{
  "best_global_step": 1000,
  "best_metric": 0.16221453249454498,
  "best_model_checkpoint": "saves/prompt-tuning/gemma-3-1b-it/train_cb_1745950309/checkpoint-1000",
  "epoch": 701.7610619469026,
  "eval_steps": 200,
  "global_step": 40000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08849557522123894,
      "grad_norm": 30.857913970947266,
      "learning_rate": 0.29999999259779675,
      "loss": 11.664,
      "num_input_tokens_seen": 3008,
      "step": 5
    },
    {
      "epoch": 0.17699115044247787,
      "grad_norm": 15.412628173828125,
      "learning_rate": 0.29999996252634736,
      "loss": 11.926,
      "num_input_tokens_seen": 6128,
      "step": 10
    },
    {
      "epoch": 0.26548672566371684,
      "grad_norm": 4.846696853637695,
      "learning_rate": 0.2999999093230187,
      "loss": 11.5005,
      "num_input_tokens_seen": 9056,
      "step": 15
    },
    {
      "epoch": 0.35398230088495575,
      "grad_norm": 3.7165660858154297,
      "learning_rate": 0.299999832987819,
      "loss": 9.2555,
      "num_input_tokens_seen": 12656,
      "step": 20
    },
    {
      "epoch": 0.4424778761061947,
      "grad_norm": 3.7266926765441895,
      "learning_rate": 0.29999973352076004,
      "loss": 6.9613,
      "num_input_tokens_seen": 15808,
      "step": 25
    },
    {
      "epoch": 0.5309734513274337,
      "grad_norm": 2.0266427993774414,
      "learning_rate": 0.2999996109218572,
      "loss": 6.1477,
      "num_input_tokens_seen": 18304,
      "step": 30
    },
    {
      "epoch": 0.6194690265486725,
      "grad_norm": 1.4981653690338135,
      "learning_rate": 0.2999994651911293,
      "loss": 3.1662,
      "num_input_tokens_seen": 21712,
      "step": 35
    },
    {
      "epoch": 0.7079646017699115,
      "grad_norm": 1.5509648323059082,
      "learning_rate": 0.2999992963285989,
      "loss": 2.0894,
      "num_input_tokens_seen": 24576,
      "step": 40
    },
    {
      "epoch": 0.7964601769911505,
      "grad_norm": 1.0404917001724243,
      "learning_rate": 0.29999910433429194,
      "loss": 1.941,
      "num_input_tokens_seen": 27088,
      "step": 45
    },
    {
      "epoch": 0.8849557522123894,
      "grad_norm": 0.5678565502166748,
      "learning_rate": 0.29999888920823814,
      "loss": 1.32,
      "num_input_tokens_seen": 29648,
      "step": 50
    },
    {
      "epoch": 0.9734513274336283,
      "grad_norm": 0.6404495239257812,
      "learning_rate": 0.29999865095047057,
      "loss": 1.5291,
      "num_input_tokens_seen": 32032,
      "step": 55
    },
    {
      "epoch": 1.0530973451327434,
      "grad_norm": 0.6864966750144958,
      "learning_rate": 0.29999838956102604,
      "loss": 0.7341,
      "num_input_tokens_seen": 34624,
      "step": 60
    },
    {
      "epoch": 1.1415929203539823,
      "grad_norm": 0.5049331188201904,
      "learning_rate": 0.29999810503994484,
      "loss": 0.5649,
      "num_input_tokens_seen": 37600,
      "step": 65
    },
    {
      "epoch": 1.2300884955752212,
      "grad_norm": 0.3618448078632355,
      "learning_rate": 0.29999779738727084,
      "loss": 0.8966,
      "num_input_tokens_seen": 40256,
      "step": 70
    },
    {
      "epoch": 1.3185840707964602,
      "grad_norm": 0.6455984711647034,
      "learning_rate": 0.29999746660305154,
      "loss": 0.7875,
      "num_input_tokens_seen": 43440,
      "step": 75
    },
    {
      "epoch": 1.407079646017699,
      "grad_norm": 0.5885500311851501,
      "learning_rate": 0.2999971126873379,
      "loss": 0.6253,
      "num_input_tokens_seen": 46192,
      "step": 80
    },
    {
      "epoch": 1.495575221238938,
      "grad_norm": 0.17574167251586914,
      "learning_rate": 0.2999967356401845,
      "loss": 0.6007,
      "num_input_tokens_seen": 49360,
      "step": 85
    },
    {
      "epoch": 1.584070796460177,
      "grad_norm": 0.34683212637901306,
      "learning_rate": 0.29999633546164944,
      "loss": 0.5341,
      "num_input_tokens_seen": 51840,
      "step": 90
    },
    {
      "epoch": 1.672566371681416,
      "grad_norm": 0.2354772537946701,
      "learning_rate": 0.29999591215179444,
      "loss": 0.3807,
      "num_input_tokens_seen": 54960,
      "step": 95
    },
    {
      "epoch": 1.7610619469026547,
      "grad_norm": 0.4168570041656494,
      "learning_rate": 0.2999954657106849,
      "loss": 0.3344,
      "num_input_tokens_seen": 57776,
      "step": 100
    },
    {
      "epoch": 1.8495575221238938,
      "grad_norm": 0.17382696270942688,
      "learning_rate": 0.2999949961383896,
      "loss": 0.3279,
      "num_input_tokens_seen": 60624,
      "step": 105
    },
    {
      "epoch": 1.9380530973451329,
      "grad_norm": 0.23753246665000916,
      "learning_rate": 0.2999945034349809,
      "loss": 0.309,
      "num_input_tokens_seen": 63424,
      "step": 110
    },
    {
      "epoch": 2.017699115044248,
      "grad_norm": 0.07051464915275574,
      "learning_rate": 0.2999939876005348,
      "loss": 0.3785,
      "num_input_tokens_seen": 65544,
      "step": 115
    },
    {
      "epoch": 2.106194690265487,
      "grad_norm": 0.11237034201622009,
      "learning_rate": 0.29999344863513094,
      "loss": 0.3453,
      "num_input_tokens_seen": 68552,
      "step": 120
    },
    {
      "epoch": 2.1946902654867255,
      "grad_norm": 0.05134290084242821,
      "learning_rate": 0.2999928865388523,
      "loss": 0.2252,
      "num_input_tokens_seen": 71256,
      "step": 125
    },
    {
      "epoch": 2.2831858407079646,
      "grad_norm": 0.3033003509044647,
      "learning_rate": 0.29999230131178567,
      "loss": 0.3022,
      "num_input_tokens_seen": 74200,
      "step": 130
    },
    {
      "epoch": 2.3716814159292037,
      "grad_norm": 0.07960475981235504,
      "learning_rate": 0.2999916929540212,
      "loss": 0.3919,
      "num_input_tokens_seen": 76696,
      "step": 135
    },
    {
      "epoch": 2.4601769911504423,
      "grad_norm": 0.12232175469398499,
      "learning_rate": 0.29999106146565285,
      "loss": 0.3556,
      "num_input_tokens_seen": 79640,
      "step": 140
    },
    {
      "epoch": 2.5486725663716814,
      "grad_norm": 0.08066463470458984,
      "learning_rate": 0.29999040684677786,
      "loss": 0.3411,
      "num_input_tokens_seen": 82632,
      "step": 145
    },
    {
      "epoch": 2.6371681415929205,
      "grad_norm": 0.12378513067960739,
      "learning_rate": 0.2999897290974972,
      "loss": 0.248,
      "num_input_tokens_seen": 85496,
      "step": 150
    },
    {
      "epoch": 2.725663716814159,
      "grad_norm": 0.07555761933326721,
      "learning_rate": 0.2999890282179155,
      "loss": 0.159,
      "num_input_tokens_seen": 89128,
      "step": 155
    },
    {
      "epoch": 2.814159292035398,
      "grad_norm": 0.20620428025722504,
      "learning_rate": 0.29998830420814077,
      "loss": 0.3356,
      "num_input_tokens_seen": 92152,
      "step": 160
    },
    {
      "epoch": 2.9026548672566372,
      "grad_norm": 0.06485848128795624,
      "learning_rate": 0.2999875570682846,
      "loss": 0.3261,
      "num_input_tokens_seen": 94632,
      "step": 165
    },
    {
      "epoch": 2.991150442477876,
      "grad_norm": 0.03192218020558357,
      "learning_rate": 0.2999867867984623,
      "loss": 0.1706,
      "num_input_tokens_seen": 97544,
      "step": 170
    },
    {
      "epoch": 3.0707964601769913,
      "grad_norm": 0.0249206330627203,
      "learning_rate": 0.29998599339879267,
      "loss": 0.1368,
      "num_input_tokens_seen": 99992,
      "step": 175
    },
    {
      "epoch": 3.15929203539823,
      "grad_norm": 0.08197808265686035,
      "learning_rate": 0.29998517686939796,
      "loss": 0.1791,
      "num_input_tokens_seen": 102824,
      "step": 180
    },
    {
      "epoch": 3.247787610619469,
      "grad_norm": 0.12062375247478485,
      "learning_rate": 0.29998433721040413,
      "loss": 0.2248,
      "num_input_tokens_seen": 105608,
      "step": 185
    },
    {
      "epoch": 3.336283185840708,
      "grad_norm": 0.08413828164339066,
      "learning_rate": 0.29998347442194073,
