
    {
      "epoch": 1.7997448979591837,
      "grad_norm": 0.6593792460663095,
      "learning_rate": 2.456469817344237e-07,
      "loss": 0.22638745307922364,
      "num_input_tokens_seen": 846138344,
      "step": 14110,
      "train_runtime": 181194.4118,
      "train_tokens_per_second": 4669.782
    },
    {
      "epoch": 1.8003826530612246,
      "grad_norm": 0.6286204175192706,
      "learning_rate": 2.4409866414841996e-07,
      "loss": 0.23471994400024415,
      "num_input_tokens_seen": 846438264,
      "step": 14115,
      "train_runtime": 181252.4813,
      "train_tokens_per_second": 4669.94
    },
    {
      "epoch": 1.8010204081632653,
      "grad_norm": 0.6523694187565792,
      "learning_rate": 2.4255511944155764e-07,
      "loss": 0.22545332908630372,
      "num_input_tokens_seen": 846733264,
      "step": 14120,
      "train_runtime": 181312.7442,
      "train_tokens_per_second": 4670.015
    },
    {
      "epoch": 1.8016581632653061,
      "grad_norm": 0.6733252335596946,
      "learning_rate": 2.410163491628925e-07,
      "loss": 0.21708171367645263,
      "num_input_tokens_seen": 847020656,
      "step": 14125,
      "train_runtime": 181377.9261,
      "train_tokens_per_second": 4669.921
    },
    {
      "epoch": 1.802295918367347,
      "grad_norm": 0.6641541846371164,
      "learning_rate": 2.394823548566877e-07,
      "loss": 0.25301213264465333,
      "num_input_tokens_seen": 847318328,
      "step": 14130,
      "train_runtime": 181443.0402,
      "train_tokens_per_second": 4669.886
    },
    {
      "epoch": 1.8029336734693877,
      "grad_norm": 0.6195771371319925,
      "learning_rate": 2.3795313806241337e-07,
      "loss": 0.2453226327896118,
      "num_input_tokens_seen": 847630544,
      "step": 14135,
      "train_runtime": 181501.4745,
      "train_tokens_per_second": 4670.103
    },
    {
      "epoch": 1.8035714285714286,
      "grad_norm": 0.5990228308868839,
      "learning_rate": 2.364287003147453e-07,
      "loss": 0.2190986156463623,
      "num_input_tokens_seen": 847944464,
      "step": 14140,
      "train_runtime": 181566.3767,
      "train_tokens_per_second": 4670.162
    },
    {
      "epoch": 1.8042091836734695,
      "grad_norm": 0.6912023506330887,
      "learning_rate": 2.3490904314356412e-07,
      "loss": 0.23070981502532958,
      "num_input_tokens_seen": 848223480,
      "step": 14145,
      "train_runtime": 181635.1823,
      "train_tokens_per_second": 4669.929
    },
    {
      "epoch": 1.8048469387755102,
      "grad_norm": 0.6258088067594457,
      "learning_rate": 2.3339416807395154e-07,
      "loss": 0.23261451721191406,
      "num_input_tokens_seen": 848527256,
      "step": 14150,
      "train_runtime": 181700.8442,
      "train_tokens_per_second": 4669.914
    },
    {
      "epoch": 1.805484693877551,
      "grad_norm": 0.6717797485189194,
      "learning_rate": 2.3188407662618983e-07,
      "loss": 0.23456430435180664,
      "num_input_tokens_seen": 848813024,
      "step": 14155,
      "train_runtime": 181766.3603,
      "train_tokens_per_second": 4669.803
    },
    {
      "epoch": 1.806122448979592,
      "grad_norm": 0.5587247419594303,
      "learning_rate": 2.3037877031576172e-07,
      "loss": 0.2323758602142334,
      "num_input_tokens_seen": 849121992,
      "step": 14160,
      "train_runtime": 181827.2831,
      "train_tokens_per_second": 4669.937
    },
    {
      "epoch": 1.8067602040816326,
      "grad_norm": 0.6967048187927307,
      "learning_rate": 2.2887825065334822e-07,
      "loss": 0.22853446006774902,
      "num_input_tokens_seen": 849428488,
      "step": 14165,
      "train_runtime": 181903.0928,
      "train_tokens_per_second": 4669.676
    },
    {
      "epoch": 1.8073979591836735,
      "grad_norm": 0.7265755754020878,
      "learning_rate": 2.2738251914482312e-07,
      "loss": 0.24710376262664796,
      "num_input_tokens_seen": 849726160,
      "step": 14170,
      "train_runtime": 181962.3829,
      "train_tokens_per_second": 4669.79
    },
    {
      "epoch": 1.8080357142857144,
      "grad_norm": 0.6767356092577087,
      "learning_rate": 2.258915772912601e-07,
      "loss": 0.24435906410217284,
      "num_input_tokens_seen": 850034112,
      "step": 14175,
      "train_runtime": 182033.5849,
      "train_tokens_per_second": 4669.655
    },
    {
      "epoch": 1.808673469387755,
      "grad_norm": 0.6072056036925845,
      "learning_rate": 2.2440542658892172e-07,
      "loss": 0.21573228836059571,
      "num_input_tokens_seen": 850345832,
      "step": 14180,
      "train_runtime": 182096.2068,
      "train_tokens_per_second": 4669.761
    },
    {
      "epoch": 1.8093112244897958,
      "grad_norm": 0.6051100953629287,
      "learning_rate": 2.2292406852926385e-07,
      "loss": 0.2449204444885254,
      "num_input_tokens_seen": 850645112,
      "step": 14185,
      "train_runtime": 182165.1754,
      "train_tokens_per_second": 4669.636
    },
    {
      "epoch": 1.8099489795918369,
      "grad_norm": 0.6962048069554118,
      "learning_rate": 2.214475045989345e-07,
      "loss": 0.2336430549621582,
      "num_input_tokens_seen": 850940792,
      "step": 14190,
      "train_runtime": 182240.0839,
      "train_tokens_per_second": 4669.339
    },
    {
      "epoch": 1.8105867346938775,
      "grad_norm": 0.7062951774689636,
      "learning_rate": 2.199757362797672e-07,
      "loss": 0.23945298194885253,
      "num_input_tokens_seen": 851231432,
      "step": 14195,
      "train_runtime": 182307.0608,
      "train_tokens_per_second": 4669.218
    },
    {
      "epoch": 1.8112244897959182,
      "grad_norm": 1.188855147983575,
      "learning_rate": 2.1850876504878492e-07,
      "loss": 0.2236766815185547,
      "num_input_tokens_seen": 851529712,
      "step": 14200,
      "train_runtime": 182376.9165,
      "train_tokens_per_second": 4669.065
    },
    {
      "epoch": 1.8118622448979593,
      "grad_norm": 0.6549158659499354,
      "learning_rate": 2.1704659237819558e-07,
      "loss": 0.2454591989517212,
      "num_input_tokens_seen": 851836424,
      "step": 14205,
      "train_runtime": 182436.9974,
      "train_tokens_per_second": 4669.209
    },
    {
      "epoch": 1.8125,
      "grad_norm": 0.6312331275412635,
      "learning_rate": 2.1558921973539092e-07,
      "loss": 0.21552486419677735,
      "num_input_tokens_seen": 852146744,
      "step": 14210,
      "train_runtime": 182511.4477,
      "train_tokens_per_second": 4669.004
    },
    {
      "epoch": 1.8131377551020407,
      "grad_norm": 0.6914111406413982,
      "learning_rate": 2.1413664858294713e-07,
      "loss": 0.2277005672454834,
      "num_input_tokens_seen": 852446456,
      "step": 14215,
      "train_runtime": 182579.4807,
      "train_tokens_per_second": 4668.906
    },
    {
      "epoch": 1.8137755102040818,
      "grad_norm": 0.623038086963682,
      "learning_rate": 2.126888803786209e-07,
      "loss": 0.23006982803344728,
      "num_input_tokens_seen": 852750528,
      "step": 14220,
      "train_runtime": 182641.2875,
      "train_tokens_per_second": 4668.991
    },
    {
      "epoch": 1.8144132653061225,
      "grad_norm": 0.6592848128998998,
      "learning_rate": 2.1124591657534776e-07,
      "loss": 0.21732797622680664,
      "num_input_tokens_seen": 853055240,
      "step": 14225,
      "train_runtime": 182707.4555,
      "train_tokens_per_second": 4668.968
    },
    {
      "epoch": 1.8150510204081631,
      "grad_norm": 0.5890391713015771,
      "learning_rate": 2.0980775862124326e-07,
      "loss": 0.2323162078857422,
      "num_input_tokens_seen": 853354320,
      "step": 14230,
      "train_runtime": 182782.2905,
      "train_tokens_per_second": 4668.693
    },
    {
      "epoch": 1.8156887755102042,
      "grad_norm": 0.6794000925800131,
      "learning_rate": 2.083744079595995e-07,
      "loss": 0.24140748977661133,
      "num_input_tokens_seen": 853663264,
      "step": 14235,
      "train_runtime": 182840.7922,
      "train_tokens_per_second": 4668.888
    },
    {
      "epoch": 1.816326530612245,
      "grad_norm": 0.6828682761767146,
      "learning_rate": 2.069458660288831e-07,
      "loss": 0.25820174217224123,
      "num_input_tokens_seen": 853957832,
      "step": 14240,
      "train_runtime": 182902.7662,
      "train_tokens_per_second": 4668.917
    },
    {
      "epoch": 1.8169642857142856,
      "grad_norm": 0.7150242243767747,
      "learning_rate": 2.0552213426273725e-07,
      "loss": 0.24172863960266114,
      "num_input_tokens_seen": 854261976,
      "step": 14245,
      "train_runtime": 182967.128,
      "train_tokens_per_second": 4668.937
    },
    {
      "epoch": 1.8176020408163265,
      "grad_norm": 0.7799388527982327,
      "learning_rate": 2.0410321408997459e-07,
      "loss": 0.2411952495574951,
      "num_input_tokens_seen": 854546712,
      "step": 14250,
      "train_runtime": 183036.3514,
      "train_tokens_per_second": 4668.727
    },
    {
      "epoch": 1.8182397959183674,
      "grad_norm": 0.6984524340201993,
      "learning_rate": 2.0268910693458154e-07,
      "loss": 0.24857621192932128,
      "num_input_tokens_seen": 854856912,
      "step": 14255,
      "train_runtime": 183090.5655,
      "train_tokens_per_second": 4669.039
    },
    {
      "epoch": 1.818877551020408,
      "grad_norm": 0.691502977060723,
      "learning_rate": 2.0127981421571295e-07,
      "loss": 0.23807902336120607,
      "num_input_tokens_seen": 855144504,
      "step": 14260,
      "train_runtime": 183158.5715,
      "train_tokens_per_second": 4668.875
    },
    {
      "epoch": 1.819515306122449,
      "grad_norm": 0.6683762091739054,
      "learning_rate": 1.998753373476936e-07,
      "loss": 0.2344792127609253,
      "num_input_tokens_seen": 855445152,
      "step": 14265,
      "train_runtime": 183218.1859,
      "train_tokens_per_second": 4668.997
    },
    {
      "epoch": 1.8201530612244898,
      "grad_norm": 0.6764785598399716,
      "learning_rate": 1.9847567774001153e-07,
      "loss": 0.24037613868713378,
      "num_input_tokens_seen": 855746640,
      "step": 14270,
      "train_runtime": 183285.9301,
      "train_tokens_per_second": 4668.916
    },
    {
      "epoch": 1.8207908163265305,
      "grad_norm": 0.68295919677254,
      "learning_rate": 1.9708083679732538e-07,
      "loss": 0.23158681392669678,
      "num_input_tokens_seen": 856050944,
      "step": 14275,
      "train_runtime": 183346.0516,
      "train_tokens_per_second": 4669.045
    },
    {
      "epoch": 1.8214285714285714,
      "grad_norm": 0.6269726263893487,
      "learning_rate": 1.9569081591945537e-07,
      "loss": 0.21793708801269532,
      "num_input_tokens_seen": 856361888,
      "step": 14280,
      "train_runtime": 183409.0954,
      "train_tokens_per_second": 4669.135
    },
    {
      "epoch": 1.8220663265306123,
      "grad_norm": 0.6775442090980904,
      "learning_rate": 1.9430561650138292e-07,
      "loss": 0.23641440868377686,
      "num_input_tokens_seen": 856656952,
      "step": 14285,
      "train_runtime": 183471.7968,
      "train_tokens_per_second": 4669.148
    },
    {
      "epoch": 1.822704081632653,
      "grad_norm": 0.694221107240311,
      "learning_rate": 1.9292523993325318e-07,
      "loss": 0.25249128341674804,
      "num_input_tokens_seen": 856967800,
      "step": 14290,
      "train_runtime": 183548.6432,
      "train_tokens_per_second": 4668.887
    },
    {
      "epoch": 1.8233418367346939,
      "grad_norm": 0.6539747262144159,
      "learning_rate": 1.9154968760037142e-07,
      "loss": 0.24744951725006104,
      "num_input_tokens_seen": 857270128,
      "step": 14295,
      "train_runtime": 183608.628,
      "train_tokens_per_second": 4669.008
    },
    {
      "epoch": 1.8239795918367347,
      "grad_norm": 0.6400158968152201,
      "learning_rate": 1.9017896088319888e-07,
      "loss": 0.2374544620513916,
      "num_input_tokens_seen": 857576328,
      "step": 14300,
      "train_runtime": 183673.4765,
      "train_tokens_per_second": 4669.026
    },
    {
      "epoch": 1.8246173469387754,
      "grad_norm": 0.7063594863011754,
      "learning_rate": 1.8881306115735632e-07,
      "loss": 0.24543800354003906,