      "loss": 0.1702,
      "num_input_tokens_seen": 108712,
      "step": 190
    },
    {
      "epoch": 3.4247787610619467,
      "grad_norm": 0.012988623231649399,
      "learning_rate": 0.2999825885041407,
      "loss": 0.1231,
      "num_input_tokens_seen": 111608,
      "step": 195
    },
    {
      "epoch": 3.5132743362831858,
      "grad_norm": 0.07738474011421204,
      "learning_rate": 0.29998167945714077,
      "loss": 0.388,
      "num_input_tokens_seen": 114504,
      "step": 200
    },
    {
      "epoch": 3.5132743362831858,
      "eval_loss": 0.26136693358421326,
      "eval_runtime": 0.9276,
      "eval_samples_per_second": 26.95,
      "eval_steps_per_second": 14.014,
      "num_input_tokens_seen": 114504,
      "step": 200
    },
    {
      "epoch": 3.601769911504425,
      "grad_norm": 0.09463059902191162,
      "learning_rate": 0.2999807472810811,
      "loss": 0.3094,
      "num_input_tokens_seen": 117592,
      "step": 205
    },
    {
      "epoch": 3.6902654867256635,
      "grad_norm": 0.017677707597613335,
      "learning_rate": 0.29997979197610536,
      "loss": 0.1854,
      "num_input_tokens_seen": 120216,
      "step": 210
    },
    {
      "epoch": 3.7787610619469025,
      "grad_norm": 0.021596524864435196,
      "learning_rate": 0.299978813542361,
      "loss": 0.3197,
      "num_input_tokens_seen": 122904,
      "step": 215
    },
    {
      "epoch": 3.8672566371681416,
      "grad_norm": 0.05760117247700691,
      "learning_rate": 0.2999778119799988,
      "loss": 0.186,
      "num_input_tokens_seen": 125832,
      "step": 220
    },
    {
      "epoch": 3.9557522123893807,
      "grad_norm": 0.047069959342479706,
      "learning_rate": 0.29997678728917326,
      "loss": 0.2049,
      "num_input_tokens_seen": 129512,
      "step": 225
    },
    {
      "epoch": 4.035398230088496,
      "grad_norm": 0.020536065101623535,
      "learning_rate": 0.2999757394700424,
      "loss": 0.1653,
      "num_input_tokens_seen": 131872,
      "step": 230
    },
    {
      "epoch": 4.123893805309734,
      "grad_norm": 0.04338550567626953,
      "learning_rate": 0.29997466852276783,
      "loss": 0.1195,
      "num_input_tokens_seen": 134624,
      "step": 235
    },
    {
      "epoch": 4.212389380530974,
      "grad_norm": 0.022253643721342087,
      "learning_rate": 0.29997357444751466,
      "loss": 0.1374,
      "num_input_tokens_seen": 137168,
      "step": 240
    },
    {
      "epoch": 4.300884955752212,
      "grad_norm": 0.03547932952642441,
      "learning_rate": 0.2999724572444516,
      "loss": 0.1834,
      "num_input_tokens_seen": 139904,
      "step": 245
    },
    {
      "epoch": 4.389380530973451,
      "grad_norm": 0.04908934235572815,
      "learning_rate": 0.29997131691375095,
      "loss": 0.1778,
      "num_input_tokens_seen": 142496,
      "step": 250
    },
    {
      "epoch": 4.477876106194691,
      "grad_norm": 0.018861019983887672,
      "learning_rate": 0.2999701534555886,
      "loss": 0.1685,
      "num_input_tokens_seen": 145632,
      "step": 255
    },
    {
      "epoch": 4.566371681415929,
      "grad_norm": 0.02899417281150818,
      "learning_rate": 0.2999689668701439,
      "loss": 0.1218,
      "num_input_tokens_seen": 148224,
      "step": 260
    },
    {
      "epoch": 4.654867256637168,
      "grad_norm": 0.027406131848692894,
      "learning_rate": 0.29996775715759993,
      "loss": 0.1349,
      "num_input_tokens_seen": 150928,
      "step": 265
    },
    {
      "epoch": 4.743362831858407,
      "grad_norm": 0.04806069657206535,
      "learning_rate": 0.2999665243181432,
      "loss": 0.1437,
      "num_input_tokens_seen": 153808,
      "step": 270
    },
    {
      "epoch": 4.831858407079646,
      "grad_norm": 0.032632045447826385,
      "learning_rate": 0.2999652683519638,
      "loss": 0.2409,
      "num_input_tokens_seen": 157248,
      "step": 275
    },
    {
      "epoch": 4.920353982300885,
      "grad_norm": 0.02714240550994873,
      "learning_rate": 0.29996398925925544,
      "loss": 0.1812,
      "num_input_tokens_seen": 160352,
      "step": 280
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.1032564714550972,
      "learning_rate": 0.2999626870402154,
      "loss": 0.1153,
      "num_input_tokens_seen": 162728,
      "step": 285
    },
    {
      "epoch": 5.088495575221239,
      "grad_norm": 0.05542123317718506,
      "learning_rate": 0.29996136169504445,
      "loss": 0.2365,
      "num_input_tokens_seen": 165368,
      "step": 290
    },
    {
      "epoch": 5.176991150442478,
      "grad_norm": 0.04795343056321144,
      "learning_rate": 0.29996001322394694,
      "loss": 0.1683,
      "num_input_tokens_seen": 168456,
      "step": 295
    },
    {
      "epoch": 5.265486725663717,
      "grad_norm": 0.06972891837358475,
      "learning_rate": 0.29995864162713093,
      "loss": 0.1462,
      "num_input_tokens_seen": 171512,
      "step": 300
    },
    {
      "epoch": 5.353982300884955,
      "grad_norm": 0.05509583279490471,
      "learning_rate": 0.2999572469048079,
      "loss": 0.181,
      "num_input_tokens_seen": 174920,
      "step": 305
    },
    {
      "epoch": 5.442477876106195,
      "grad_norm": 0.03771742060780525,
      "learning_rate": 0.29995582905719287,
      "loss": 0.131,
      "num_input_tokens_seen": 177496,
      "step": 310
    },
    {
      "epoch": 5.530973451327434,
      "grad_norm": 0.026887349784374237,
      "learning_rate": 0.2999543880845046,
      "loss": 0.0887,
      "num_input_tokens_seen": 180328,
      "step": 315
    },
    {
      "epoch": 5.619469026548672,
      "grad_norm": 0.04695015028119087,
      "learning_rate": 0.2999529239869652,
      "loss": 0.1749,
      "num_input_tokens_seen": 183000,
      "step": 320
    },
    {
      "epoch": 5.707964601769912,
      "grad_norm": 0.037778183817863464,
      "learning_rate": 0.2999514367648005,
      "loss": 0.1148,
      "num_input_tokens_seen": 185512,
      "step": 325
    },
    {
      "epoch": 5.79646017699115,
      "grad_norm": 0.039306797087192535,
      "learning_rate": 0.29994992641823987,
      "loss": 0.1924,
      "num_input_tokens_seen": 188776,
      "step": 330
    },
    {
      "epoch": 5.88495575221239,
      "grad_norm": 0.07070355862379074,
      "learning_rate": 0.29994839294751613,
      "loss": 0.1748,
      "num_input_tokens_seen": 191816,
      "step": 335
    },
    {
      "epoch": 5.9734513274336285,
      "grad_norm": 0.06912440806627274,
      "learning_rate": 0.29994683635286584,
      "loss": 0.1692,
      "num_input_tokens_seen": 194344,
      "step": 340
    },
    {
      "epoch": 6.053097345132743,
      "grad_norm": 0.013760467059910297,
      "learning_rate": 0.2999452566345291,
      "loss": 0.0577,
      "num_input_tokens_seen": 197000,
      "step": 345
    },
    {
      "epoch": 6.1415929203539825,
      "grad_norm": 0.05414609983563423,
      "learning_rate": 0.2999436537927494,
      "loss": 0.2424,
      "num_input_tokens_seen": 199992,
      "step": 350
    },
    {
      "epoch": 6.230088495575221,
      "grad_norm": 0.04373975098133087,
      "learning_rate": 0.299942027827774,
      "loss": 0.0888,
      "num_input_tokens_seen": 203208,
      "step": 355
    },
    {
      "epoch": 6.31858407079646,
      "grad_norm": 0.07380948215723038,
      "learning_rate": 0.29994037873985363,
      "loss": 0.169,
      "num_input_tokens_seen": 205976,
      "step": 360
    },
    {
      "epoch": 6.407079646017699,
      "grad_norm": 0.02306806482374668,
      "learning_rate": 0.29993870652924254,
      "loss": 0.1052,
      "num_input_tokens_seen": 209176,
      "step": 365
    },
    {
      "epoch": 6.495575221238938,
      "grad_norm": 0.010759019292891026,
      "learning_rate": 0.29993701119619876,
      "loss": 0.1041,
      "num_input_tokens_seen": 211896,
      "step": 370
    },
    {
      "epoch": 6.584070796460177,
      "grad_norm": 0.07841109484434128,
      "learning_rate": 0.2999352927409835,
      "loss": 0.1436,
      "num_input_tokens_seen": 214824,
      "step": 375
    },
    {