      "num_input_tokens_seen": 857864376,
      "step": 14305,
      "train_runtime": 183741.1776,
      "train_tokens_per_second": 4668.874
    },
    {
      "epoch": 1.8252551020408163,
      "grad_norm": 0.6335893908540836,
      "learning_rate": 1.8745198979361944e-07,
      "loss": 0.23575279712677003,
      "num_input_tokens_seen": 858166808,
      "step": 14310,
      "train_runtime": 183802.1091,
      "train_tokens_per_second": 4668.971
    },
    {
      "epoch": 1.8258928571428572,
      "grad_norm": 0.6032855222955488,
      "learning_rate": 1.8609574815791843e-07,
      "loss": 0.23586275577545165,
      "num_input_tokens_seen": 858466344,
      "step": 14315,
      "train_runtime": 183865.3456,
      "train_tokens_per_second": 4668.995
    },
    {
      "epoch": 1.8265306122448979,
      "grad_norm": 0.6065067315991968,
      "learning_rate": 1.8474433761133614e-07,
      "loss": 0.22159624099731445,
      "num_input_tokens_seen": 858751184,
      "step": 14320,
      "train_runtime": 183930.6342,
      "train_tokens_per_second": 4668.886
    },
    {
      "epoch": 1.8271683673469388,
      "grad_norm": 0.6386792360458959,
      "learning_rate": 1.833977595101083e-07,
      "loss": 0.25848994255065916,
      "num_input_tokens_seen": 859064664,
      "step": 14325,
      "train_runtime": 184005.4578,
      "train_tokens_per_second": 4668.691
    },
    {
      "epoch": 1.8278061224489797,
      "grad_norm": 0.7243826811963424,
      "learning_rate": 1.8205601520561945e-07,
      "loss": 0.2323843240737915,
      "num_input_tokens_seen": 859362472,
      "step": 14330,
      "train_runtime": 184060.8739,
      "train_tokens_per_second": 4668.904
    },
    {
      "epoch": 1.8284438775510203,
      "grad_norm": 0.7113576316394074,
      "learning_rate": 1.8071910604440356e-07,
      "loss": 0.2448119878768921,
      "num_input_tokens_seen": 859662920,
      "step": 14335,
      "train_runtime": 184136.5773,
      "train_tokens_per_second": 4668.616
    },
    {
      "epoch": 1.8290816326530612,
      "grad_norm": 0.5755378370316325,
      "learning_rate": 1.79387033368143e-07,
      "loss": 0.2230464220046997,
      "num_input_tokens_seen": 859961152,
      "step": 14340,
      "train_runtime": 184207.2545,
      "train_tokens_per_second": 4668.443
    },
    {
      "epoch": 1.8297193877551021,
      "grad_norm": 0.6114534758619493,
      "learning_rate": 1.7805979851366505e-07,
      "loss": 0.23956425189971925,
      "num_input_tokens_seen": 860261560,
      "step": 14345,
      "train_runtime": 184275.2068,
      "train_tokens_per_second": 4668.352
    },
    {
      "epoch": 1.8303571428571428,
      "grad_norm": 0.6469560822067821,
      "learning_rate": 1.7673740281294316e-07,
      "loss": 0.25022525787353517,
      "num_input_tokens_seen": 860558928,
      "step": 14350,
      "train_runtime": 184332.7459,
      "train_tokens_per_second": 4668.508
    },
    {
      "epoch": 1.8309948979591837,
      "grad_norm": 0.6437016015002156,
      "learning_rate": 1.7541984759309294e-07,
      "loss": 0.24035637378692626,
      "num_input_tokens_seen": 860855480,
      "step": 14355,
      "train_runtime": 184395.2999,
      "train_tokens_per_second": 4668.533
    },
    {
      "epoch": 1.8316326530612246,
      "grad_norm": 0.6600952592336741,
      "learning_rate": 1.7410713417637338e-07,
      "loss": 0.25348525047302245,
      "num_input_tokens_seen": 861151800,
      "step": 14360,
      "train_runtime": 184460.4383,
      "train_tokens_per_second": 4668.49
    },
    {
      "epoch": 1.8322704081632653,
      "grad_norm": 0.6087445950838055,
      "learning_rate": 1.7279926388018564e-07,
      "loss": 0.23996715545654296,
      "num_input_tokens_seen": 861455912,
      "step": 14365,
      "train_runtime": 184522.8519,
      "train_tokens_per_second": 4668.559
    },
    {
      "epoch": 1.8329081632653061,
      "grad_norm": 0.6504701631415639,
      "learning_rate": 1.714962380170665e-07,
      "loss": 0.23213858604431153,
      "num_input_tokens_seen": 861758272,
      "step": 14370,
      "train_runtime": 184581.0034,
      "train_tokens_per_second": 4668.727
    },
    {
      "epoch": 1.833545918367347,
      "grad_norm": 0.6261043429395752,
      "learning_rate": 1.701980578946938e-07,
      "loss": 0.24935603141784668,
      "num_input_tokens_seen": 862063536,
      "step": 14375,
      "train_runtime": 184657.0754,
      "train_tokens_per_second": 4668.457
    },
    {
      "epoch": 1.8341836734693877,
      "grad_norm": 0.6270297807543881,
      "learning_rate": 1.6890472481588426e-07,
      "loss": 0.23564741611480713,
      "num_input_tokens_seen": 862356432,
      "step": 14380,
      "train_runtime": 184726.1435,
      "train_tokens_per_second": 4668.297
    },
    {
      "epoch": 1.8348214285714286,
      "grad_norm": 0.6711032745179101,
      "learning_rate": 1.6761624007858524e-07,
      "loss": 0.2496790885925293,
      "num_input_tokens_seen": 862663984,
      "step": 14385,
      "train_runtime": 184788.2754,
      "train_tokens_per_second": 4668.391
    },
    {
      "epoch": 1.8354591836734695,
      "grad_norm": 0.641643337265233,
      "learning_rate": 1.6633260497588233e-07,
      "loss": 0.22472338676452636,
      "num_input_tokens_seen": 862959176,
      "step": 14390,
      "train_runtime": 184852.6304,
      "train_tokens_per_second": 4668.363
    },
    {
      "epoch": 1.8360969387755102,
      "grad_norm": 0.5647985821520639,
      "learning_rate": 1.6505382079599341e-07,
      "loss": 0.22203531265258789,
      "num_input_tokens_seen": 863267536,
      "step": 14395,
      "train_runtime": 184921.6317,
      "train_tokens_per_second": 4668.289
    },
    {
      "epoch": 1.836734693877551,
      "grad_norm": 0.6576296676798646,
      "learning_rate": 1.6377988882226691e-07,
      "loss": 0.2492135286331177,
      "num_input_tokens_seen": 863564488,
      "step": 14400,
      "train_runtime": 184993.3183,
      "train_tokens_per_second": 4668.085
    },
    {
      "epoch": 1.837372448979592,
      "grad_norm": 0.6642034714755283,
      "learning_rate": 1.6251081033318294e-07,
      "loss": 0.2369211196899414,
      "num_input_tokens_seen": 863853768,
      "step": 14405,
      "train_runtime": 185060.0842,
      "train_tokens_per_second": 4667.964
    },
    {
      "epoch": 1.8380102040816326,
      "grad_norm": 0.6598234979501645,
      "learning_rate": 1.6124658660234993e-07,
      "loss": 0.22833032608032228,
      "num_input_tokens_seen": 864159608,
      "step": 14410,
      "train_runtime": 185115.1498,
      "train_tokens_per_second": 4668.227
    },
    {
      "epoch": 1.8386479591836735,
      "grad_norm": 0.6709623601559725,
      "learning_rate": 1.5998721889850577e-07,
      "loss": 0.22344694137573243,
      "num_input_tokens_seen": 864464216,
      "step": 14415,
      "train_runtime": 185172.0534,
      "train_tokens_per_second": 4668.438
    },
    {
      "epoch": 1.8392857142857144,
      "grad_norm": 0.662379928517177,
      "learning_rate": 1.5873270848551227e-07,
      "loss": 0.23959665298461913,
      "num_input_tokens_seen": 864761688,
      "step": 14420,
      "train_runtime": 185241.837,
      "train_tokens_per_second": 4668.285
    },
    {
      "epoch": 1.839923469387755,
      "grad_norm": 0.6046807574839765,
      "learning_rate": 1.5748305662236008e-07,
      "loss": 0.23878827095031738,
      "num_input_tokens_seen": 865072656,
      "step": 14425,
      "train_runtime": 185301.7857,
      "train_tokens_per_second": 4668.453
    },
    {
      "epoch": 1.8405612244897958,
      "grad_norm": 0.5907992672075336,
      "learning_rate": 1.562382645631605e-07,
      "loss": 0.24034523963928223,
      "num_input_tokens_seen": 865366568,
      "step": 14430,
      "train_runtime": 185374.3333,
      "train_tokens_per_second": 4668.211
    },
    {
      "epoch": 1.8411989795918369,
      "grad_norm": 0.6488857955022216,
      "learning_rate": 1.549983335571509e-07,
      "loss": 0.24318280220031738,
      "num_input_tokens_seen": 865673504,
      "step": 14435,
      "train_runtime": 185436.3672,
      "train_tokens_per_second": 4668.305
    },
    {
      "epoch": 1.8418367346938775,
      "grad_norm": 0.6345954947782853,
      "learning_rate": 1.537632648486864e-07,
      "loss": 0.2233747959136963,
      "num_input_tokens_seen": 865966584,
      "step": 14440,
      "train_runtime": 185502.4597,
      "train_tokens_per_second": 4668.222
    },
    {
      "epoch": 1.8424744897959182,
      "grad_norm": 0.6093667539618414,
      "learning_rate": 1.5253305967724674e-07,
      "loss": 0.23251943588256835,
      "num_input_tokens_seen": 866262616,
      "step": 14445,
      "train_runtime": 185567.1899,
      "train_tokens_per_second": 4668.188
    },
    {
      "epoch": 1.8431122448979593,
      "grad_norm": 0.6576062354450807,
      "learning_rate": 1.513077192774287e-07,
      "loss": 0.2469010829925537,
      "num_input_tokens_seen": 866575568,
      "step": 14450,
      "train_runtime": 185620.909,
      "train_tokens_per_second": 4668.523
    },
    {
      "epoch": 1.84375,
      "grad_norm": 0.665884965746731,
      "learning_rate": 1.500872448789459e-07,
      "loss": 0.22332825660705566,
      "num_input_tokens_seen": 866884552,
      "step": 14455,
      "train_runtime": 185690.0122,
      "train_tokens_per_second": 4668.45
    },
    {
      "epoch": 1.8443877551020407,
      "grad_norm": 0.6526420329302832,
      "learning_rate": 1.4887163770662905e-07,
      "loss": 0.226957368850708,
      "num_input_tokens_seen": 867174168,
      "step": 14460,
      "train_runtime": 185753.506,
      "train_tokens_per_second": 4668.413
    },
    {
      "epoch": 1.8450255102040818,
      "grad_norm": 0.6001558081237469,
      "learning_rate": 1.4766089898042678e-07,
      "loss": 0.2408813238143921,
      "num_input_tokens_seen": 867469624,
      "step": 14465,
      "train_runtime": 185821.4483,
      "train_tokens_per_second": 4668.297
    },
    {
      "epoch": 1.8456632653061225,
      "grad_norm": 0.6956641287999953,
      "learning_rate": 1.4645502991539883e-07,
      "loss": 0.23400726318359374,
      "num_input_tokens_seen": 867765896,
      "step": 14470,
      "train_runtime": 185880.418,
      "train_tokens_per_second": 4668.409
    },
    {
      "epoch": 1.8463010204081631,
      "grad_norm": 0.6665358852671845,
      "learning_rate": 1.452540317217188e-07,
      "loss": 0.2394644021987915,
      "num_input_tokens_seen": 868071176,
      "step": 14475,
      "train_runtime": 185946.7671,
      "train_tokens_per_second": 4668.385
    },
    {
      "epoch": 1.8469387755102042,
      "grad_norm": 0.7054297858351227,
      "learning_rate": 1.440579056046737e-07,
      "loss": 0.2262880563735962,
      "num_input_tokens_seen": 868362264,
      "step": 14480,
      "train_runtime": 186010.9837,
      "train_tokens_per_second": 4668.339
    },
    {
      "epoch": 1.847576530612245,
      "grad_norm": 0.6808849931657256,
      "learning_rate": 1.428666527646577e-07,
      "loss": 0.23255569934844972,
      "num_input_tokens_seen": 868646368,
      "step": 14485,
      "train_runtime": 186068.3786,
      "train_tokens_per_second": 4668.426
    },
    {
      "epoch": 1.8482142857142856,
      "grad_norm": 0.7317704340164283,
      "learning_rate": 1.4168027439717735e-07,
      "loss": 0.2767033100128174,
      "num_input_tokens_seen": 868964096,
      "step": 14490,
      "train_runtime": 186121.5642,
      "train_tokens_per_second": 4668.799
    },
    {
      "epoch": 1.8488520408163265,
      "grad_norm": 0.892917228411661,
      "learning_rate": 1.4049877169284575e-07,
      "loss": 0.2565030574798584,
      "num_input_tokens_seen": 869264864,
      "step": 14495,
      "train_runtime": 186184.094,
      "train_tokens_per_second": 4668.846
    },
    {
      "epoch": 1.8494897959183674,
      "grad_norm": 0.5994045672594449,
      "learning_rate": 1.3932214583738334e-07,
      "loss": 0.2315276861190796,
      "num_input_tokens_seen": 869578544,
      "step": 14500,
      "train_runtime": 186243.2777,
      "train_tokens_per_second": 4669.047
    },
    {
      "epoch": 1.850127551020408,
      "grad_norm": 0.6367895708219833,
      "learning_rate": 1.3815039801161723e-07,
      "loss": 0.25290961265563966,
      "num_input_tokens_seen": 869874568,
      "step": 14505,
      "train_runtime": 186309.0778,
      "train_tokens_per_second": 4668.986
    },
    {
      "epoch": 1.850765306122449,
      "grad_norm": 0.637022963753881,
      "learning_rate": 1.369835293914773e-07,
      "loss": 0.2709094524383545,
      "num_input_tokens_seen": 870193664,
      "step": 14510,
      "train_runtime": 186368.8791,
      "train_tokens_per_second": 4669.201
    },
    {
      "epoch": 1.8514030612244898,
      "grad_norm": 0.6272452650829071,
      "learning_rate": 1.358215411479974e-07,
      "loss": 0.2437835454940796,
      "num_input_tokens_seen": 870507328,
      "step": 14515,
      "train_runtime": 186441.4503,
      "train_tokens_per_second": 4669.065
    },
    {
      "epoch": 1.8520408163265305,
      "grad_norm": 0.6047628176878721,
      "learning_rate": 1.3466443444731525e-07,
      "loss": 0.23250491619110109,
      "num_input_tokens_seen": 870806344,
      "step": 14520,
      "train_runtime": 186512.5533,
      "train_tokens_per_second": 4668.889
    },
    {
      "epoch": 1.8526785714285714,
      "grad_norm": 0.6340814502818172,
      "learning_rate": 1.3351221045066698e-07,
      "loss": 0.23869748115539552,
      "num_input_tokens_seen": 871111792,
      "step": 14525,
      "train_runtime": 186578.1182,
      "train_tokens_per_second": 4668.885
    },
    {
      "epoch": 1.8533163265306123,
      "grad_norm": 0.6143465477228329,
      "learning_rate": 1.3236487031439037e-07,
      "loss": 0.23951935768127441,
      "num_input_tokens_seen": 871406384,
      "step": 14530,
      "train_runtime": 186646.2647,
      "train_tokens_per_second": 4668.759
    },
    {
      "epoch": 1.853954081632653,
      "grad_norm": 0.6278573970486169,
      "learning_rate": 1.3122241518992107e-07,
      "loss": 0.21364061832427977,
      "num_input_tokens_seen": 871696536,
      "step": 14535,
      "train_runtime": 186704.199,
      "train_tokens_per_second": 4668.864
    },
    {
      "epoch": 1.8545918367346939,
      "grad_norm": 0.6947604802332978,
      "learning_rate": 1.3008484622379358e-07,
      "loss": 0.24540648460388184,
      "num_input_tokens_seen": 871993096,
      "step": 14540,
      "train_runtime": 186768.0732,
      "train_tokens_per_second": 4668.855
    },
    {
      "epoch": 1.8552295918367347,
      "grad_norm": 0.6678310631928734,
      "learning_rate": 1.2895216455763582e-07,
      "loss": 0.21504359245300292,
      "num_input_tokens_seen": 872292056,
      "step": 14545,
      "train_runtime": 186838.1946,
      "train_tokens_per_second": 4668.703
    },
    {
      "epoch": 1.8558673469387754,
      "grad_norm": 0.6101568298619875,
      "learning_rate": 1.278243713281757e-07,
      "loss": 0.22426114082336426,
      "num_input_tokens_seen": 872596776,
      "step": 14550,
      "train_runtime": 186885.3811,
      "train_tokens_per_second": 4669.155
    },
    {
      "epoch": 1.8565051020408163,
      "grad_norm": 0.630720751877991,
      "learning_rate": 1.2670146766723013e-07,
      "loss": 0.24573097229003907,
      "num_input_tokens_seen": 872897784,
      "step": 14555,
      "train_runtime": 186946.4855,
      "train_tokens_per_second": 4669.239
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.7352603329608624,
      "learning_rate": 1.2558345470171153e-07,
      "loss": 0.22406630516052245,
      "num_input_tokens_seen": 873195496,
      "step": 14560,
      "train_runtime": 187015.3542,
      "train_tokens_per_second": 4669.111
    },
    {
      "epoch": 1.8577806122448979,
      "grad_norm": 0.6045920803471069,
      "learning_rate": 1.244703335536257e-07,
      "loss": 0.21021738052368164,
      "num_input_tokens_seen": 873510560,
      "step": 14565,
      "train_runtime": 187084.6298,
      "train_tokens_per_second": 4669.066
    },
    {
      "epoch": 1.8584183673469388,
      "grad_norm": 0.5968331843220145,
      "learning_rate": 1.2336210534006576e-07,
      "loss": 0.22899558544158935,
      "num_input_tokens_seen": 873811360,
      "step": 14570,
      "train_runtime": 187141.7871,
      "train_tokens_per_second": 4669.248
    },
    {
      "epoch": 1.8590561224489797,
      "grad_norm": 0.6543841944293568,
      "learning_rate": 1.2225877117321705e-07,
      "loss": 0.2300337314605713,
      "num_input_tokens_seen": 874115088,
      "step": 14575,
      "train_runtime": 187193.497,
      "train_tokens_per_second": 4669.58
    },
    {
      "epoch": 1.8596938775510203,
      "grad_norm": 0.7193327651779114,
      "learning_rate": 1.2116033216035162e-07,
      "loss": 0.24243698120117188,
      "num_input_tokens_seen": 874413560,
      "step": 14580,
      "train_runtime": 187256.3253,
      "train_tokens_per_second": 4669.608
    },
    {
      "epoch": 1.8603316326530612,
      "grad_norm": 0.605379428490787,
      "learning_rate": 1.20066789403831e-07,
      "loss": 0.2280111789703369,
      "num_input_tokens_seen": 874718728,
      "step": 14585,
      "train_runtime": 187312.103,
      "train_tokens_per_second": 4669.846
    },
    {
      "epoch": 1.8609693877551021,
      "grad_norm": 0.6197025506402201,
      "learning_rate": 1.1897814400110064e-07,
      "loss": 0.21057276725769042,
      "num_input_tokens_seen": 875021888,
      "step": 14590,
      "train_runtime": 187373.8204,
      "train_tokens_per_second": 4669.926
    },
    {
      "epoch": 1.8616071428571428,
      "grad_norm": 0.6212197732973205,
      "learning_rate": 1.1789439704469219e-07,
      "loss": 0.23755302429199218,
      "num_input_tokens_seen": 875319048,
      "step": 14595,
      "train_runtime": 187422.7258,
      "train_tokens_per_second": 4670.293
    },
    {
      "epoch": 1.8622448979591837,
      "grad_norm": 0.5975752241743172,
      "learning_rate": 1.168155496222223e-07,
      "loss": 0.24098501205444336,
      "num_input_tokens_seen": 875626728,
      "step": 14600,
      "train_runtime": 187481.5352,
      "train_tokens_per_second": 4670.469
    },
    {
      "epoch": 1.8628826530612246,
      "grad_norm": 0.6388992963446776,
      "learning_rate": 1.1574160281638935e-07,
      "loss": 0.23636155128479003,
      "num_input_tokens_seen": 875919600,
      "step": 14605,
      "train_runtime": 187541.4211,
      "train_tokens_per_second": 4670.539
    },
    {
      "epoch": 1.8635204081632653,
      "grad_norm": 0.7144456643858664,
      "learning_rate": 1.1467255770497343e-07,
      "loss": 0.23244943618774414,
      "num_input_tokens_seen": 876221560,
      "step": 14610,
      "train_runtime": 187603.397,
      "train_tokens_per_second": 4670.606
    },
    {
      "epoch": 1.8641581632653061,
      "grad_norm": 0.6817582985285732,
      "learning_rate": 1.1360841536083633e-07,
      "loss": 0.21701469421386718,
      "num_input_tokens_seen": 876511832,
      "step": 14615,
      "train_runtime": 187670.4552,
      "train_tokens_per_second": 4670.484
    },
    {
      "epoch": 1.864795918367347,
      "grad_norm": 0.627570811391596,
      "learning_rate": 1.125491768519199e-07,
      "loss": 0.24056625366210938,
      "num_input_tokens_seen": 876828424,
      "step": 14620,
      "train_runtime": 187732.5249,
      "train_tokens_per_second": 4670.626
    },
    {
      "epoch": 1.8654336734693877,
      "grad_norm": 0.7294722509912248,
      "learning_rate": 1.1149484324124326e-07,
      "loss": 0.2572359561920166,
      "num_input_tokens_seen": 877131488,
      "step": 14625,
      "train_runtime": 187806.3447,
      "train_tokens_per_second": 4670.404
    },
    {
      "epoch": 1.8660714285714286,
      "grad_norm": 0.693302961896091,
      "learning_rate": 1.1044541558690281e-07,
      "loss": 0.24057960510253906,
      "num_input_tokens_seen": 877422776,
      "step": 14630,
      "train_runtime": 187872.14,
      "train_tokens_per_second": 4670.319
    },
    {
      "epoch": 1.8667091836734695,
      "grad_norm": 0.6165680252752774,
      "learning_rate": 1.094008949420744e-07,
      "loss": 0.24250717163085939,
      "num_input_tokens_seen": 877722544,
      "step": 14635,
      "train_runtime": 187938.663,
      "train_tokens_per_second": 4670.261
    },
    {
      "epoch": 1.8673469387755102,
      "grad_norm": 0.6002669462477733,
      "learning_rate": 1.0836128235500676e-07,
      "loss": 0.23071460723876952,
      "num_input_tokens_seen": 878029288,
      "step": 14640,
      "train_runtime": 188007.6859,
      "train_tokens_per_second": 4670.178
    },
    {
      "epoch": 1.867984693877551,
      "grad_norm": 0.6771460444986078,
      "learning_rate": 1.0732657886902308e-07,
      "loss": 0.23574914932250976,
      "num_input_tokens_seen": 878335192,
      "step": 14645,
      "train_runtime": 188076.2415,
      "train_tokens_per_second": 4670.102
    },
    {
      "epoch": 1.868622448979592,
      "grad_norm": 0.5840827704305357,
      "learning_rate": 1.0629678552252109e-07,
      "loss": 0.21893777847290039,
      "num_input_tokens_seen": 878636112,
      "step": 14650,
      "train_runtime": 188143.9317,
      "train_tokens_per_second": 4670.021
    },
    {
      "epoch": 1.8692602040816326,
      "grad_norm": 0.7404579482784661,
      "learning_rate": 1.0527190334897076e-07,
      "loss": 0.246806001663208,
      "num_input_tokens_seen": 878933752,
      "step": 14655,
      "train_runtime": 188207.2491,
      "train_tokens_per_second": 4670.031
    },
    {
      "epoch": 1.8698979591836735,
      "grad_norm": 0.6590259305605255,
      "learning_rate": 1.0425193337691208e-07,
      "loss": 0.22685532569885253,
      "num_input_tokens_seen": 879246712,
      "step": 14660,
      "train_runtime": 188273.9892,
      "train_tokens_per_second": 4670.038
    },
    {
      "epoch": 1.8705357142857144,
      "grad_norm": 0.7058607713702875,
      "learning_rate": 1.0323687662995685e-07,
      "loss": 0.22674460411071778,
      "num_input_tokens_seen": 879549144,
      "step": 14665,
      "train_runtime": 188337.3823,
      "train_tokens_per_second": 4670.072
    },
    {
      "epoch": 1.871173469387755,
      "grad_norm": 0.6503089387113022,
      "learning_rate": 1.0222673412678518e-07,
      "loss": 0.2473440170288086,
      "num_input_tokens_seen": 879859696,
      "step": 14670,
      "train_runtime": 188398.9085,
      "train_tokens_per_second": 4670.195
    },
    {
      "epoch": 1.8718112244897958,
      "grad_norm": 0.5604952931271503,
      "learning_rate": 1.0122150688114507e-07,
      "loss": 0.24516563415527343,
      "num_input_tokens_seen": 880169896,
      "step": 14675,
      "train_runtime": 188464.7909,
      "train_tokens_per_second": 4670.209
    },
    {
      "epoch": 1.8724489795918369,
      "grad_norm": 0.6788083731679111,
      "learning_rate": 1.0022119590185287e-07,
      "loss": 0.22207980155944823,
      "num_input_tokens_seen": 880480736,
      "step": 14680,
      "train_runtime": 188520.1778,
      "train_tokens_per_second": 4670.485
    },
    {
      "epoch": 1.8730867346938775,
      "grad_norm": 0.6197056753863878,
      "learning_rate": 9.922580219279055e-08,
      "loss": 0.24501399993896483,
      "num_input_tokens_seen": 880784904,
      "step": 14685,
      "train_runtime": 188578.492,
      "train_tokens_per_second": 4670.654
    },
    {
      "epoch": 1.8737244897959182,
      "grad_norm": 0.6536242686249847,
      "learning_rate": 9.82353267529046e-08,
      "loss": 0.2373288631439209,
      "num_input_tokens_seen": 881082192,
      "step": 14690,
      "train_runtime": 188647.7988,
      "train_tokens_per_second": 4670.514
    },
    {
      "epoch": 1.8743622448979593,
      "grad_norm": 0.6091687376173869,
      "learning_rate": 9.72497705762071e-08,
      "loss": 0.24420948028564454,
      "num_input_tokens_seen": 881380784,
      "step": 14695,
      "train_runtime": 188704.9611,
      "train_tokens_per_second": 4670.682
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.6559991166313782,
      "learning_rate": 9.626913465177246e-08,
      "loss": 0.25005807876586916,
      "num_input_tokens_seen": 881690656,