      "epoch": 6.672566371681416,
      "grad_norm": 0.09694568067789078,
      "learning_rate": 0.29993355116386194,
      "loss": 0.1655,
      "num_input_tokens_seen": 217192,
      "step": 380
    },
    {
      "epoch": 6.761061946902655,
      "grad_norm": 0.04355792701244354,
      "learning_rate": 0.29993178646510266,
      "loss": 0.1346,
      "num_input_tokens_seen": 219816,
      "step": 385
    },
    {
      "epoch": 6.849557522123893,
      "grad_norm": 0.03253980726003647,
      "learning_rate": 0.2999299986449777,
      "loss": 0.1362,
      "num_input_tokens_seen": 222360,
      "step": 390
    },
    {
      "epoch": 6.938053097345133,
      "grad_norm": 0.03275546059012413,
      "learning_rate": 0.29992818770376284,
      "loss": 0.0809,
      "num_input_tokens_seen": 225864,
      "step": 395
    },
    {
      "epoch": 7.017699115044247,
      "grad_norm": 0.004386965651065111,
      "learning_rate": 0.29992635364173725,
      "loss": 0.1626,
      "num_input_tokens_seen": 228504,
      "step": 400
    },
    {
      "epoch": 7.017699115044247,
      "eval_loss": 0.26157596707344055,
      "eval_runtime": 0.9127,
      "eval_samples_per_second": 27.39,
      "eval_steps_per_second": 14.243,
      "num_input_tokens_seen": 228504,
      "step": 400
    },
    {
      "epoch": 7.106194690265487,
      "grad_norm": 0.02609449438750744,
      "learning_rate": 0.2999244964591839,
      "loss": 0.1058,
      "num_input_tokens_seen": 231416,
      "step": 405
    },
    {
      "epoch": 7.1946902654867255,
      "grad_norm": 0.04712270200252533,
      "learning_rate": 0.2999226161563891,
      "loss": 0.071,
      "num_input_tokens_seen": 233832,
      "step": 410
    },
    {
      "epoch": 7.283185840707965,
      "grad_norm": 0.1328945904970169,
      "learning_rate": 0.2999207127336429,
      "loss": 0.3077,
      "num_input_tokens_seen": 236776,
      "step": 415
    },
    {
      "epoch": 7.371681415929204,
      "grad_norm": 0.1069122925400734,
      "learning_rate": 0.2999187861912387,
      "loss": 0.303,
      "num_input_tokens_seen": 239720,
      "step": 420
    },
    {
      "epoch": 7.460176991150442,
      "grad_norm": 0.014653612859547138,
      "learning_rate": 0.2999168365294737,
      "loss": 0.2254,
      "num_input_tokens_seen": 242344,
      "step": 425
    },
    {
      "epoch": 7.548672566371682,
      "grad_norm": 0.0725458487868309,
      "learning_rate": 0.29991486374864856,
      "loss": 0.2119,
      "num_input_tokens_seen": 245208,
      "step": 430
    },
    {
      "epoch": 7.6371681415929205,
      "grad_norm": 0.17669686675071716,
      "learning_rate": 0.29991286784906745,
      "loss": 0.2821,
      "num_input_tokens_seen": 248232,
      "step": 435
    },
    {
      "epoch": 7.725663716814159,
      "grad_norm": 0.03473649173974991,
      "learning_rate": 0.2999108488310382,
      "loss": 0.2033,
      "num_input_tokens_seen": 251080,
      "step": 440
    },
    {
      "epoch": 7.814159292035399,
      "grad_norm": 0.1072806790471077,
      "learning_rate": 0.29990880669487213,
      "loss": 0.1306,
      "num_input_tokens_seen": 253576,
      "step": 445
    },
    {
      "epoch": 7.902654867256637,
      "grad_norm": 0.07879234850406647,
      "learning_rate": 0.29990674144088425,
      "loss": 0.287,
      "num_input_tokens_seen": 256616,
      "step": 450
    },
    {
      "epoch": 7.991150442477876,
      "grad_norm": 0.010200544260442257,
      "learning_rate": 0.299904653069393,
      "loss": 0.1301,
      "num_input_tokens_seen": 259544,
      "step": 455
    },
    {
      "epoch": 8.070796460176991,
      "grad_norm": 0.05468774959445,
      "learning_rate": 0.29990254158072044,
      "loss": 0.107,
      "num_input_tokens_seen": 262232,
      "step": 460
    },
    {
      "epoch": 8.15929203539823,
      "grad_norm": 0.050852928310632706,
      "learning_rate": 0.2999004069751921,
      "loss": 0.2407,
      "num_input_tokens_seen": 265048,
      "step": 465
    },
    {
      "epoch": 8.247787610619469,
      "grad_norm": 0.048885371536016464,
      "learning_rate": 0.2998982492531373,
      "loss": 0.1074,
      "num_input_tokens_seen": 268008,
      "step": 470
    },
    {
      "epoch": 8.336283185840708,
      "grad_norm": 0.01333150826394558,
      "learning_rate": 0.2998960684148887,
      "loss": 0.1284,
      "num_input_tokens_seen": 270568,
      "step": 475
    },
    {
      "epoch": 8.424778761061948,
      "grad_norm": 0.061909109354019165,
      "learning_rate": 0.29989386446078264,
      "loss": 0.1909,
      "num_input_tokens_seen": 273224,
      "step": 480
    },
    {
      "epoch": 8.513274336283185,
      "grad_norm": 0.04440929368138313,
      "learning_rate": 0.299891637391159,
      "loss": 0.2364,
      "num_input_tokens_seen": 276472,
      "step": 485
    },
    {
      "epoch": 8.601769911504425,
      "grad_norm": 0.015465333126485348,
      "learning_rate": 0.2998893872063612,
      "loss": 0.1328,
      "num_input_tokens_seen": 279432,
      "step": 490
    },
    {
      "epoch": 8.690265486725664,
      "grad_norm": 0.021732686087489128,
      "learning_rate": 0.2998871139067363,
      "loss": 0.159,
      "num_input_tokens_seen": 282232,
      "step": 495
    },
    {
      "epoch": 8.778761061946902,
      "grad_norm": 0.0326734259724617,
      "learning_rate": 0.2998848174926348,
      "loss": 0.1221,
      "num_input_tokens_seen": 284776,
      "step": 500
    },
    {
      "epoch": 8.867256637168142,
      "grad_norm": 0.048362698405981064,
      "learning_rate": 0.2998824979644109,
      "loss": 0.1715,
      "num_input_tokens_seen": 287864,
      "step": 505
    },
    {
      "epoch": 8.955752212389381,
      "grad_norm": 0.03577310964465141,
      "learning_rate": 0.29988015532242224,
      "loss": 0.1428,
      "num_input_tokens_seen": 290712,
      "step": 510
    },
    {
      "epoch": 9.035398230088495,
      "grad_norm": 0.03865543380379677,
      "learning_rate": 0.29987778956703015,
      "loss": 0.0928,
      "num_input_tokens_seen": 293160,
      "step": 515
    },
    {
      "epoch": 9.123893805309734,
      "grad_norm": 0.009994987398386002,
      "learning_rate": 0.2998754006985994,
      "loss": 0.0693,
      "num_input_tokens_seen": 296472,
      "step": 520
    },
    {
      "epoch": 9.212389380530974,
      "grad_norm": 0.05774105340242386,
      "learning_rate": 0.29987298871749846,
      "loss": 0.1607,
      "num_input_tokens_seen": 299112,
      "step": 525
    },
    {
      "epoch": 9.300884955752213,
      "grad_norm": 0.04266613349318504,
      "learning_rate": 0.2998705536240992,
      "loss": 0.1653,
      "num_input_tokens_seen": 302024,
      "step": 530
    },
    {
      "epoch": 9.389380530973451,
      "grad_norm": 0.04025699943304062,
      "learning_rate": 0.2998680954187772,
      "loss": 0.1146,
      "num_input_tokens_seen": 304872,
      "step": 535
    },
    {
      "epoch": 9.47787610619469,
      "grad_norm": 0.08113261312246323,
      "learning_rate": 0.2998656141019115,
      "loss": 0.1603,
      "num_input_tokens_seen": 307560,
      "step": 540
    },
    {
      "epoch": 9.56637168141593,
      "grad_norm": 0.06283559650182724,
      "learning_rate": 0.2998631096738848,
      "loss": 0.0989,
      "num_input_tokens_seen": 310840,
      "step": 545
    },
    {
      "epoch": 9.654867256637168,
      "grad_norm": 0.045276716351509094,
      "learning_rate": 0.29986058213508326,
      "loss": 0.1774,
      "num_input_tokens_seen": 313800,
      "step": 550
    },
    {
      "epoch": 9.743362831858407,
      "grad_norm": 0.02669377438724041,
      "learning_rate": 0.29985803148589674,
      "loss": 0.1324,
      "num_input_tokens_seen": 316664,
      "step": 555
    },
    {
      "epoch": 9.831858407079647,
      "grad_norm": 0.06914814561605453,
      "learning_rate": 0.2998554577267185,
      "loss": 0.2518,
      "num_input_tokens_seen": 319656,
      "step": 560
    },
    {
      "epoch": 9.920353982300885,
      "grad_norm": 0.050139423459768295,
      "learning_rate": 0.2998528608579455,
      "loss": 0.1795,
      "num_input_tokens_seen": 322536,
      "step": 565
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.0408666767179966,
      "learning_rate": 0.2998502408799781,
      "loss": 0.2118,
      "num_input_tokens_seen": 324624,
      "step": 570
    },
    {
      "epoch": 10.08849557522124,
      "grad_norm": 0.07781104743480682,
      "learning_rate": 0.2998475977932205,
      "loss": 0.2035,
      "num_input_tokens_seen": 327488,
      "step": 575
    },
    {
      "epoch": 10.176991150442477,
      "grad_norm": 0.05363497510552406,
      "learning_rate": 0.29984493159808023,
      "loss": 0.1568,
      "num_input_tokens_seen": 330240,
      "step": 580
    },
    {
      "epoch": 10.265486725663717,
      "grad_norm": 0.06371191143989563,
      "learning_rate": 0.29984224229496836,
      "loss": 0.1581,
      "num_input_tokens_seen": 332976,
      "step": 585
    },
    {
      "epoch": 10.353982300884956,
      "grad_norm": 0.030843645334243774,
      "learning_rate": 0.2998395298842998,
      "loss": 0.1529,
      "num_input_tokens_seen": 335504,
      "step": 590
    },
    {
      "epoch": 10.442477876106194,
      "grad_norm": 0.06816186755895615,
      "learning_rate": 0.29983679436649263,
      "loss": 0.1245,
      "num_input_tokens_seen": 338512,
      "step": 595
    },
    {
      "epoch": 10.530973451327434,
      "grad_norm": 0.038672056049108505,
      "learning_rate": 0.2998340357419689,
      "loss": 0.1683,
      "num_input_tokens_seen": 341136,
      "step": 600
    },
    {
      "epoch": 10.530973451327434,
      "eval_loss": 0.18268337845802307,
      "eval_runtime": 0.9301,
      "eval_samples_per_second": 26.88,
      "eval_steps_per_second": 13.977,
      "num_input_tokens_seen": 341136,
      "step": 600
    },
    {
      "epoch": 10.619469026548673,
      "grad_norm": 0.039640266448259354,
      "learning_rate": 0.29983125401115385,
      "loss": 0.094,
      "num_input_tokens_seen": 343664,
      "step": 605
    },
    {
      "epoch": 10.70796460176991,
      "grad_norm": 0.05746905133128166,
      "learning_rate": 0.29982844917447654,
      "loss": 0.1391,
      "num_input_tokens_seen": 346656,
      "step": 610
    },
    {
      "epoch": 10.79646017699115,
      "grad_norm": 0.06383468955755234,
      "learning_rate": 0.2998256212323695,
      "loss": 0.0731,
      "num_input_tokens_seen": 349712,
      "step": 615
    },
    {
      "epoch": 10.88495575221239,
      "grad_norm": 0.023070938885211945,
      "learning_rate": 0.29982277018526887,
      "loss": 0.1777,
      "num_input_tokens_seen": 352896,
      "step": 620
    },
    {
      "epoch": 10.973451327433628,
      "grad_norm": 0.01819828897714615,
      "learning_rate": 0.2998198960336143,
      "loss": 0.0921,
      "num_input_tokens_seen": 356016,
      "step": 625
    },
    {
      "epoch": 11.053097345132743,
      "grad_norm": 0.044772058725357056,
      "learning_rate": 0.299816998777849,
      "loss": 0.1844,
      "num_input_tokens_seen": 358432,
      "step": 630
    },
    {
      "epoch": 11.141592920353983,
      "grad_norm": 0.05431145057082176,
      "learning_rate": 0.2998140784184197,
      "loss": 0.1205,
      "num_input_tokens_seen": 361008,
      "step": 635
    },
    {
      "epoch": 11.230088495575222,
      "grad_norm": 0.025473183020949364,
      "learning_rate": 0.2998111349557769,
      "loss": 0.1439,
      "num_input_tokens_seen": 363728,
      "step": 640
    },
    {
      "epoch": 11.31858407079646,
      "grad_norm": 0.12129873782396317,
      "learning_rate": 0.29980816839037444,
      "loss": 0.1391,
      "num_input_tokens_seen": 367232,
      "step": 645
    },
    {
      "epoch": 11.4070796460177,
      "grad_norm": 0.07408632338047028,
      "learning_rate": 0.2998051787226698,
      "loss": 0.1066,
      "num_input_tokens_seen": 370448,
      "step": 650
    },
    {
      "epoch": 11.495575221238939,
      "grad_norm": 0.07822711020708084,
      "learning_rate": 0.29980216595312403,
      "loss": 0.138,
      "num_input_tokens_seen": 373504,
      "step": 655
    },
    {
      "epoch": 11.584070796460177,
      "grad_norm": 0.035831745713949203,
      "learning_rate": 0.29979913008220177,
      "loss": 0.1095,
      "num_input_tokens_seen": 376464,
      "step": 660
    },
    {
      "epoch": 11.672566371681416,
      "grad_norm": 0.05501428619027138,
      "learning_rate": 0.2997960711103711,
      "loss": 0.1716,
      "num_input_tokens_seen": 379216,
      "step": 665
    },
    {
      "epoch": 11.761061946902656,
      "grad_norm": 0.036521703004837036,
      "learning_rate": 0.29979298903810386,
      "loss": 0.0997,
      "num_input_tokens_seen": 381664,
      "step": 670
    },
    {
      "epoch": 11.849557522123893,
      "grad_norm": 0.017547767609357834,
      "learning_rate": 0.29978988386587524,
      "loss": 0.1861,
      "num_input_tokens_seen": 384432,
      "step": 675
    },
    {
      "epoch": 11.938053097345133,
      "grad_norm": 0.021192388609051704,
      "learning_rate": 0.2997867555941642,
      "loss": 0.0548,
      "num_input_tokens_seen": 387296,
      "step": 680
    },
    {
      "epoch": 12.017699115044248,
      "grad_norm": 0.0301345344632864,
      "learning_rate": 0.299783604223453,
      "loss": 0.1076,
      "num_input_tokens_seen": 390160,
      "step": 685
    },
    {
      "epoch": 12.106194690265486,
      "grad_norm": 0.09441268444061279,
      "learning_rate": 0.29978042975422786,
      "loss": 0.0864,
      "num_input_tokens_seen": 392944,
      "step": 690
    },
    {
      "epoch": 12.194690265486726,
      "grad_norm": 0.010798695497214794,
      "learning_rate": 0.29977723218697816,
      "loss": 0.0513,
      "num_input_tokens_seen": 395984,
      "step": 695
    },
    {
      "epoch": 12.283185840707965,
      "grad_norm": 0.07217500358819962,
      "learning_rate": 0.299774011522197,
      "loss": 0.1622,
      "num_input_tokens_seen": 398848,
      "step": 700
    },
    {
      "epoch": 12.371681415929203,
      "grad_norm": 0.01950668916106224,
      "learning_rate": 0.29977076776038114,
      "loss": 0.1213,
      "num_input_tokens_seen": 401536,
      "step": 705
    },
    {
      "epoch": 12.460176991150442,
      "grad_norm": 0.037151433527469635,
      "learning_rate": 0.2997675009020307,
      "loss": 0.0885,
      "num_input_tokens_seen": 404864,
      "step": 710
    },
    {
      "epoch": 12.548672566371682,
      "grad_norm": 0.03521841764450073,
      "learning_rate": 0.2997642109476496,
      "loss": 0.0841,
      "num_input_tokens_seen": 407744,
      "step": 715
    },
    {
      "epoch": 12.63716814159292,
      "grad_norm": 0.05878576263785362,
      "learning_rate": 0.299760897897745,
      "loss": 0.1851,
      "num_input_tokens_seen": 410416,
      "step": 720
    },
    {
      "epoch": 12.725663716814159,
      "grad_norm": 0.05503930523991585,
      "learning_rate": 0.29975756175282803,
      "loss": 0.154,
      "num_input_tokens_seen": 413488,
      "step": 725
    },
    {
      "epoch": 12.814159292035399,
      "grad_norm": 0.00581141235306859,
      "learning_rate": 0.29975420251341306,
      "loss": 0.143,
      "num_input_tokens_seen": 416048,
      "step": 730
    },
    {
      "epoch": 12.902654867256636,
      "grad_norm": 0.10581871122121811,
      "learning_rate": 0.29975082018001814,
      "loss": 0.2202,
      "num_input_tokens_seen": 418944,
      "step": 735
    },
    {
      "epoch": 12.991150442477876,
      "grad_norm": 0.04614975303411484,
      "learning_rate": 0.2997474147531648,
      "loss": 0.1573,
      "num_input_tokens_seen": 421632,
      "step": 740
    },
    {
      "epoch": 13.070796460176991,
      "grad_norm": 0.01075072307139635,
      "learning_rate": 0.29974398623337833,
      "loss": 0.0464,
      "num_input_tokens_seen": 424608,
      "step": 745
    },
    {
      "epoch": 13.15929203539823,
      "grad_norm": 0.017960282042622566,
      "learning_rate": 0.2997405346211873,
      "loss": 0.1037,
      "num_input_tokens_seen": 427088,