      "step": 14700,
      "train_runtime": 188762.798,
      "train_tokens_per_second": 4670.892
    },
    {
      "epoch": 1.8756377551020407,
      "grad_norm": 0.612906653695731,
      "learning_rate": 9.529341996373675e-08,
      "loss": 0.22320895195007323,
      "num_input_tokens_seen": 881983480,
      "step": 14705,
      "train_runtime": 188837.3232,
      "train_tokens_per_second": 4670.599
    },
    {
      "epoch": 1.8762755102040818,
      "grad_norm": 0.6738789264807418,
      "learning_rate": 9.432262749129784e-08,
      "loss": 0.23938822746276855,
      "num_input_tokens_seen": 882261616,
      "step": 14710,
      "train_runtime": 188903.7599,
      "train_tokens_per_second": 4670.429
    },
    {
      "epoch": 1.8769132653061225,
      "grad_norm": 0.6590420467075672,
      "learning_rate": 9.335675820871415e-08,
      "loss": 0.2600644588470459,
      "num_input_tokens_seen": 882559920,
      "step": 14715,
      "train_runtime": 188979.7054,
      "train_tokens_per_second": 4670.131
    },
    {
      "epoch": 1.8775510204081631,
      "grad_norm": 0.7053299947071794,
      "learning_rate": 9.239581308530254e-08,
      "loss": 0.23621573448181152,
      "num_input_tokens_seen": 882858320,
      "step": 14720,
      "train_runtime": 189048.0866,
      "train_tokens_per_second": 4670.02
    },
    {
      "epoch": 1.8781887755102042,
      "grad_norm": 0.6725471849133484,
      "learning_rate": 9.143979308543993e-08,
      "loss": 0.24531195163726807,
      "num_input_tokens_seen": 883153856,
      "step": 14725,
      "train_runtime": 189112.9534,
      "train_tokens_per_second": 4669.981
    },
    {
      "epoch": 1.878826530612245,
      "grad_norm": 0.7108238210464412,
      "learning_rate": 9.048869916855662e-08,
      "loss": 0.21591539382934571,
      "num_input_tokens_seen": 883437160,
      "step": 14730,
      "train_runtime": 189174.4025,
      "train_tokens_per_second": 4669.961
    },
    {
      "epoch": 1.8794642857142856,
      "grad_norm": 0.6541061455855879,
      "learning_rate": 8.954253228914356e-08,
      "loss": 0.23390002250671388,
      "num_input_tokens_seen": 883720000,
      "step": 14735,
      "train_runtime": 189235.0474,
      "train_tokens_per_second": 4669.959
    },
    {
      "epoch": 1.8801020408163265,
      "grad_norm": 0.7518516631614485,
      "learning_rate": 8.860129339674562e-08,
      "loss": 0.24814205169677733,
      "num_input_tokens_seen": 884030264,
      "step": 14740,
      "train_runtime": 189298.0993,
      "train_tokens_per_second": 4670.043
    },
    {
      "epoch": 1.8807397959183674,
      "grad_norm": 0.726048580189418,
      "learning_rate": 8.766498343596053e-08,
      "loss": 0.21693217754364014,
      "num_input_tokens_seen": 884318136,
      "step": 14745,
      "train_runtime": 189365.5517,
      "train_tokens_per_second": 4669.9
    },
    {
      "epoch": 1.881377551020408,
      "grad_norm": 0.6600998374561002,
      "learning_rate": 8.67336033464411e-08,
      "loss": 0.24096481800079345,
      "num_input_tokens_seen": 884605888,
      "step": 14750,
      "train_runtime": 189428.1347,
      "train_tokens_per_second": 4669.876
    },
    {
      "epoch": 1.882015306122449,
      "grad_norm": 0.5852704161884672,
      "learning_rate": 8.580715406289353e-08,
      "loss": 0.23575496673583984,
      "num_input_tokens_seen": 884905752,
      "step": 14755,
      "train_runtime": 189494.6066,
      "train_tokens_per_second": 4669.82
    },
    {
      "epoch": 1.8826530612244898,
      "grad_norm": 0.6730731304858276,
      "learning_rate": 8.488563651507242e-08,
      "loss": 0.23407430648803712,
      "num_input_tokens_seen": 885201480,
      "step": 14760,
      "train_runtime": 189556.2553,
      "train_tokens_per_second": 4669.862
    },
    {
      "epoch": 1.8832908163265305,
      "grad_norm": 0.6611050619633174,
      "learning_rate": 8.396905162778523e-08,
      "loss": 0.23900527954101564,
      "num_input_tokens_seen": 885513816,
      "step": 14765,
      "train_runtime": 189614.3506,
      "train_tokens_per_second": 4670.078
    },
    {
      "epoch": 1.8839285714285714,
      "grad_norm": 0.6575305055614173,
      "learning_rate": 8.305740032089116e-08,
      "loss": 0.2386373519897461,
      "num_input_tokens_seen": 885804312,
      "step": 14770,
      "train_runtime": 189674.9686,
      "train_tokens_per_second": 4670.117
    },
    {
      "epoch": 1.8845663265306123,
      "grad_norm": 0.7400761573412182,
      "learning_rate": 8.215068350929334e-08,
      "loss": 0.24945921897888185,
      "num_input_tokens_seen": 886090304,
      "step": 14775,
      "train_runtime": 189746.0203,
      "train_tokens_per_second": 4669.876
    },
    {
      "epoch": 1.885204081632653,
      "grad_norm": 0.650024826467964,
      "learning_rate": 8.124890210294667e-08,
      "loss": 0.23193161487579345,
      "num_input_tokens_seen": 886383608,
      "step": 14780,
      "train_runtime": 189812.0482,
      "train_tokens_per_second": 4669.796
    },
    {
      "epoch": 1.8858418367346939,
      "grad_norm": 0.7150616702077507,
      "learning_rate": 8.035205700685167e-08,
      "loss": 0.2643417835235596,
      "num_input_tokens_seen": 886684968,
      "step": 14785,
      "train_runtime": 189880.3557,
      "train_tokens_per_second": 4669.704
    },
    {
      "epoch": 1.8864795918367347,
      "grad_norm": 0.6653687091903648,
      "learning_rate": 7.946014912105505e-08,
      "loss": 0.23194210529327391,
      "num_input_tokens_seen": 886989336,
      "step": 14790,
      "train_runtime": 189936.09,
      "train_tokens_per_second": 4669.936
    },
    {
      "epoch": 1.8871173469387754,
      "grad_norm": 0.6347424894790333,
      "learning_rate": 7.857317934064857e-08,
      "loss": 0.23241157531738282,
      "num_input_tokens_seen": 887304888,
      "step": 14795,
      "train_runtime": 189989.1431,
      "train_tokens_per_second": 4670.293
    },
    {
      "epoch": 1.8877551020408163,
      "grad_norm": 0.657102108937192,
      "learning_rate": 7.769114855576799e-08,
      "loss": 0.22253942489624023,
      "num_input_tokens_seen": 887613072,
      "step": 14800,
      "train_runtime": 190054.016,
      "train_tokens_per_second": 4670.32
    },
    {
      "epoch": 1.8883928571428572,
      "grad_norm": 0.684282852276083,
      "learning_rate": 7.6814057651593e-08,
      "loss": 0.23771584033966064,
      "num_input_tokens_seen": 887915816,
      "step": 14805,
      "train_runtime": 190118.3645,
      "train_tokens_per_second": 4670.332
    },
    {
      "epoch": 1.8890306122448979,
      "grad_norm": 0.6501394834124957,
      "learning_rate": 7.594190750834617e-08,
      "loss": 0.26170597076416013,
      "num_input_tokens_seen": 888220584,
      "step": 14810,
      "train_runtime": 190194.9364,
      "train_tokens_per_second": 4670.054
    },
    {
      "epoch": 1.8896683673469388,
      "grad_norm": 0.6861093797898603,
      "learning_rate": 7.507469900129016e-08,
      "loss": 0.22220227718353272,
      "num_input_tokens_seen": 888526080,
      "step": 14815,
      "train_runtime": 190242.9994,
      "train_tokens_per_second": 4670.48
    },
    {
      "epoch": 1.8903061224489797,
      "grad_norm": 0.6730178203742853,
      "learning_rate": 7.421243300072989e-08,
      "loss": 0.2384312629699707,
      "num_input_tokens_seen": 888816712,
      "step": 14820,
      "train_runtime": 190301.3519,
      "train_tokens_per_second": 4670.575
    },
    {
      "epoch": 1.8909438775510203,
      "grad_norm": 0.7650803401896588,
      "learning_rate": 7.335511037200982e-08,
      "loss": 0.24442434310913086,
      "num_input_tokens_seen": 889113872,
      "step": 14825,
      "train_runtime": 190366.3449,
      "train_tokens_per_second": 4670.541
    },
    {
      "epoch": 1.8915816326530612,
      "grad_norm": 0.7261122563642717,
      "learning_rate": 7.250273197551227e-08,
      "loss": 0.25747177600860593,
      "num_input_tokens_seen": 889419088,
      "step": 14830,
      "train_runtime": 190427.6903,
      "train_tokens_per_second": 4670.639
    },
    {
      "epoch": 1.8922193877551021,
      "grad_norm": 0.6046613288996454,
      "learning_rate": 7.165529866665855e-08,
      "loss": 0.23916244506835938,
      "num_input_tokens_seen": 889714880,
      "step": 14835,
      "train_runtime": 190481.327,
      "train_tokens_per_second": 4670.877
    },
    {
      "epoch": 1.8928571428571428,
      "grad_norm": 0.7351870622704426,
      "learning_rate": 7.081281129590777e-08,
      "loss": 0.2395155906677246,
      "num_input_tokens_seen": 889996216,
      "step": 14840,
      "train_runtime": 190542.6664,
      "train_tokens_per_second": 4670.85
    },
    {
      "epoch": 1.8934948979591837,
      "grad_norm": 0.7085706373965491,
      "learning_rate": 6.99752707087542e-08,
      "loss": 0.25458741188049316,
      "num_input_tokens_seen": 890302536,
      "step": 14845,
      "train_runtime": 190608.7403,
      "train_tokens_per_second": 4670.838
    },
    {
      "epoch": 1.8941326530612246,
      "grad_norm": 0.7101358935326122,
      "learning_rate": 6.91426777457288e-08,
      "loss": 0.26941733360290526,
      "num_input_tokens_seen": 890607864,
      "step": 14850,
      "train_runtime": 190676.7186,
      "train_tokens_per_second": 4670.774
    },
    {
      "epoch": 1.8947704081632653,
      "grad_norm": 0.6953427813899536,
      "learning_rate": 6.831503324239707e-08,
      "loss": 0.24118461608886718,
      "num_input_tokens_seen": 890907856,
      "step": 14855,
      "train_runtime": 190730.2153,
      "train_tokens_per_second": 4671.037
    },
    {
      "epoch": 1.8954081632653061,
      "grad_norm": 0.999849336589608,
      "learning_rate": 6.749233802935684e-08,
      "loss": 0.22293548583984374,
      "num_input_tokens_seen": 891195896,
      "step": 14860,
      "train_runtime": 190793.986,
      "train_tokens_per_second": 4670.985
    },
    {
      "epoch": 1.896045918367347,
      "grad_norm": 0.5972988403218099,
      "learning_rate": 6.667459293224155e-08,
      "loss": 0.22655868530273438,
      "num_input_tokens_seen": 891502736,
      "step": 14865,
      "train_runtime": 190855.7123,
      "train_tokens_per_second": 4671.082
    },
    {
      "epoch": 1.8966836734693877,
      "grad_norm": 0.6875299488691295,
      "learning_rate": 6.586179877171473e-08,
      "loss": 0.23338775634765624,
      "num_input_tokens_seen": 891806256,
      "step": 14870,
      "train_runtime": 190925.9032,
      "train_tokens_per_second": 4670.955
    },
    {
      "epoch": 1.8973214285714286,
      "grad_norm": 0.6580488873267832,
      "learning_rate": 6.505395636347222e-08,
      "loss": 0.26167232990264894,
      "num_input_tokens_seen": 892109288,
      "step": 14875,
      "train_runtime": 190993.3851,
      "train_tokens_per_second": 4670.891
    },
    {
      "epoch": 1.8979591836734695,
      "grad_norm": 0.738698676658821,
      "learning_rate": 6.425106651824054e-08,
      "loss": 0.23957352638244628,
      "num_input_tokens_seen": 892408864,
      "step": 14880,
      "train_runtime": 191060.084,
      "train_tokens_per_second": 4670.828
    },
    {
      "epoch": 1.8985969387755102,
      "grad_norm": 0.6194161475157497,
      "learning_rate": 6.345313004177511e-08,
      "loss": 0.23000946044921874,
      "num_input_tokens_seen": 892701112,
      "step": 14885,
      "train_runtime": 191120.4848,
      "train_tokens_per_second": 4670.881
    },
    {
      "epoch": 1.899234693877551,
      "grad_norm": 0.7407638762684772,
      "learning_rate": 6.266014773486207e-08,
      "loss": 0.2402566909790039,
      "num_input_tokens_seen": 892990744,
      "step": 14890,
      "train_runtime": 191182.511,
      "train_tokens_per_second": 4670.881
    },
    {
      "epoch": 1.899872448979592,
      "grad_norm": 0.62624988091952,
      "learning_rate": 6.187212039331314e-08,
      "loss": 0.25193309783935547,
      "num_input_tokens_seen": 893301904,
      "step": 14895,
      "train_runtime": 191255.442,
      "train_tokens_per_second": 4670.727
    },
    {
      "epoch": 1.9005102040816326,
      "grad_norm": 0.7579850216336217,