      "step": 750
    },
    {
      "epoch": 13.247787610619469,
      "grad_norm": 0.0385766439139843,
      "learning_rate": 0.2997370599171241,
      "loss": 0.1049,
      "num_input_tokens_seen": 429744,
      "step": 755
    },
    {
      "epoch": 13.336283185840708,
      "grad_norm": 0.05268830806016922,
      "learning_rate": 0.2997335621217246,
      "loss": 0.0809,
      "num_input_tokens_seen": 432752,
      "step": 760
    },
    {
      "epoch": 13.424778761061948,
      "grad_norm": 0.03978714719414711,
      "learning_rate": 0.29973004123552816,
      "loss": 0.1434,
      "num_input_tokens_seen": 435456,
      "step": 765
    },
    {
      "epoch": 13.513274336283185,
      "grad_norm": 0.0829022005200386,
      "learning_rate": 0.2997264972590777,
      "loss": 0.0889,
      "num_input_tokens_seen": 438608,
      "step": 770
    },
    {
      "epoch": 13.601769911504425,
      "grad_norm": 0.05147767812013626,
      "learning_rate": 0.29972293019291973,
      "loss": 0.1205,
      "num_input_tokens_seen": 441248,
      "step": 775
    },
    {
      "epoch": 13.690265486725664,
      "grad_norm": 0.02918502688407898,
      "learning_rate": 0.2997193400376045,
      "loss": 0.1391,
      "num_input_tokens_seen": 444160,
      "step": 780
    },
    {
      "epoch": 13.778761061946902,
      "grad_norm": 0.011774185113608837,
      "learning_rate": 0.2997157267936854,
      "loss": 0.0952,
      "num_input_tokens_seen": 447200,
      "step": 785
    },
    {
      "epoch": 13.867256637168142,
      "grad_norm": 0.04630854353308678,
      "learning_rate": 0.2997120904617199,
      "loss": 0.0895,
      "num_input_tokens_seen": 450320,
      "step": 790
    },
    {
      "epoch": 13.955752212389381,
      "grad_norm": 0.05750810354948044,
      "learning_rate": 0.29970843104226863,
      "loss": 0.0835,
      "num_input_tokens_seen": 453328,
      "step": 795
    },
    {
      "epoch": 14.035398230088495,
      "grad_norm": 0.05526066944003105,
      "learning_rate": 0.2997047485358959,
      "loss": 0.1169,
      "num_input_tokens_seen": 455488,
      "step": 800
    },
    {
      "epoch": 14.035398230088495,
      "eval_loss": 0.3692370653152466,
      "eval_runtime": 0.9282,
      "eval_samples_per_second": 26.933,
      "eval_steps_per_second": 14.005,
      "num_input_tokens_seen": 455488,
      "step": 800
    },
    {
      "epoch": 14.123893805309734,
      "grad_norm": 0.001955971820279956,
      "learning_rate": 0.2997010429431697,
      "loss": 0.068,
      "num_input_tokens_seen": 458400,
      "step": 805
    },
    {
      "epoch": 14.212389380530974,
      "grad_norm": 0.003783697262406349,
      "learning_rate": 0.29969731426466134,
      "loss": 0.0242,
      "num_input_tokens_seen": 461168,
      "step": 810
    },
    {
      "epoch": 14.300884955752213,
      "grad_norm": 0.049082037061452866,
      "learning_rate": 0.299693562500946,
      "loss": 0.1366,
      "num_input_tokens_seen": 463840,
      "step": 815
    },
    {
      "epoch": 14.389380530973451,
      "grad_norm": 0.044649045914411545,
      "learning_rate": 0.29968978765260207,
      "loss": 0.0568,
      "num_input_tokens_seen": 467008,
      "step": 820
    },
    {
      "epoch": 14.47787610619469,
      "grad_norm": 0.06274111568927765,
      "learning_rate": 0.2996859897202118,
      "loss": 0.0451,
      "num_input_tokens_seen": 469856,
      "step": 825
    },
    {
      "epoch": 14.56637168141593,
      "grad_norm": 0.008572902530431747,
      "learning_rate": 0.2996821687043609,
      "loss": 0.0694,
      "num_input_tokens_seen": 472672,
      "step": 830
    },
    {
      "epoch": 14.654867256637168,
      "grad_norm": 0.007443713489919901,
      "learning_rate": 0.2996783246056384,
      "loss": 0.0542,
      "num_input_tokens_seen": 475648,
      "step": 835
    },
    {
      "epoch": 14.743362831858407,
      "grad_norm": 0.04691655933856964,
      "learning_rate": 0.29967445742463744,
      "loss": 0.0625,
      "num_input_tokens_seen": 479216,
      "step": 840
    },
    {
      "epoch": 14.831858407079647,
      "grad_norm": 0.0310326237231493,
      "learning_rate": 0.29967056716195417,
      "loss": 0.1367,
      "num_input_tokens_seen": 481744,
      "step": 845
    },
    {
      "epoch": 14.920353982300885,
      "grad_norm": 0.019747117534279823,
      "learning_rate": 0.2996666538181885,
      "loss": 0.1892,
      "num_input_tokens_seen": 484640,
      "step": 850
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.04650654271245003,
      "learning_rate": 0.29966271739394407,
      "loss": 0.1188,
      "num_input_tokens_seen": 486672,
      "step": 855
    },
    {
      "epoch": 15.08849557522124,
      "grad_norm": 0.039386965334415436,
      "learning_rate": 0.29965875788982776,
      "loss": 0.0739,
      "num_input_tokens_seen": 489680,
      "step": 860
    },
    {
      "epoch": 15.176991150442477,
      "grad_norm": 0.05197660252451897,
      "learning_rate": 0.2996547753064503,
      "loss": 0.2028,
      "num_input_tokens_seen": 492864,
      "step": 865
    },
    {
      "epoch": 15.265486725663717,
      "grad_norm": 0.04256889596581459,
      "learning_rate": 0.29965076964442583,
      "loss": 0.2029,
      "num_input_tokens_seen": 495568,
      "step": 870
    },
    {
      "epoch": 15.353982300884956,
      "grad_norm": 0.03237804397940636,
      "learning_rate": 0.299646740904372,
      "loss": 0.1158,
      "num_input_tokens_seen": 498544,
      "step": 875
    },
    {
      "epoch": 15.442477876106194,
      "grad_norm": 0.028797201812267303,
      "learning_rate": 0.29964268908691016,
      "loss": 0.129,
      "num_input_tokens_seen": 501472,
      "step": 880
    },
    {
      "epoch": 15.530973451327434,
      "grad_norm": 0.06987904012203217,
      "learning_rate": 0.29963861419266513,
      "loss": 0.1092,
      "num_input_tokens_seen": 504064,
      "step": 885
    },
    {
      "epoch": 15.619469026548673,
      "grad_norm": 0.103307344019413,
      "learning_rate": 0.29963451622226533,
      "loss": 0.1868,
      "num_input_tokens_seen": 506816,
      "step": 890
    },
    {
      "epoch": 15.70796460176991,
      "grad_norm": 0.044770486652851105,
      "learning_rate": 0.29963039517634277,
      "loss": 0.1244,
      "num_input_tokens_seen": 509552,
      "step": 895
    },
    {
      "epoch": 15.79646017699115,
      "grad_norm": 0.009984035976231098,
      "learning_rate": 0.2996262510555328,
      "loss": 0.0608,
      "num_input_tokens_seen": 512576,
      "step": 900
    },
    {
      "epoch": 15.88495575221239,
      "grad_norm": 0.0823654904961586,
      "learning_rate": 0.2996220838604746,
      "loss": 0.0569,
      "num_input_tokens_seen": 515664,
      "step": 905
    },
    {
      "epoch": 15.973451327433628,
      "grad_norm": 0.04599125683307648,
      "learning_rate": 0.29961789359181085,
      "loss": 0.0578,
      "num_input_tokens_seen": 518464,
      "step": 910
    },
    {
      "epoch": 16.053097345132745,
      "grad_norm": 0.05799437314271927,
      "learning_rate": 0.29961368025018764,
      "loss": 0.4042,
      "num_input_tokens_seen": 520584,
      "step": 915
    },
    {
      "epoch": 16.141592920353983,
      "grad_norm": 0.0686408057808876,
      "learning_rate": 0.2996094438362548,
      "loss": 0.1675,
      "num_input_tokens_seen": 524104,
      "step": 920
    },
    {
      "epoch": 16.23008849557522,
      "grad_norm": 0.06178189441561699,
      "learning_rate": 0.2996051843506657,
      "loss": 0.1185,
      "num_input_tokens_seen": 527352,
      "step": 925
    },
    {
      "epoch": 16.31858407079646,
      "grad_norm": 0.07655368000268936,
      "learning_rate": 0.299600901794077,
      "loss": 0.1685,
      "num_input_tokens_seen": 530328,
      "step": 930
    },
    {
      "epoch": 16.4070796460177,
      "grad_norm": 0.007738374173641205,
      "learning_rate": 0.29959659616714923,
      "loss": 0.1346,
      "num_input_tokens_seen": 532936,
      "step": 935
    },
    {
      "epoch": 16.495575221238937,