      "learning_rate": 6.108904880797017e-08,
      "loss": 0.2599358081817627,
      "num_input_tokens_seen": 893614944,
      "step": 14900,
      "train_runtime": 191308.3724,
      "train_tokens_per_second": 4671.071
    },
    {
      "epoch": 1.9011479591836735,
      "grad_norm": 0.6862768217826825,
      "learning_rate": 6.031093376469899e-08,
      "loss": 0.250386381149292,
      "num_input_tokens_seen": 893909080,
      "step": 14905,
      "train_runtime": 191380.9808,
      "train_tokens_per_second": 4670.836
    },
    {
      "epoch": 1.9017857142857144,
      "grad_norm": 0.6631016851451885,
      "learning_rate": 5.9537776044393256e-08,
      "loss": 0.2259291172027588,
      "num_input_tokens_seen": 894216832,
      "step": 14910,
      "train_runtime": 191447.6052,
      "train_tokens_per_second": 4670.818
    },
    {
      "epoch": 1.902423469387755,
      "grad_norm": 0.6027942096396334,
      "learning_rate": 5.87695764229701e-08,
      "loss": 0.2525123834609985,
      "num_input_tokens_seen": 894522600,
      "step": 14915,
      "train_runtime": 191512.1418,
      "train_tokens_per_second": 4670.84
    },
    {
      "epoch": 1.9030612244897958,
      "grad_norm": 0.6092285515600385,
      "learning_rate": 5.80063356713717e-08,
      "loss": 0.24264698028564452,
      "num_input_tokens_seen": 894826800,
      "step": 14920,
      "train_runtime": 191580.1686,
      "train_tokens_per_second": 4670.769
    },
    {
      "epoch": 1.9036989795918369,
      "grad_norm": 0.6588368285104348,
      "learning_rate": 5.7248054555563704e-08,
      "loss": 0.24944958686828614,
      "num_input_tokens_seen": 895116520,
      "step": 14925,
      "train_runtime": 191638.7761,
      "train_tokens_per_second": 4670.853
    },
    {
      "epoch": 1.9043367346938775,
      "grad_norm": 0.6477524467877152,
      "learning_rate": 5.6494733836534035e-08,
      "loss": 0.225768518447876,
      "num_input_tokens_seen": 895401512,
      "step": 14930,
      "train_runtime": 191699.6981,
      "train_tokens_per_second": 4670.855
    },
    {
      "epoch": 1.9049744897959182,
      "grad_norm": 0.6588457990258274,
      "learning_rate": 5.574637427029239e-08,
      "loss": 0.22354941368103026,
      "num_input_tokens_seen": 895700776,
      "step": 14935,
      "train_runtime": 191769.6542,
      "train_tokens_per_second": 4670.712
    },
    {
      "epoch": 1.9056122448979593,
      "grad_norm": 0.6939714517067077,
      "learning_rate": 5.50029766078708e-08,
      "loss": 0.2440859317779541,
      "num_input_tokens_seen": 895998240,
      "step": 14940,
      "train_runtime": 191845.4352,
      "train_tokens_per_second": 4670.417
    },
    {
      "epoch": 1.90625,
      "grad_norm": 0.566132980534101,
      "learning_rate": 5.426454159531913e-08,
      "loss": 0.2135913610458374,
      "num_input_tokens_seen": 896314136,
      "step": 14945,
      "train_runtime": 191911.839,
      "train_tokens_per_second": 4670.447
    },
    {
      "epoch": 1.9068877551020407,
      "grad_norm": 0.6906905587473503,
      "learning_rate": 5.3531069973709607e-08,
      "loss": 0.22824203968048096,
      "num_input_tokens_seen": 896611744,
      "step": 14950,
      "train_runtime": 191980.4463,
      "train_tokens_per_second": 4670.328
    },
    {
      "epoch": 1.9075255102040818,
      "grad_norm": 0.6372209541472462,
      "learning_rate": 5.280256247913229e-08,
      "loss": 0.2516682863235474,
      "num_input_tokens_seen": 896937600,
      "step": 14955,
      "train_runtime": 192052.6662,
      "train_tokens_per_second": 4670.269
    },
    {
      "epoch": 1.9081632653061225,
      "grad_norm": 0.6464146872929862,
      "learning_rate": 5.207901984269459e-08,
      "loss": 0.23169450759887694,
      "num_input_tokens_seen": 897243616,
      "step": 14960,
      "train_runtime": 192118.1025,
      "train_tokens_per_second": 4670.271
    },
    {
      "epoch": 1.9088010204081631,
      "grad_norm": 0.6440297794914668,
      "learning_rate": 5.136044279052288e-08,
      "loss": 0.25643513202667234,
      "num_input_tokens_seen": 897550240,
      "step": 14965,
      "train_runtime": 192187.8115,
      "train_tokens_per_second": 4670.173
    },
    {
      "epoch": 1.9094387755102042,
      "grad_norm": 0.6749457211671639,
      "learning_rate": 5.0646832043758645e-08,
      "loss": 0.24692878723144532,
      "num_input_tokens_seen": 897847152,
      "step": 14970,
      "train_runtime": 192258.5077,
      "train_tokens_per_second": 4670.0
    },
    {
      "epoch": 1.910076530612245,
      "grad_norm": 0.6478322305302919,
      "learning_rate": 4.993818831856123e-08,
      "loss": 0.23909859657287597,
      "num_input_tokens_seen": 898131880,
      "step": 14975,
      "train_runtime": 192320.6567,
      "train_tokens_per_second": 4669.971
    },
    {
      "epoch": 1.9107142857142856,
      "grad_norm": 0.7018693541826305,
      "learning_rate": 4.923451232610288e-08,
      "loss": 0.25514373779296873,
      "num_input_tokens_seen": 898427688,
      "step": 14980,
      "train_runtime": 192380.6107,
      "train_tokens_per_second": 4670.053
    },
    {
      "epoch": 1.9113520408163265,
      "grad_norm": 0.6564444124647395,
      "learning_rate": 4.8535804772572024e-08,
      "loss": 0.2652846336364746,
      "num_input_tokens_seen": 898728800,
      "step": 14985,
      "train_runtime": 192443.2259,
      "train_tokens_per_second": 4670.098
    },
    {
      "epoch": 1.9119897959183674,
      "grad_norm": 0.6716127219950674,
      "learning_rate": 4.7842066359170544e-08,
      "loss": 0.25480942726135253,
      "num_input_tokens_seen": 899027712,
      "step": 14990,
      "train_runtime": 192510.9509,
      "train_tokens_per_second": 4670.008
    },
    {
      "epoch": 1.912627551020408,
      "grad_norm": 0.674740294633714,
      "learning_rate": 4.715329778211375e-08,
      "loss": 0.23373541831970215,
      "num_input_tokens_seen": 899326320,
      "step": 14995,
      "train_runtime": 192579.882,
      "train_tokens_per_second": 4669.887
    },
    {
      "epoch": 1.913265306122449,
      "grad_norm": 0.6340465402770782,
      "learning_rate": 4.646949973262871e-08,
      "loss": 0.2550114870071411,
      "num_input_tokens_seen": 899624928,
      "step": 15000,
      "train_runtime": 192640.1618,
      "train_tokens_per_second": 4669.976
    },
    {
      "epoch": 1.9139030612244898,
      "grad_norm": 0.6207776252521124,
      "learning_rate": 4.579067289695427e-08,
      "loss": 0.24618887901306152,
      "num_input_tokens_seen": 899911880,
      "step": 15005,
      "train_runtime": 192699.2674,
      "train_tokens_per_second": 4670.033
    },
    {
      "epoch": 1.9145408163265305,
      "grad_norm": 0.7526401101813818,
      "learning_rate": 4.511681795634049e-08,
      "loss": 0.22883908748626708,
      "num_input_tokens_seen": 900200640,
      "step": 15010,
      "train_runtime": 192767.0538,
      "train_tokens_per_second": 4669.888
    },
    {
      "epoch": 1.9151785714285714,
      "grad_norm": 0.6939288258041673,
      "learning_rate": 4.4447935587049185e-08,
      "loss": 0.23292334079742433,
      "num_input_tokens_seen": 900501656,
      "step": 15015,
      "train_runtime": 192825.7253,
      "train_tokens_per_second": 4670.029
    },
    {
      "epoch": 1.9158163265306123,
      "grad_norm": 0.645913207816087,
      "learning_rate": 4.37840264603484e-08,
      "loss": 0.24563350677490234,
      "num_input_tokens_seen": 900798080,
      "step": 15020,
      "train_runtime": 192883.2208,
      "train_tokens_per_second": 4670.173
    },
    {
      "epoch": 1.916454081632653,
      "grad_norm": 0.7211337876438396,
      "learning_rate": 4.312509124251907e-08,
      "loss": 0.25634796619415284,
      "num_input_tokens_seen": 901104512,
      "step": 15025,
      "train_runtime": 192954.1675,
      "train_tokens_per_second": 4670.044
    },
    {
      "epoch": 1.9170918367346939,
      "grad_norm": 0.6328512646454919,
      "learning_rate": 4.247113059484775e-08,
      "loss": 0.23440108299255372,
      "num_input_tokens_seen": 901396168,
      "step": 15030,
      "train_runtime": 193013.4268,
      "train_tokens_per_second": 4670.122
    },
    {
      "epoch": 1.9177295918367347,
      "grad_norm": 0.6326676759399048,
      "learning_rate": 4.1822145173630033e-08,
      "loss": 0.2312554359436035,
      "num_input_tokens_seen": 901702576,
      "step": 15035,
      "train_runtime": 193069.9139,
      "train_tokens_per_second": 4670.342
    },
    {
      "epoch": 1.9183673469387754,
      "grad_norm": 0.6059941073498739,
      "learning_rate": 4.117813563016826e-08,
      "loss": 0.23285727500915526,
      "num_input_tokens_seen": 901999496,
      "step": 15040,
      "train_runtime": 193132.6869,
      "train_tokens_per_second": 4670.362
    },
    {
      "epoch": 1.9190051020408163,
      "grad_norm": 0.5869284707146232,
      "learning_rate": 4.0539102610770434e-08,
      "loss": 0.2457603931427002,
      "num_input_tokens_seen": 902302920,
      "step": 15045,
      "train_runtime": 193201.8869,
      "train_tokens_per_second": 4670.259
    },
    {
      "epoch": 1.9196428571428572,
      "grad_norm": 0.5822091240848306,
      "learning_rate": 3.990504675675133e-08,
      "loss": 0.2333512306213379,
      "num_input_tokens_seen": 902607528,
      "step": 15050,
      "train_runtime": 193265.0667,
      "train_tokens_per_second": 4670.309
    },
    {
      "epoch": 1.9202806122448979,
      "grad_norm": 0.5942966343245862,
      "learning_rate": 3.927596870442973e-08,
      "loss": 0.2389463186264038,
      "num_input_tokens_seen": 902914504,
      "step": 15055,
      "train_runtime": 193323.4537,
      "train_tokens_per_second": 4670.486
    },
    {
      "epoch": 1.9209183673469388,
      "grad_norm": 0.5396680335059741,
      "learning_rate": 3.8651869085130076e-08,
      "loss": 0.24712717533111572,
      "num_input_tokens_seen": 903209392,
      "step": 15060,
      "train_runtime": 193390.8722,
      "train_tokens_per_second": 4670.383
    },
    {
      "epoch": 1.9215561224489797,
      "grad_norm": 0.6268102745983977,
      "learning_rate": 3.8032748525179684e-08,
      "loss": 0.22878117561340333,
      "num_input_tokens_seen": 903501080,
      "step": 15065,
      "train_runtime": 193456.8732,
      "train_tokens_per_second": 4670.297
    },
    {
      "epoch": 1.9221938775510203,
      "grad_norm": 0.6629009625921484,
      "learning_rate": 3.741860764590877e-08,
      "loss": 0.25066375732421875,
      "num_input_tokens_seen": 903822752,
      "step": 15070,
      "train_runtime": 193513.3293,
      "train_tokens_per_second": 4670.597
    },
    {
      "epoch": 1.9228316326530612,
      "grad_norm": 0.5795866770696644,
      "learning_rate": 3.680944706365097e-08,
      "loss": 0.22920804023742675,
      "num_input_tokens_seen": 904138144,
      "step": 15075,
      "train_runtime": 193572.0552,
      "train_tokens_per_second": 4670.809
    },
    {
      "epoch": 1.9234693877551021,
      "grad_norm": 0.6199219051878265,
      "learning_rate": 3.6205267389741724e-08,
      "loss": 0.230307674407959,
      "num_input_tokens_seen": 904440488,
      "step": 15080,
      "train_runtime": 193633.0442,
      "train_tokens_per_second": 4670.899
    },
    {
      "epoch": 1.9241071428571428,
      "grad_norm": 0.7081937235558688,
      "learning_rate": 3.560606923051768e-08,
      "loss": 0.2649749994277954,
      "num_input_tokens_seen": 904749440,
      "step": 15085,
      "train_runtime": 193691.7229,
      "train_tokens_per_second": 4671.08
    },
    {
      "epoch": 1.9247448979591837,
      "grad_norm": 0.6850531620583067,
      "learning_rate": 3.5011853187315035e-08,
      "loss": 0.232527494430542,
      "num_input_tokens_seen": 905040448,
      "step": 15090,
      "train_runtime": 193756.5747,
      "train_tokens_per_second": 4671.018
    },
    {
      "epoch": 1.9253826530612246,
      "grad_norm": 0.6649364094207673,
      "learning_rate": 3.442261985647177e-08,
      "loss": 0.25315489768981936,
      "num_input_tokens_seen": 905343480,
      "step": 15095,
      "train_runtime": 193819.7759,