      "grad_norm": 0.01790127158164978,
      "learning_rate": 0.2995922674705464,
      "loss": 0.174,
      "num_input_tokens_seen": 536168,
      "step": 940
    },
    {
      "epoch": 16.58407079646018,
      "grad_norm": 0.10434616357088089,
      "learning_rate": 0.2995879157049361,
      "loss": 0.2692,
      "num_input_tokens_seen": 538728,
      "step": 945
    },
    {
      "epoch": 16.672566371681416,
      "grad_norm": 0.06209958344697952,
      "learning_rate": 0.2995835408709893,
      "loss": 0.2169,
      "num_input_tokens_seen": 541832,
      "step": 950
    },
    {
      "epoch": 16.761061946902654,
      "grad_norm": 0.03988637402653694,
      "learning_rate": 0.29957914296938076,
      "loss": 0.2327,
      "num_input_tokens_seen": 544440,
      "step": 955
    },
    {
      "epoch": 16.849557522123895,
      "grad_norm": 0.012704829685389996,
      "learning_rate": 0.2995747220007886,
      "loss": 0.118,
      "num_input_tokens_seen": 547256,
      "step": 960
    },
    {
      "epoch": 16.938053097345133,
      "grad_norm": 0.0582464300096035,
      "learning_rate": 0.2995702779658947,
      "loss": 0.1066,
      "num_input_tokens_seen": 550360,
      "step": 965
    },
    {
      "epoch": 17.01769911504425,
      "grad_norm": 0.02164345234632492,
      "learning_rate": 0.29956581086538425,
      "loss": 0.1278,
      "num_input_tokens_seen": 552464,
      "step": 970
    },
    {
      "epoch": 17.106194690265486,
      "grad_norm": 0.03588252514600754,
      "learning_rate": 0.2995613206999462,
      "loss": 0.1696,
      "num_input_tokens_seen": 555040,
      "step": 975
    },
    {
      "epoch": 17.194690265486727,
      "grad_norm": 0.045338165014982224,
      "learning_rate": 0.29955680747027297,
      "loss": 0.1647,
      "num_input_tokens_seen": 557968,
      "step": 980
    },
    {
      "epoch": 17.283185840707965,
      "grad_norm": 0.017174439504742622,
      "learning_rate": 0.2995522711770607,
      "loss": 0.2074,
      "num_input_tokens_seen": 560768,
      "step": 985
    },
    {
      "epoch": 17.371681415929203,
      "grad_norm": 0.04313628003001213,
      "learning_rate": 0.2995477118210087,
      "loss": 0.1886,
      "num_input_tokens_seen": 563424,
      "step": 990
    },
    {
      "epoch": 17.460176991150444,
      "grad_norm": 0.027376322075724602,
      "learning_rate": 0.29954312940282024,
      "loss": 0.0531,
      "num_input_tokens_seen": 566384,
      "step": 995
    },
    {
      "epoch": 17.548672566371682,
      "grad_norm": 0.07877122610807419,
      "learning_rate": 0.29953852392320196,
      "loss": 0.2206,
      "num_input_tokens_seen": 569504,
      "step": 1000
    },
    {
      "epoch": 17.548672566371682,
      "eval_loss": 0.16221453249454498,
      "eval_runtime": 0.9346,
      "eval_samples_per_second": 26.748,
      "eval_steps_per_second": 13.909,
      "num_input_tokens_seen": 569504,
      "step": 1000
    },
    {
      "epoch": 17.63716814159292,
      "grad_norm": 0.04619221016764641,
      "learning_rate": 0.2995338953828641,
      "loss": 0.1302,
      "num_input_tokens_seen": 572608,
      "step": 1005
    },
    {
      "epoch": 17.72566371681416,
      "grad_norm": 0.029165692627429962,
      "learning_rate": 0.2995292437825204,
      "loss": 0.1459,
      "num_input_tokens_seen": 575504,
      "step": 1010
    },
    {
      "epoch": 17.8141592920354,
      "grad_norm": 0.009039947763085365,
      "learning_rate": 0.29952456912288816,
      "loss": 0.1063,
      "num_input_tokens_seen": 578064,
      "step": 1015
    },
    {
      "epoch": 17.902654867256636,
      "grad_norm": 0.039725981652736664,
      "learning_rate": 0.2995198714046884,
      "loss": 0.1626,
      "num_input_tokens_seen": 581312,
      "step": 1020
    },
    {
      "epoch": 17.991150442477878,
      "grad_norm": 0.054699528962373734,
      "learning_rate": 0.2995151506286454,
      "loss": 0.1926,
      "num_input_tokens_seen": 584240,
      "step": 1025
    },
    {
      "epoch": 18.07079646017699,
      "grad_norm": 0.03119960054755211,
      "learning_rate": 0.2995104067954873,
      "loss": 0.1908,
      "num_input_tokens_seen": 586296,
      "step": 1030
    },
    {
      "epoch": 18.15929203539823,
      "grad_norm": 0.029157282784581184,
      "learning_rate": 0.2995056399059456,
      "loss": 0.3949,
      "num_input_tokens_seen": 588936,
      "step": 1035
    },
    {
      "epoch": 18.24778761061947,
      "grad_norm": 0.020410750061273575,
      "learning_rate": 0.2995008499607554,
      "loss": 0.1605,
      "num_input_tokens_seen": 591816,
      "step": 1040
    },
    {
      "epoch": 18.336283185840706,
      "grad_norm": 0.013808061368763447,
      "learning_rate": 0.2994960369606554,
      "loss": 0.1222,
      "num_input_tokens_seen": 595000,
      "step": 1045
    },
    {
      "epoch": 18.424778761061948,
      "grad_norm": 0.034983083605766296,
      "learning_rate": 0.2994912009063878,
      "loss": 0.0674,
      "num_input_tokens_seen": 597960,
      "step": 1050
    },
    {
      "epoch": 18.513274336283185,
      "grad_norm": 0.046385470777750015,
      "learning_rate": 0.29948634179869843,
      "loss": 0.0976,
      "num_input_tokens_seen": 600728,
      "step": 1055
    },
    {
      "epoch": 18.601769911504427,
      "grad_norm": 0.01408429816365242,
      "learning_rate": 0.29948145963833656,
      "loss": 0.0755,
      "num_input_tokens_seen": 603496,
      "step": 1060
    },
    {
      "epoch": 18.690265486725664,
      "grad_norm": 0.05294256657361984,
      "learning_rate": 0.29947655442605514,
      "loss": 0.1158,
      "num_input_tokens_seen": 606104,
      "step": 1065
    },
    {
      "epoch": 18.778761061946902,
      "grad_norm": 0.018804816529154778,
      "learning_rate": 0.2994716261626106,
      "loss": 0.0952,
      "num_input_tokens_seen": 609160,
      "step": 1070
    },
    {
      "epoch": 18.86725663716814,
      "grad_norm": 0.020481033250689507,
      "learning_rate": 0.2994666748487629,
      "loss": 0.0937,
      "num_input_tokens_seen": 612264,
      "step": 1075
    },
    {
      "epoch": 18.95575221238938,
      "grad_norm": 0.02373555861413479,
      "learning_rate": 0.2994617004852756,
      "loss": 0.0929,
      "num_input_tokens_seen": 615256,
      "step": 1080
    },
    {
      "epoch": 19.035398230088497,
      "grad_norm": 0.03828621283173561,
      "learning_rate": 0.2994567030729159,
      "loss": 0.0655,
      "num_input_tokens_seen": 617880,
      "step": 1085
    },
    {
      "epoch": 19.123893805309734,
      "grad_norm": 0.0347868986427784,
      "learning_rate": 0.29945168261245436,
      "loss": 0.1002,
      "num_input_tokens_seen": 620360,
      "step": 1090
    },
    {
      "epoch": 19.212389380530972,
      "grad_norm": 0.0012306994758546352,
      "learning_rate": 0.29944663910466524,
      "loss": 0.0101,
      "num_input_tokens_seen": 623224,
      "step": 1095
    },
    {
      "epoch": 19.300884955752213,
      "grad_norm": 0.01410099770873785,
      "learning_rate": 0.2994415725503263,
      "loss": 0.059,
      "num_input_tokens_seen": 625560,
      "step": 1100
    },
    {
      "epoch": 19.38938053097345,
      "grad_norm": 0.029469553381204605,
      "learning_rate": 0.29943648295021885,
      "loss": 0.0519,
      "num_input_tokens_seen": 628488,
      "step": 1105
    },
    {
      "epoch": 19.47787610619469,
      "grad_norm": 0.04847237095236778,
      "learning_rate": 0.2994313703051278,
      "loss": 0.145,
      "num_input_tokens_seen": 631400,
      "step": 1110
    },
    {
      "epoch": 19.56637168141593,
      "grad_norm": 0.007048665080219507,
      "learning_rate": 0.29942623461584156,
      "loss": 0.109,
      "num_input_tokens_seen": 634408,
      "step": 1115
    },
    {
      "epoch": 19.654867256637168,
      "grad_norm": 0.02925610914826393,
      "learning_rate": 0.29942107588315214,
      "loss": 0.0506,
      "num_input_tokens_seen": 637448,
      "step": 1120
    },
    {
      "epoch": 19.743362831858406,
      "grad_norm": 0.033245909959077835,