      "train_tokens_per_second": 4671.058
    },
    {
      "epoch": 1.9260204081632653,
      "grad_norm": 0.7367104201106388,
      "learning_rate": 3.383836982932542e-08,
      "loss": 0.23107264041900635,
      "num_input_tokens_seen": 905632408,
      "step": 15100,
      "train_runtime": 193875.9655,
      "train_tokens_per_second": 4671.195
    },
    {
      "epoch": 1.9266581632653061,
      "grad_norm": 0.6422996191071108,
      "learning_rate": 3.325910369220975e-08,
      "loss": 0.2546488523483276,
      "num_input_tokens_seen": 905934184,
      "step": 15105,
      "train_runtime": 193947.093,
      "train_tokens_per_second": 4671.038
    },
    {
      "epoch": 1.927295918367347,
      "grad_norm": 0.6387982927830601,
      "learning_rate": 3.268482202646084e-08,
      "loss": 0.23939924240112304,
      "num_input_tokens_seen": 906237608,
      "step": 15110,
      "train_runtime": 194020.9974,
      "train_tokens_per_second": 4670.822
    },
    {
      "epoch": 1.9279336734693877,
      "grad_norm": 0.6809442061626129,
      "learning_rate": 3.211552540840934e-08,
      "loss": 0.2389169692993164,
      "num_input_tokens_seen": 906551528,
      "step": 15115,
      "train_runtime": 194093.7717,
      "train_tokens_per_second": 4670.688
    },
    {
      "epoch": 1.9285714285714286,
      "grad_norm": 0.6062182732420422,
      "learning_rate": 3.1551214409384355e-08,
      "loss": 0.2684735536575317,
      "num_input_tokens_seen": 906864488,
      "step": 15120,
      "train_runtime": 194147.0547,
      "train_tokens_per_second": 4671.019
    },
    {
      "epoch": 1.9292091836734695,
      "grad_norm": 0.6858638615391687,
      "learning_rate": 3.09918895957112e-08,
      "loss": 0.24612531661987305,
      "num_input_tokens_seen": 907132432,
      "step": 15125,
      "train_runtime": 194214.467,
      "train_tokens_per_second": 4670.777
    },
    {
      "epoch": 1.9298469387755102,
      "grad_norm": 0.688916621792478,
      "learning_rate": 3.0437551528711974e-08,
      "loss": 0.24551730155944823,
      "num_input_tokens_seen": 907423464,
      "step": 15130,
      "train_runtime": 194277.6119,
      "train_tokens_per_second": 4670.757
    },
    {
      "epoch": 1.930484693877551,
      "grad_norm": 0.6635125567256317,
      "learning_rate": 2.988820076470278e-08,
      "loss": 0.2580201864242554,
      "num_input_tokens_seen": 907719760,
      "step": 15135,
      "train_runtime": 194339.0798,
      "train_tokens_per_second": 4670.804
    },
    {
      "epoch": 1.931122448979592,
      "grad_norm": 0.814995636446256,
      "learning_rate": 2.934383785499595e-08,
      "loss": 0.23907341957092285,
      "num_input_tokens_seen": 908028760,
      "step": 15140,
      "train_runtime": 194403.9862,
      "train_tokens_per_second": 4670.834
    },
    {
      "epoch": 1.9317602040816326,
      "grad_norm": 0.665618420325683,
      "learning_rate": 2.880446334589837e-08,
      "loss": 0.2549652099609375,
      "num_input_tokens_seen": 908335864,
      "step": 15145,
      "train_runtime": 194466.2845,
      "train_tokens_per_second": 4670.917
    },
    {
      "epoch": 1.9323979591836735,
      "grad_norm": 0.640722233968815,
      "learning_rate": 2.8270077778708714e-08,
      "loss": 0.2548914909362793,
      "num_input_tokens_seen": 908623464,
      "step": 15150,
      "train_runtime": 194536.8756,
      "train_tokens_per_second": 4670.7
    },
    {
      "epoch": 1.9330357142857144,
      "grad_norm": 0.653741482098068,
      "learning_rate": 2.7740681689721327e-08,
      "loss": 0.2401275157928467,
      "num_input_tokens_seen": 908925248,
      "step": 15155,
      "train_runtime": 194603.3872,
      "train_tokens_per_second": 4670.655
    },
    {
      "epoch": 1.933673469387755,
      "grad_norm": 0.8538638590132563,
      "learning_rate": 2.721627561022122e-08,
      "loss": 0.25029151439666747,
      "num_input_tokens_seen": 909225024,
      "step": 15160,
      "train_runtime": 194664.7841,
      "train_tokens_per_second": 4670.722
    },
    {
      "epoch": 1.9343112244897958,
      "grad_norm": 0.6287202779561625,
      "learning_rate": 2.6696860066487417e-08,
      "loss": 0.21725854873657227,
      "num_input_tokens_seen": 909529280,
      "step": 15165,
      "train_runtime": 194721.458,
      "train_tokens_per_second": 4670.925
    },
    {
      "epoch": 1.9349489795918369,
      "grad_norm": 0.6622045332718416,
      "learning_rate": 2.6182435579789055e-08,
      "loss": 0.2391425371170044,
      "num_input_tokens_seen": 909837264,
      "step": 15170,
      "train_runtime": 194789.9301,
      "train_tokens_per_second": 4670.864
    },
    {
      "epoch": 1.9355867346938775,
      "grad_norm": 0.6455356673733826,
      "learning_rate": 2.567300266638706e-08,
      "loss": 0.25289885997772216,
      "num_input_tokens_seen": 910137832,
      "step": 15175,
      "train_runtime": 194862.8765,
      "train_tokens_per_second": 4670.658
    },
    {
      "epoch": 1.9362244897959182,
      "grad_norm": 0.6807299064078204,
      "learning_rate": 2.516856183753358e-08,
      "loss": 0.23154253959655763,
      "num_input_tokens_seen": 910455200,
      "step": 15180,
      "train_runtime": 194933.524,
      "train_tokens_per_second": 4670.593
    },
    {
      "epoch": 1.9368622448979593,
      "grad_norm": 0.7313470463205378,
      "learning_rate": 2.4669113599469774e-08,
      "loss": 0.24894938468933106,
      "num_input_tokens_seen": 910766760,
      "step": 15185,
      "train_runtime": 194984.5745,
      "train_tokens_per_second": 4670.968
    },
    {
      "epoch": 1.9375,
      "grad_norm": 0.6806718686332535,
      "learning_rate": 2.4174658453426368e-08,
      "loss": 0.25083222389221194,
      "num_input_tokens_seen": 911073496,
      "step": 15190,
      "train_runtime": 195050.2362,
      "train_tokens_per_second": 4670.968
    },
    {
      "epoch": 1.9381377551020407,
      "grad_norm": 0.6310448786168161,
      "learning_rate": 2.3685196895624206e-08,
      "loss": 0.23468017578125,
      "num_input_tokens_seen": 911382928,
      "step": 15195,
      "train_runtime": 195112.2691,
      "train_tokens_per_second": 4671.069
    },
    {
      "epoch": 1.9387755102040818,
      "grad_norm": 0.674200703449938,
      "learning_rate": 2.3200729417272028e-08,
      "loss": 0.22753286361694336,
      "num_input_tokens_seen": 911694488,
      "step": 15200,
      "train_runtime": 195183.338,
      "train_tokens_per_second": 4670.965
    },
    {
      "epoch": 1.9394132653061225,
      "grad_norm": 0.642863695602926,
      "learning_rate": 2.2721256504567026e-08,
      "loss": 0.25208315849304197,
      "num_input_tokens_seen": 911986016,
      "step": 15205,
      "train_runtime": 195252.5857,
      "train_tokens_per_second": 4670.801
    },
    {
      "epoch": 1.9400510204081631,
      "grad_norm": 0.6219130487996105,
      "learning_rate": 2.2246778638692623e-08,
      "loss": 0.2282183885574341,
      "num_input_tokens_seen": 912285512,
      "step": 15210,
      "train_runtime": 195315.6121,
      "train_tokens_per_second": 4670.827
    },
    {
      "epoch": 1.9406887755102042,
      "grad_norm": 0.6720021915213673,
      "learning_rate": 2.177729629582237e-08,
      "loss": 0.22368407249450684,
      "num_input_tokens_seen": 912594320,
      "step": 15215,
      "train_runtime": 195376.0522,
      "train_tokens_per_second": 4670.963
    },
    {
      "epoch": 1.941326530612245,
      "grad_norm": 0.5769656601013554,
      "learning_rate": 2.1312809947113267e-08,
      "loss": 0.21522626876831055,
      "num_input_tokens_seen": 912890168,
      "step": 15220,
      "train_runtime": 195437.2223,
      "train_tokens_per_second": 4671.015
    },
    {
      "epoch": 1.9419642857142856,
      "grad_norm": 0.6302180994971214,
      "learning_rate": 2.0853320058710214e-08,
      "loss": 0.23648781776428224,
      "num_input_tokens_seen": 913182776,
      "step": 15225,
      "train_runtime": 195494.0997,
      "train_tokens_per_second": 4671.153
    },
    {
      "epoch": 1.9426020408163265,
      "grad_norm": 0.7437601080428625,
      "learning_rate": 2.0398827091743234e-08,
      "loss": 0.2461167812347412,
      "num_input_tokens_seen": 913465992,
      "step": 15230,
      "train_runtime": 195559.0722,
      "train_tokens_per_second": 4671.049
    },
    {
      "epoch": 1.9432397959183674,
      "grad_norm": 0.6919033357778054,
      "learning_rate": 1.9949331502327475e-08,
      "loss": 0.22862133979797364,
      "num_input_tokens_seen": 913747936,
      "step": 15235,
      "train_runtime": 195621.9343,
      "train_tokens_per_second": 4670.989
    },
    {
      "epoch": 1.943877551020408,
      "grad_norm": 0.6223614709195145,
      "learning_rate": 1.950483374156431e-08,
      "loss": 0.253930401802063,
      "num_input_tokens_seen": 914049728,
      "step": 15240,
      "train_runtime": 195676.3182,
      "train_tokens_per_second": 4671.233
    },
    {
      "epoch": 1.944515306122449,
      "grad_norm": 0.6337973715833365,
      "learning_rate": 1.9065334255536916e-08,
      "loss": 0.24294543266296387,
      "num_input_tokens_seen": 914343256,
      "step": 15245,
      "train_runtime": 195744.5468,
      "train_tokens_per_second": 4671.105
    },
    {
      "epoch": 1.9451530612244898,
      "grad_norm": 0.6210057272388032,
      "learning_rate": 1.8630833485314693e-08,
      "loss": 0.2540744304656982,
      "num_input_tokens_seen": 914645640,
      "step": 15250,
      "train_runtime": 195809.6772,
      "train_tokens_per_second": 4671.095
    },
    {
      "epoch": 1.9457908163265305,
      "grad_norm": 0.6715394091560279,
      "learning_rate": 1.8201331866948834e-08,
      "loss": 0.23338706493377687,
      "num_input_tokens_seen": 914941704,
      "step": 15255,
      "train_runtime": 195874.8375,
      "train_tokens_per_second": 4671.053
    },
    {
      "epoch": 1.9464285714285714,
      "grad_norm": 0.6634502906165924,
      "learning_rate": 1.7776829831474553e-08,
      "loss": 0.24994971752166747,
      "num_input_tokens_seen": 915241552,
      "step": 15260,
      "train_runtime": 195942.347,
      "train_tokens_per_second": 4670.974
    },
    {
      "epoch": 1.9470663265306123,
      "grad_norm": 0.6218536804696106,
      "learning_rate": 1.735732780490884e-08,
      "loss": 0.24652099609375,
      "num_input_tokens_seen": 915547064,
      "step": 15265,
      "train_runtime": 195991.8415,
      "train_tokens_per_second": 4671.353
    },
    {
      "epoch": 1.947704081632653,
      "grad_norm": 0.6663908818224021,
      "learning_rate": 1.694282620825216e-08,
      "loss": 0.22615201473236085,
      "num_input_tokens_seen": 915860168,
      "step": 15270,
      "train_runtime": 196050.4943,
      "train_tokens_per_second": 4671.552
    },
    {
      "epoch": 1.9483418367346939,
      "grad_norm": 0.6377532593066776,
      "learning_rate": 1.6533325457485093e-08,
      "loss": 0.24736628532409669,
      "num_input_tokens_seen": 916178968,
      "step": 15275,
      "train_runtime": 196109.351,
      "train_tokens_per_second": 4671.776
    },
    {
      "epoch": 1.9489795918367347,
      "grad_norm": 0.6079270441622019,
      "learning_rate": 1.612882596357057e-08,
      "loss": 0.2625592231750488,
      "num_input_tokens_seen": 916474696,
      "step": 15280,
      "train_runtime": 196175.2937,
      "train_tokens_per_second": 4671.713
    },
    {
      "epoch": 1.9496173469387754,
      "grad_norm": 0.7086343688070547,
      "learning_rate": 1.5729328132452203e-08,
      "loss": 0.23019585609436036,
      "num_input_tokens_seen": 916767440,
      "step": 15285,
      "train_runtime": 196233.0518,
      "train_tokens_per_second": 4671.83
    },
    {
      "epoch": 1.9502551020408163,
      "grad_norm": 0.6864295940398258,
      "learning_rate": 1.533483236505373e-08,
      "loss": 0.23982906341552734,
      "num_input_tokens_seen": 917065400,
      "step": 15290,
      "train_runtime": 196290.197,
      "train_tokens_per_second": 4671.988
    },
    {
      "epoch": 1.9508928571428572,
      "grad_norm": 0.8042241709134048,
      "learning_rate": 1.494533905727902e-08,