      "learning_rate": 0.29941589410785513,
      "loss": 0.0651,
      "num_input_tokens_seen": 640280,
      "step": 1125
    },
    {
      "epoch": 19.831858407079647,
      "grad_norm": 0.009714157320559025,
      "learning_rate": 0.29941068929074954,
      "loss": 0.14,
      "num_input_tokens_seen": 642984,
      "step": 1130
    },
    {
      "epoch": 19.920353982300885,
      "grad_norm": 0.01279335655272007,
      "learning_rate": 0.2994054614326381,
      "loss": 0.071,
      "num_input_tokens_seen": 646072,
      "step": 1135
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.09708935767412186,
      "learning_rate": 0.29940021053432686,
      "loss": 0.1207,
      "num_input_tokens_seen": 648736,
      "step": 1140
    },
    {
      "epoch": 20.088495575221238,
      "grad_norm": 0.11073679476976395,
      "learning_rate": 0.29939493659662575,
      "loss": 0.1115,
      "num_input_tokens_seen": 651264,
      "step": 1145
    },
    {
      "epoch": 20.17699115044248,
      "grad_norm": 0.012537207454442978,
      "learning_rate": 0.299389639620348,
      "loss": 0.1204,
      "num_input_tokens_seen": 654352,
      "step": 1150
    },
    {
      "epoch": 20.265486725663717,
      "grad_norm": 0.03436347469687462,
      "learning_rate": 0.29938431960631046,
      "loss": 0.0668,
      "num_input_tokens_seen": 657088,
      "step": 1155
    },
    {
      "epoch": 20.353982300884955,
      "grad_norm": 0.03144063800573349,
      "learning_rate": 0.2993789765553335,
      "loss": 0.1168,
      "num_input_tokens_seen": 659968,
      "step": 1160
    },
    {
      "epoch": 20.442477876106196,
      "grad_norm": 0.0384085439145565,
      "learning_rate": 0.2993736104682412,
      "loss": 0.0787,
      "num_input_tokens_seen": 663152,
      "step": 1165
    },
    {
      "epoch": 20.530973451327434,
      "grad_norm": 0.07563851773738861,
      "learning_rate": 0.299368221345861,
      "loss": 0.076,
      "num_input_tokens_seen": 666240,
      "step": 1170
    },
    {
      "epoch": 20.61946902654867,
      "grad_norm": 0.04717668518424034,
      "learning_rate": 0.29936280918902397,
      "loss": 0.1242,
      "num_input_tokens_seen": 668896,
      "step": 1175
    },
    {
      "epoch": 20.707964601769913,
      "grad_norm": 0.08111153542995453,
      "learning_rate": 0.2993573739985648,
      "loss": 0.0807,
      "num_input_tokens_seen": 671376,
      "step": 1180
    },
    {
      "epoch": 20.79646017699115,
      "grad_norm": 0.01971960999071598,
      "learning_rate": 0.2993519157753216,
      "loss": 0.061,
      "num_input_tokens_seen": 674304,
      "step": 1185
    },
    {
      "epoch": 20.884955752212388,
      "grad_norm": 0.008201570250093937,
      "learning_rate": 0.2993464345201361,
      "loss": 0.1526,
      "num_input_tokens_seen": 677152,
      "step": 1190
    },
    {
      "epoch": 20.97345132743363,
      "grad_norm": 0.040368225425481796,
      "learning_rate": 0.2993409302338536,
      "loss": 0.0303,
      "num_input_tokens_seen": 679840,
      "step": 1195
    },
    {
      "epoch": 21.053097345132745,
      "grad_norm": 0.06068111211061478,
      "learning_rate": 0.2993354029173229,
      "loss": 0.0923,
      "num_input_tokens_seen": 682024,
      "step": 1200
    },
    {
      "epoch": 21.053097345132745,
      "eval_loss": 0.3018835783004761,
      "eval_runtime": 0.9297,
      "eval_samples_per_second": 26.891,
      "eval_steps_per_second": 13.983,
      "num_input_tokens_seen": 682024,
      "step": 1200
    },
    {
      "epoch": 21.141592920353983,
      "grad_norm": 0.10846542567014694,
      "learning_rate": 0.2993298525713965,
      "loss": 0.1178,
      "num_input_tokens_seen": 685320,
      "step": 1205
    },
    {
      "epoch": 21.23008849557522,
      "grad_norm": 0.0035468339920043945,
      "learning_rate": 0.29932427919693017,
      "loss": 0.0531,
      "num_input_tokens_seen": 688200,
      "step": 1210
    },
    {
      "epoch": 21.31858407079646,
      "grad_norm": 0.02589237317442894,
      "learning_rate": 0.2993186827947834,
      "loss": 0.0576,
      "num_input_tokens_seen": 691432,
      "step": 1215
    },
    {
      "epoch": 21.4070796460177,
      "grad_norm": 0.07809528708457947,
      "learning_rate": 0.2993130633658194,
      "loss": 0.0765,
      "num_input_tokens_seen": 694088,
      "step": 1220
    },
    {
      "epoch": 21.495575221238937,
      "grad_norm": 0.004746662452816963,
      "learning_rate": 0.29930742091090456,
      "loss": 0.0183,
      "num_input_tokens_seen": 696568,
      "step": 1225
    },
    {
      "epoch": 21.58407079646018,
      "grad_norm": 0.07144982367753983,
      "learning_rate": 0.29930175543090914,
      "loss": 0.1573,
      "num_input_tokens_seen": 699128,
      "step": 1230
    },
    {
      "epoch": 21.672566371681416,
      "grad_norm": 0.024281533434987068,
      "learning_rate": 0.2992960669267068,
      "loss": 0.1005,
      "num_input_tokens_seen": 702936,
      "step": 1235
    },
    {
      "epoch": 21.761061946902654,
      "grad_norm": 0.052919622510671616,
      "learning_rate": 0.29929035539917476,
      "loss": 0.0991,
      "num_input_tokens_seen": 705560,
      "step": 1240
    },
    {
      "epoch": 21.849557522123895,
      "grad_norm": 0.021938621997833252,
      "learning_rate": 0.2992846208491938,
      "loss": 0.0506,
      "num_input_tokens_seen": 708312,
      "step": 1245
    },
    {
      "epoch": 21.938053097345133,
      "grad_norm": 0.01015006098896265,
      "learning_rate": 0.2992788632776483,
      "loss": 0.0501,
      "num_input_tokens_seen": 711224,
      "step": 1250
    },
    {
      "epoch": 22.01769911504425,
      "grad_norm": 0.05136142298579216,
      "learning_rate": 0.29927308268542613,
      "loss": 0.0967,
      "num_input_tokens_seen": 713400,
      "step": 1255
    },
    {
      "epoch": 22.106194690265486,
      "grad_norm": 0.05894184857606888,
      "learning_rate": 0.2992672790734187,
      "loss": 0.053,
      "num_input_tokens_seen": 716440,
      "step": 1260
    },
    {
      "epoch": 22.194690265486727,
      "grad_norm": 0.0012867054902017117,
      "learning_rate": 0.299261452442521,
      "loss": 0.0791,
      "num_input_tokens_seen": 719112,
      "step": 1265
    },
    {
      "epoch": 22.283185840707965,
      "grad_norm": 0.06244965270161629,
      "learning_rate": 0.29925560279363167,
      "loss": 0.07,
      "num_input_tokens_seen": 722520,
      "step": 1270
    },
    {
      "epoch": 22.371681415929203,
      "grad_norm": 0.0011850915616378188,
      "learning_rate": 0.29924973012765266,
      "loss": 0.0368,
      "num_input_tokens_seen": 726040,
      "step": 1275
    },
    {
      "epoch": 22.460176991150444,
      "grad_norm": 0.005481795407831669,
      "learning_rate": 0.29924383444548974,
      "loss": 0.026,
      "num_input_tokens_seen": 728712,
      "step": 1280
    },
    {
      "epoch": 22.548672566371682,
      "grad_norm": 0.0618191659450531,
      "learning_rate": 0.299237915748052,
      "loss": 0.1531,
      "num_input_tokens_seen": 731848,
      "step": 1285
    },
    {
      "epoch": 22.63716814159292,
      "grad_norm": 0.006729009561240673,
      "learning_rate": 0.2992319740362522,
      "loss": 0.0592,
      "num_input_tokens_seen": 734504,
      "step": 1290
    },
    {
      "epoch": 22.72566371681416,
      "grad_norm": 0.036971136927604675,
      "learning_rate": 0.2992260093110066,
      "loss": 0.0356,
      "num_input_tokens_seen": 737240,
      "step": 1295
    },
    {
      "epoch": 22.8141592920354,
      "grad_norm": 0.008717183955013752,
      "learning_rate": 0.2992200215732352,
      "loss": 0.0241,
      "num_input_tokens_seen": 740024,
      "step": 1300
    },
    {
      "epoch": 22.902654867256636,
      "grad_norm": 0.005968763958662748,
      "learning_rate": 0.2992140108238611,
      "loss": 0.0138,
      "num_input_tokens_seen": 742568,
      "step": 1305
    },
    {
      "epoch": 22.991150442477878,
      "grad_norm": 0.009583886712789536,
      "learning_rate": 0.2992079770638115,
      "loss": 0.0277,