      "loss": 0.24926409721374512,
      "num_input_tokens_seen": 917347400,
      "step": 15295,
      "train_runtime": 196357.3383,
      "train_tokens_per_second": 4671.826
    },
    {
      "epoch": 1.9515306122448979,
      "grad_norm": 0.6859727032503002,
      "learning_rate": 1.4560848600012056e-08,
      "loss": 0.22752974033355713,
      "num_input_tokens_seen": 917652104,
      "step": 15300,
      "train_runtime": 196423.9354,
      "train_tokens_per_second": 4671.794
    },
    {
      "epoch": 1.9521683673469388,
      "grad_norm": 0.6582219909764592,
      "learning_rate": 1.4181361379115854e-08,
      "loss": 0.2244406223297119,
      "num_input_tokens_seen": 917948512,
      "step": 15305,
      "train_runtime": 196486.1865,
      "train_tokens_per_second": 4671.822
    },
    {
      "epoch": 1.9528061224489797,
      "grad_norm": 0.614642962700177,
      "learning_rate": 1.3806877775432436e-08,
      "loss": 0.23401083946228027,
      "num_input_tokens_seen": 918255864,
      "step": 15310,
      "train_runtime": 196553.3965,
      "train_tokens_per_second": 4671.788
    },
    {
      "epoch": 1.9534438775510203,
      "grad_norm": 0.66017110524275,
      "learning_rate": 1.3437398164781734e-08,
      "loss": 0.24059834480285644,
      "num_input_tokens_seen": 918556096,
      "step": 15315,
      "train_runtime": 196611.9627,
      "train_tokens_per_second": 4671.924
    },
    {
      "epoch": 1.9540816326530612,
      "grad_norm": 0.7409477782750582,
      "learning_rate": 1.3072922917963249e-08,
      "loss": 0.2520226001739502,
      "num_input_tokens_seen": 918838080,
      "step": 15320,
      "train_runtime": 196668.993,
      "train_tokens_per_second": 4672.003
    },
    {
      "epoch": 1.9547193877551021,
      "grad_norm": 0.6710534884970901,
      "learning_rate": 1.2713452400752724e-08,
      "loss": 0.23367981910705565,
      "num_input_tokens_seen": 919142800,
      "step": 15325,
      "train_runtime": 196733.5083,
      "train_tokens_per_second": 4672.02
    },
    {
      "epoch": 1.9553571428571428,
      "grad_norm": 0.6531668013394583,
      "learning_rate": 1.2358986973903807e-08,
      "loss": 0.2511638641357422,
      "num_input_tokens_seen": 919424568,
      "step": 15330,
      "train_runtime": 196795.3707,
      "train_tokens_per_second": 4671.983
    },
    {
      "epoch": 1.9559948979591837,
      "grad_norm": 0.6896304841409016,
      "learning_rate": 1.2009526993147503e-08,
      "loss": 0.25591254234313965,
      "num_input_tokens_seen": 919716328,
      "step": 15335,
      "train_runtime": 196859.8654,
      "train_tokens_per_second": 4671.934
    },
    {
      "epoch": 1.9566326530612246,
      "grad_norm": 0.6822575203236972,
      "learning_rate": 1.1665072809191602e-08,
      "loss": 0.2302922487258911,
      "num_input_tokens_seen": 920002952,
      "step": 15340,
      "train_runtime": 196930.711,
      "train_tokens_per_second": 4671.709
    },
    {
      "epoch": 1.9572704081632653,
      "grad_norm": 0.5860022280306675,
      "learning_rate": 1.132562476771959e-08,
      "loss": 0.24354929924011232,
      "num_input_tokens_seen": 920303216,
      "step": 15345,
      "train_runtime": 196991.6172,
      "train_tokens_per_second": 4671.789
    },
    {
      "epoch": 1.9579081632653061,
      "grad_norm": 0.6499682638962776,
      "learning_rate": 1.0991183209391188e-08,
      "loss": 0.24647674560546876,
      "num_input_tokens_seen": 920605432,
      "step": 15350,
      "train_runtime": 197055.1354,
      "train_tokens_per_second": 4671.816
    },
    {
      "epoch": 1.958545918367347,
      "grad_norm": 0.6286218716370053,
      "learning_rate": 1.0661748469842359e-08,
      "loss": 0.25051705837249755,
      "num_input_tokens_seen": 920908424,
      "step": 15355,
      "train_runtime": 197109.8827,
      "train_tokens_per_second": 4672.056
    },
    {
      "epoch": 1.9591836734693877,
      "grad_norm": 0.6629175399799272,
      "learning_rate": 1.0337320879683087e-08,
      "loss": 0.25404863357543944,
      "num_input_tokens_seen": 921225616,
      "step": 15360,
      "train_runtime": 197177.1406,
      "train_tokens_per_second": 4672.071
    },
    {
      "epoch": 1.9598214285714286,
      "grad_norm": 0.6129552783441397,
      "learning_rate": 1.0017900764500155e-08,
      "loss": 0.22357313632965087,
      "num_input_tokens_seen": 921526472,
      "step": 15365,
      "train_runtime": 197245.7374,
      "train_tokens_per_second": 4671.972
    },
    {
      "epoch": 1.9604591836734695,
      "grad_norm": 0.6482298341541117,
      "learning_rate": 9.703488444853248e-09,
      "loss": 0.2499098300933838,
      "num_input_tokens_seen": 921837576,
      "step": 15370,
      "train_runtime": 197306.159,
      "train_tokens_per_second": 4672.118
    },
    {
      "epoch": 1.9610969387755102,
      "grad_norm": 0.6262291708987566,
      "learning_rate": 9.39408423627719e-09,
      "loss": 0.24408447742462158,
      "num_input_tokens_seen": 922150280,
      "step": 15375,
      "train_runtime": 197373.1285,
      "train_tokens_per_second": 4672.117
    },
    {
      "epoch": 1.961734693877551,
      "grad_norm": 0.6982328131491065,
      "learning_rate": 9.089688449280265e-09,
      "loss": 0.2181765079498291,
      "num_input_tokens_seen": 922423208,
      "step": 15380,
      "train_runtime": 197444.0008,
      "train_tokens_per_second": 4671.822
    },
    {
      "epoch": 1.962372448979592,
      "grad_norm": 0.7360925768701235,
      "learning_rate": 8.79030138934589e-09,
      "loss": 0.25997095108032225,
      "num_input_tokens_seen": 922728704,
      "step": 15385,
      "train_runtime": 197508.3757,
      "train_tokens_per_second": 4671.846
    },
    {
      "epoch": 1.9630102040816326,
      "grad_norm": 0.6554598098231311,
      "learning_rate": 8.49592335692928e-09,
      "loss": 0.22874805927276612,
      "num_input_tokens_seen": 923043512,
      "step": 15390,
      "train_runtime": 197575.4255,
      "train_tokens_per_second": 4671.854
    },
    {
      "epoch": 1.9636479591836735,
      "grad_norm": 0.657567910295264,
      "learning_rate": 8.206554647459675e-09,
      "loss": 0.22446055412292482,
      "num_input_tokens_seen": 923337928,
      "step": 15395,
      "train_runtime": 197629.3283,
      "train_tokens_per_second": 4672.069
    },
    {
      "epoch": 1.9642857142857144,
      "grad_norm": 0.5868118437814902,
      "learning_rate": 7.922195551338107e-09,
      "loss": 0.24218473434448243,
      "num_input_tokens_seen": 923636376,
      "step": 15400,
      "train_runtime": 197692.0939,
      "train_tokens_per_second": 4672.096
    },
    {
      "epoch": 1.964923469387755,
      "grad_norm": 0.6584794701314968,
      "learning_rate": 7.642846353939637e-09,
      "loss": 0.25373034477233886,
      "num_input_tokens_seen": 923946456,
      "step": 15405,
      "train_runtime": 197764.7228,
      "train_tokens_per_second": 4671.948
    },
    {
      "epoch": 1.9655612244897958,
      "grad_norm": 0.6229709246682923,
      "learning_rate": 7.368507335610009e-09,
      "loss": 0.22513649463653565,
      "num_input_tokens_seen": 924260824,
      "step": 15410,
      "train_runtime": 197837.2885,
      "train_tokens_per_second": 4671.823
    },
    {
      "epoch": 1.9661989795918369,
      "grad_norm": 0.6721020512127416,
      "learning_rate": 7.099178771668436e-09,
      "loss": 0.24282946586608886,
      "num_input_tokens_seen": 924556432,
      "step": 15415,
      "train_runtime": 197891.742,
      "train_tokens_per_second": 4672.031
    },
    {
      "epoch": 1.9668367346938775,
      "grad_norm": 0.6566852368029317,
      "learning_rate": 6.834860932403709e-09,
      "loss": 0.23992621898651123,
      "num_input_tokens_seen": 924853080,
      "step": 15420,
      "train_runtime": 197956.4709,
      "train_tokens_per_second": 4672.002
    },
    {
      "epoch": 1.9674744897959182,
      "grad_norm": 0.7169766079391501,
      "learning_rate": 6.575554083078084e-09,
      "loss": 0.23851332664489747,
      "num_input_tokens_seen": 925157776,
      "step": 15425,
      "train_runtime": 198025.9657,
      "train_tokens_per_second": 4671.901
    },
    {
      "epoch": 1.9681122448979593,
      "grad_norm": 0.7219266135535019,
      "learning_rate": 6.321258483924508e-09,
      "loss": 0.26246089935302735,
      "num_input_tokens_seen": 925452872,
      "step": 15430,
      "train_runtime": 198096.1079,
      "train_tokens_per_second": 4671.737
    },
    {
      "epoch": 1.96875,
      "grad_norm": 0.649961248925387,
      "learning_rate": 6.071974390144953e-09,
      "loss": 0.24496083259582518,
      "num_input_tokens_seen": 925750224,
      "step": 15435,
      "train_runtime": 198155.8596,
      "train_tokens_per_second": 4671.829
    },
    {
      "epoch": 1.9693877551020407,
      "grad_norm": 0.6554683356430376,
      "learning_rate": 5.827702051914852e-09,
      "loss": 0.23990488052368164,
      "num_input_tokens_seen": 926041384,
      "step": 15440,
      "train_runtime": 198216.2995,
      "train_tokens_per_second": 4671.873
    },
    {
      "epoch": 1.9700255102040818,
      "grad_norm": 0.654062922493736,
      "learning_rate": 5.5884417143775574e-09,
      "loss": 0.22308700084686278,
      "num_input_tokens_seen": 926329424,
      "step": 15445,
      "train_runtime": 198279.4839,
      "train_tokens_per_second": 4671.837
    },
    {
      "epoch": 1.9706632653061225,
      "grad_norm": 0.6365872097156556,
      "learning_rate": 5.354193617648218e-09,
      "loss": 0.24755873680114746,
      "num_input_tokens_seen": 926628136,
      "step": 15450,
      "train_runtime": 198341.7408,
      "train_tokens_per_second": 4671.877
    },
    {
      "epoch": 1.9713010204081631,
      "grad_norm": 0.7283488730178487,
      "learning_rate": 5.124957996811009e-09,
      "loss": 0.23184123039245605,
      "num_input_tokens_seen": 926921008,
      "step": 15455,
      "train_runtime": 198398.5706,
      "train_tokens_per_second": 4672.015
    },
    {
      "epoch": 1.9719387755102042,
      "grad_norm": 0.6616142170700493,
      "learning_rate": 4.900735081919128e-09,
      "loss": 0.24162561893463136,
      "num_input_tokens_seen": 927227984,
      "step": 15460,
      "train_runtime": 198465.786,
      "train_tokens_per_second": 4671.979
    },
    {
      "epoch": 1.972576530612245,
      "grad_norm": 0.7143983122777157,
      "learning_rate": 4.6815250979970195e-09,
      "loss": 0.24975261688232422,
      "num_input_tokens_seen": 927535288,
      "step": 15465,
      "train_runtime": 198535.1702,
      "train_tokens_per_second": 4671.894
    },
    {
      "epoch": 1.9732142857142856,
      "grad_norm": 0.6474965141272541,
      "learning_rate": 4.467328265037041e-09,
      "loss": 0.24119699001312256,
      "num_input_tokens_seen": 927841632,
      "step": 15470,
      "train_runtime": 198597.5761,
      "train_tokens_per_second": 4671.969
    },
    {
      "epoch": 1.9738520408163265,
      "grad_norm": 0.6285616866725054,
      "learning_rate": 4.258144798000019e-09,
      "loss": 0.2362743377685547,
      "num_input_tokens_seen": 928138384,
      "step": 15475,
      "train_runtime": 198654.2027,
      "train_tokens_per_second": 4672.131
    },
    {
      "epoch": 1.9744897959183674,
      "grad_norm": 0.6901046959151206,
      "learning_rate": 4.05397490681636e-09,
      "loss": 0.23540866374969482,
      "num_input_tokens_seen": 928443960,
      "step": 15480,
      "train_runtime": 198723.0405,
      "train_tokens_per_second": 4672.05
    },
    {
      "epoch": 1.975127551020408,
      "grad_norm": 0.6897354113635594,
      "learning_rate": 3.854818796385495e-09,
      "loss": 0.22936525344848632,
      "num_input_tokens_seen": 928722664,
      "step": 15485,
      "train_runtime": 198788.4506,
      "train_tokens_per_second": 4671.915
    },
    {
      "epoch": 1.975765306122449,
      "grad_norm": 0.5984417283572043,
      "learning_rate": 3.660676666573659e-09,
      "loss": 0.24026241302490234,
      "num_input_tokens_seen": 929023592,
      "step": 15490,
      "train_runtime": 198854.1961,
      "train_tokens_per_second": 4671.883
    },
    {
      "epoch": 1.9764030612244898,
      "grad_norm": 0.6105615965645449,
      "learning_rate": 3.4715487122161107e-09,
      "loss": 0.23082895278930665,
      "num_input_tokens_seen": 929326440,
      "step": 15495,
      "train_runtime": 198922.1289,
      "train_tokens_per_second": 4671.81
    },
    {
      "epoch": 1.9770408163265305,
      "grad_norm": 0.6720123345499762,
      "learning_rate": 3.287435123116578e-09,
      "loss": 0.2418750286102295,
      "num_input_tokens_seen": 929634512,
      "step": 15500,
      "train_runtime": 198989.5209,
      "train_tokens_per_second": 4671.776
    },
    {
      "epoch": 1.9776785714285714,
      "grad_norm": 0.5643089157852808,
      "learning_rate": 3.1083360840455935e-09,
      "loss": 0.24054617881774903,
      "num_input_tokens_seen": 929938048,
      "step": 15505,
      "train_runtime": 199054.5814,
      "train_tokens_per_second": 4671.774
    },
    {
      "epoch": 1.9783163265306123,
      "grad_norm": 0.6910800941279635,
      "learning_rate": 2.9342517747410482e-09,
      "loss": 0.2343662738800049,
      "num_input_tokens_seen": 930244016,
      "step": 15510,
      "train_runtime": 199122.538,
      "train_tokens_per_second": 4671.716
    },
    {
      "epoch": 1.978954081632653,
      "grad_norm": 0.8194130967587886,
      "learning_rate": 2.7651823699093027e-09,
      "loss": 0.25220522880554197,
      "num_input_tokens_seen": 930548096,
      "step": 15515,
      "train_runtime": 199182.2168,
      "train_tokens_per_second": 4671.843
    },
    {
      "epoch": 1.9795918367346939,
      "grad_norm": 0.6444734667559321,
      "learning_rate": 2.6011280392235215e-09,
      "loss": 0.24388303756713867,
      "num_input_tokens_seen": 930848856,
      "step": 15520,
      "train_runtime": 199241.7599,
      "train_tokens_per_second": 4671.957
    },
    {
      "epoch": 1.9802295918367347,
      "grad_norm": 0.6709665440148682,
      "learning_rate": 2.442088947323118e-09,
      "loss": 0.24142193794250488,
      "num_input_tokens_seen": 931124408,
      "step": 15525,
      "train_runtime": 199304.9027,
      "train_tokens_per_second": 4671.859
    },
    {
      "epoch": 1.9808673469387754,
      "grad_norm": 0.7106841627245938,
      "learning_rate": 2.288065253814864e-09,
      "loss": 0.2382129669189453,
      "num_input_tokens_seen": 931429872,
      "step": 15530,
      "train_runtime": 199369.6505,
      "train_tokens_per_second": 4671.874
    },
    {
      "epoch": 1.9815051020408163,
      "grad_norm": 0.6747427447210661,
      "learning_rate": 2.139057113272891e-09,
      "loss": 0.24044766426086425,
      "num_input_tokens_seen": 931723432,
      "step": 15535,
      "train_runtime": 199434.3705,
      "train_tokens_per_second": 4671.83
    },
    {
      "epoch": 1.9821428571428572,
      "grad_norm": 0.6645292262642306,
      "learning_rate": 1.995064675237024e-09,
      "loss": 0.23690540790557862,
      "num_input_tokens_seen": 932015016,
      "step": 15540,
      "train_runtime": 199495.9766,
      "train_tokens_per_second": 4671.849
    },
    {
      "epoch": 1.9827806122448979,
      "grad_norm": 0.6428403076400457,
      "learning_rate": 1.8560880842133366e-09,
      "loss": 0.2564822196960449,
      "num_input_tokens_seen": 932321664,
      "step": 15545,
      "train_runtime": 199561.703,
      "train_tokens_per_second": 4671.847
    },
    {
      "epoch": 1.9834183673469388,
      "grad_norm": 0.6590699643792678,
      "learning_rate": 1.7221274796752618e-09,
      "loss": 0.24346504211425782,
      "num_input_tokens_seen": 932613040,
      "step": 15550,
      "train_runtime": 199626.3384,
      "train_tokens_per_second": 4671.794
    },
    {
      "epoch": 1.9840561224489797,
      "grad_norm": 0.5929330109388921,
      "learning_rate": 1.5931829960608158e-09,
      "loss": 0.21612107753753662,
      "num_input_tokens_seen": 932901632,
      "step": 15555,
      "train_runtime": 199691.1884,
      "train_tokens_per_second": 4671.722
    },
    {
      "epoch": 1.9846938775510203,
      "grad_norm": 0.6927753327683915,
      "learning_rate": 1.469254762775374e-09,
      "loss": 0.23641691207885743,
      "num_input_tokens_seen": 933200712,
      "step": 15560,
      "train_runtime": 199762.3402,
      "train_tokens_per_second": 4671.555
    },
    {
      "epoch": 1.9853316326530612,
      "grad_norm": 0.8471620771249144,
      "learning_rate": 1.350342904188895e-09,
      "loss": 0.22833261489868165,
      "num_input_tokens_seen": 933486000,
      "step": 15565,
      "train_runtime": 199831.7859,
      "train_tokens_per_second": 4671.359
    },
    {
      "epoch": 1.9859693877551021,
      "grad_norm": 0.6521928183363583,
      "learning_rate": 1.2364475396386966e-09,
      "loss": 0.24839613437652588,
      "num_input_tokens_seen": 933799240,
      "step": 15570,
      "train_runtime": 199899.9354,
      "train_tokens_per_second": 4671.333
    },
    {
      "epoch": 1.9866071428571428,
      "grad_norm": 0.5994256520871023,
      "learning_rate": 1.1275687834266803e-09,
      "loss": 0.24686167240142823,
      "num_input_tokens_seen": 934119000,
      "step": 15575,
      "train_runtime": 199961.015,
      "train_tokens_per_second": 4671.506
    },
    {
      "epoch": 1.9872448979591837,
      "grad_norm": 0.6513508678236617,
      "learning_rate": 1.0237067448193306e-09,
      "loss": 0.23922209739685057,
      "num_input_tokens_seen": 934422016,
      "step": 15580,
      "train_runtime": 200036.8427,
      "train_tokens_per_second": 4671.25
    },
    {
      "epoch": 1.9878826530612246,
      "grad_norm": 0.7236862794639907,
      "learning_rate": 9.248615280499362e-10,
      "loss": 0.2451181411743164,
      "num_input_tokens_seen": 934722472,
      "step": 15585,
      "train_runtime": 200094.6839,
      "train_tokens_per_second": 4671.401
    },
    {
      "epoch": 1.9885204081632653,
      "grad_norm": 0.6585262786305752,
      "learning_rate": 8.310332323169246e-10,
      "loss": 0.22677361965179443,
      "num_input_tokens_seen": 935021672,
      "step": 15590,
      "train_runtime": 200164.9187,
      "train_tokens_per_second": 4671.256
    },
    {
      "epoch": 1.9891581632653061,
      "grad_norm": 0.6565613266627721,
      "learning_rate": 7.422219517833062e-10,
      "loss": 0.21729130744934083,
      "num_input_tokens_seen": 935339160,
      "step": 15595,
      "train_runtime": 200217.336,
      "train_tokens_per_second": 4671.619
    },
    {
      "epoch": 1.989795918367347,
      "grad_norm": 0.6493386512660779,
      "learning_rate": 6.584277755772306e-10,
      "loss": 0.21915626525878906,
      "num_input_tokens_seen": 935637120,
      "step": 15600,
      "train_runtime": 200282.4262,
      "train_tokens_per_second": 4671.589
    },
    {
      "epoch": 1.9904336734693877,
      "grad_norm": 0.7134473120100414,
      "learning_rate": 5.796507877919855e-10,
      "loss": 0.2331714153289795,
      "num_input_tokens_seen": 935934880,
      "step": 15605,
      "train_runtime": 200347.2233,
      "train_tokens_per_second": 4671.564
    },
    {
      "epoch": 1.9910714285714286,
      "grad_norm": 0.6361365769395474,
      "learning_rate": 5.058910674859973e-10,
      "loss": 0.22351865768432616,
      "num_input_tokens_seen": 936237184,
      "step": 15610,
      "train_runtime": 200410.6136,
      "train_tokens_per_second": 4671.595
    },
    {
      "epoch": 1.9917091836734695,
      "grad_norm": 0.64839750830652,
      "learning_rate": 4.371486886822762e-10,
      "loss": 0.2331063985824585,
      "num_input_tokens_seen": 936513656,
      "step": 15615,
      "train_runtime": 200465.0828,
      "train_tokens_per_second": 4671.705
    },
    {
      "epoch": 1.9923469387755102,
      "grad_norm": 0.7466120551652838,
      "learning_rate": 3.7342372036841546e-10,
      "loss": 0.232620906829834,
      "num_input_tokens_seen": 936814896,
      "step": 15620,
      "train_runtime": 200522.7778,
      "train_tokens_per_second": 4671.863
    },
    {
      "epoch": 1.992984693877551,
      "grad_norm": 0.6860694105968212,
      "learning_rate": 3.147162264971471e-10,
      "loss": 0.2312720537185669,
      "num_input_tokens_seen": 937107704,
      "step": 15625,
      "train_runtime": 200584.4207,
      "train_tokens_per_second": 4671.887
    },
    {
      "epoch": 1.993622448979592,
      "grad_norm": 0.6889309531395138,
      "learning_rate": 2.610262659852314e-10,
      "loss": 0.22203030586242675,
      "num_input_tokens_seen": 937400936,
      "step": 15630,
      "train_runtime": 200641.7539,
      "train_tokens_per_second": 4672.013
    },
    {
      "epoch": 1.9942602040816326,
      "grad_norm": 0.7258747239264295,
      "learning_rate": 2.123538927145674e-10,
      "loss": 0.2579025745391846,
      "num_input_tokens_seen": 937697776,
      "step": 15635,
      "train_runtime": 200702.7711,
      "train_tokens_per_second": 4672.072
    },
    {
      "epoch": 1.9948979591836735,
      "grad_norm": 0.7542805069702584,
      "learning_rate": 1.686991555310824e-10,
      "loss": 0.23705878257751464,
      "num_input_tokens_seen": 937980704,
      "step": 15640,
      "train_runtime": 200765.3389,
      "train_tokens_per_second": 4672.025
    },
    {
      "epoch": 1.9955357142857144,
      "grad_norm": 0.708619693440659,
      "learning_rate": 1.3006209824584226e-10,
      "loss": 0.22812957763671876,
      "num_input_tokens_seen": 938273128,
      "step": 15645,
      "train_runtime": 200836.7758,
      "train_tokens_per_second": 4671.819
    },
    {
      "epoch": 1.996173469387755,
      "grad_norm": 0.6945117523175717,
      "learning_rate": 9.6442759632831e-11,
      "loss": 0.24734277725219728,
      "num_input_tokens_seen": 938574952,
      "step": 15650,
      "train_runtime": 200899.4054,
      "train_tokens_per_second": 4671.865
    },
    {
      "epoch": 1.9968112244897958,
      "grad_norm": 0.6905716380218004,
      "learning_rate": 6.784117343228147e-11,
      "loss": 0.2404686689376831,
      "num_input_tokens_seen": 938879584,
      "step": 15655,
      "train_runtime": 200966.372,
      "train_tokens_per_second": 4671.824
    },
    {
      "epoch": 1.9974489795918369,
      "grad_norm": 0.6406483668769022,
      "learning_rate": 4.425736834789973e-11,
      "loss": 0.2283982515335083,
      "num_input_tokens_seen": 939182000,
      "step": 15660,
      "train_runtime": 201029.3363,
      "train_tokens_per_second": 4671.865
    },
    {
      "epoch": 1.9980867346938775,
      "grad_norm": 0.598588173441811,
      "learning_rate": 2.5691368046865118e-11,
      "loss": 0.2239130973815918,
      "num_input_tokens_seen": 939491240,
      "step": 15665,
      "train_runtime": 201084.494,
      "train_tokens_per_second": 4672.122
    },
    {
      "epoch": 1.9987244897959182,
      "grad_norm": 0.6020425993622818,
      "learning_rate": 1.214319116260576e-11,
      "loss": 0.23047285079956054,
      "num_input_tokens_seen": 939787104,
      "step": 15670,
      "train_runtime": 201150.2888,
      "train_tokens_per_second": 4672.064
    },
    {
      "epoch": 1.9993622448979593,
      "grad_norm": 0.6049905153503449,
      "learning_rate": 3.6128512903577105e-12,
      "loss": 0.24902353286743165,
      "num_input_tokens_seen": 940079408,
      "step": 15675,
      "train_runtime": 201220.7949,
      "train_tokens_per_second": 4671.88
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6681944838914244,
      "learning_rate": 1.0035699216093265e-13,
      "loss": 0.24981300830841063,
      "num_input_tokens_seen": 940386304,
      "step": 15680,
      "train_runtime": 201279.4867,
      "train_tokens_per_second": 4672.042
    },
    {
      "epoch": 2.0,
      "num_input_tokens_seen": 940386304,
      "step": 15680,
      "total_flos": 1739097771048960.0,
      "train_loss": 0.28079748448668695,
      "train_runtime": 201315.0101,
      "train_samples_per_second": 0.312,
      "train_steps_per_second": 0.078
    }
  ],
  "logging_steps": 5,
  "max_steps": 15680,
  "num_input_tokens_seen": 940386304,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1739097771048960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}