      "num_input_tokens_seen": 745224,
      "step": 1310
    },
    {
      "epoch": 23.07079646017699,
      "grad_norm": 0.0741044208407402,
      "learning_rate": 0.29920192029401677,
      "loss": 0.1856,
      "num_input_tokens_seen": 747824,
      "step": 1315
    },
    {
      "epoch": 23.15929203539823,
      "grad_norm": 0.00021991966059431434,
      "learning_rate": 0.2991958405154109,
      "loss": 0.0163,
      "num_input_tokens_seen": 750752,
      "step": 1320
    },
    {
      "epoch": 23.24778761061947,
      "grad_norm": 0.05521894618868828,
      "learning_rate": 0.29918973772893154,
      "loss": 0.0799,
      "num_input_tokens_seen": 753872,
      "step": 1325
    },
    {
      "epoch": 23.336283185840706,
      "grad_norm": 0.0023225238546729088,
      "learning_rate": 0.29918361193551973,
      "loss": 0.0335,
      "num_input_tokens_seen": 756752,
      "step": 1330
    },
    {
      "epoch": 23.424778761061948,
      "grad_norm": 0.0033994922414422035,
      "learning_rate": 0.29917746313612026,
      "loss": 0.0395,
      "num_input_tokens_seen": 759328,
      "step": 1335
    },
    {
      "epoch": 23.513274336283185,
      "grad_norm": 0.008613375015556812,
      "learning_rate": 0.29917129133168124,
      "loss": 0.0506,
      "num_input_tokens_seen": 762224,
      "step": 1340
    },
    {
      "epoch": 23.601769911504427,
      "grad_norm": 0.04974130913615227,
      "learning_rate": 0.2991650965231546,
      "loss": 0.0822,
      "num_input_tokens_seen": 765344,
      "step": 1345
    },
    {
      "epoch": 23.690265486725664,
      "grad_norm": 0.005501857027411461,
      "learning_rate": 0.29915887871149544,
      "loss": 0.0781,
      "num_input_tokens_seen": 768256,
      "step": 1350
    },
    {
      "epoch": 23.778761061946902,
      "grad_norm": 0.017547132447361946,
      "learning_rate": 0.2991526378976628,
      "loss": 0.0634,
      "num_input_tokens_seen": 770960,
      "step": 1355
    },
    {
      "epoch": 23.86725663716814,
      "grad_norm": 0.028280172497034073,
      "learning_rate": 0.29914637408261896,
      "loss": 0.1649,
      "num_input_tokens_seen": 773584,
      "step": 1360
    },
    {
      "epoch": 23.95575221238938,
      "grad_norm": 0.008519790135324001,
      "learning_rate": 0.29914008726733,
      "loss": 0.0458,
      "num_input_tokens_seen": 776848,
      "step": 1365
    },
    {
      "epoch": 24.035398230088497,
      "grad_norm": 0.004154284950345755,
      "learning_rate": 0.2991337774527653,
      "loss": 0.0702,
      "num_input_tokens_seen": 779304,
      "step": 1370
    },
    {
      "epoch": 24.123893805309734,
      "grad_norm": 0.0028321868740022182,
      "learning_rate": 0.2991274446398981,
      "loss": 0.0177,
      "num_input_tokens_seen": 782088,
      "step": 1375
    },
    {
      "epoch": 24.212389380530972,
      "grad_norm": 0.014611666090786457,
      "learning_rate": 0.29912108882970484,
      "loss": 0.0222,
      "num_input_tokens_seen": 785080,
      "step": 1380
    },
    {
      "epoch": 24.300884955752213,
      "grad_norm": 0.008415494114160538,
      "learning_rate": 0.2991147100231657,
      "loss": 0.0544,
      "num_input_tokens_seen": 787800,
      "step": 1385
    },
    {
      "epoch": 24.38938053097345,
      "grad_norm": 0.07670147716999054,
      "learning_rate": 0.2991083082212644,
      "loss": 0.0466,
      "num_input_tokens_seen": 790424,
      "step": 1390
    },
    {
      "epoch": 24.47787610619469,
      "grad_norm": 0.0016266423044726253,
      "learning_rate": 0.2991018834249881,
      "loss": 0.0114,
      "num_input_tokens_seen": 793032,
      "step": 1395
    },
    {
      "epoch": 24.56637168141593,
      "grad_norm": 0.07207012176513672,
      "learning_rate": 0.29909543563532764,
      "loss": 0.0482,
      "num_input_tokens_seen": 796328,
      "step": 1400
    },
    {
      "epoch": 24.56637168141593,
      "eval_loss": 0.20980682969093323,
      "eval_runtime": 0.9402,
      "eval_samples_per_second": 26.59,
      "eval_steps_per_second": 13.827,
      "num_input_tokens_seen": 796328,
      "step": 1400
    },
    {
      "epoch": 24.654867256637168,
      "grad_norm": 0.011100090108811855,
      "learning_rate": 0.29908896485327746,
      "loss": 0.0384,
      "num_input_tokens_seen": 799176,
      "step": 1405
    },
    {
      "epoch": 24.743362831858406,
      "grad_norm": 0.036893852055072784,
      "learning_rate": 0.29908247107983527,
      "loss": 0.0925,
      "num_input_tokens_seen": 802072,
      "step": 1410
    },
    {
      "epoch": 24.831858407079647,
      "grad_norm": 0.007857692427933216,
      "learning_rate": 0.29907595431600253,
      "loss": 0.0166,
      "num_input_tokens_seen": 804648,
      "step": 1415
    },
    {
      "epoch": 24.920353982300885,
      "grad_norm": 0.1398167908191681,
      "learning_rate": 0.29906941456278424,
      "loss": 0.0461,
      "num_input_tokens_seen": 807368,
      "step": 1420
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.0017228537471964955,
      "learning_rate": 0.2990628518211889,
      "loss": 0.038,
      "num_input_tokens_seen": 810192,
      "step": 1425
    },
    {
      "epoch": 25.088495575221238,
      "grad_norm": 0.003742335131391883,
      "learning_rate": 0.2990562660922286,
      "loss": 0.0423,
      "num_input_tokens_seen": 812960,
      "step": 1430
    },
    {
      "epoch": 25.17699115044248,
      "grad_norm": 0.027222873643040657,
      "learning_rate": 0.2990496573769189,
      "loss": 0.067,
      "num_input_tokens_seen": 816256,
      "step": 1435
    },
    {
      "epoch": 25.265486725663717,
      "grad_norm": 0.0814589336514473,
      "learning_rate": 0.29904302567627894,
      "loss": 0.0506,
      "num_input_tokens_seen": 819232,
      "step": 1440
    },
    {
      "epoch": 25.353982300884955,
      "grad_norm": 0.005998051259666681,
      "learning_rate": 0.2990363709913314,
      "loss": 0.0522,
      "num_input_tokens_seen": 822080,
      "step": 1445
    },
    {
      "epoch": 25.442477876106196,
      "grad_norm": 0.06759881973266602,
      "learning_rate": 0.29902969332310264,
      "loss": 0.0548,
      "num_input_tokens_seen": 824688,
      "step": 1450
    },
    {
      "epoch": 25.530973451327434,
      "grad_norm": 0.03362750634551048,
      "learning_rate": 0.2990229926726223,
      "loss": 0.0222,
      "num_input_tokens_seen": 827360,
      "step": 1455
    },
    {
      "epoch": 25.61946902654867,
      "grad_norm": 0.059995878487825394,
      "learning_rate": 0.29901626904092365,
      "loss": 0.0337,
      "num_input_tokens_seen": 830848,
      "step": 1460
    },
    {
      "epoch": 25.707964601769913,
      "grad_norm": 0.008152305148541927,
      "learning_rate": 0.2990095224290438,
      "loss": 0.0441,
      "num_input_tokens_seen": 833264,
      "step": 1465
    },
    {
      "epoch": 25.79646017699115,
      "grad_norm": 0.028977930545806885,
      "learning_rate": 0.29900275283802297,
      "loss": 0.0138,
      "num_input_tokens_seen": 836480,
      "step": 1470
    },
    {
      "epoch": 25.884955752212388,
      "grad_norm": 0.0021785518620163202,
      "learning_rate": 0.2989959602689051,
      "loss": 0.0224,
      "num_input_tokens_seen": 839168,
      "step": 1475
    },
    {
      "epoch": 25.97345132743363,
      "grad_norm": 0.0018287242855876684,
      "learning_rate": 0.2989891447227379,
      "loss": 0.0495,
      "num_input_tokens_seen": 842160,
      "step": 1480
    },
    {
      "epoch": 26.053097345132745,
      "grad_norm": 0.022007077932357788,
      "learning_rate": 0.29898230620057215,
      "loss": 0.017,
      "num_input_tokens_seen": 844392,
      "step": 1485
    },
    {
      "epoch": 26.141592920353983,
      "grad_norm": 0.006985585205256939,
      "learning_rate": 0.2989754447034626,
      "loss": 0.007,
      "num_input_tokens_seen": 847464,
      "step": 1490
    },
    {
      "epoch": 26.23008849557522,
      "grad_norm": 0.027698179706931114,
      "learning_rate": 0.2989685602324673,
      "loss": 0.018,
      "num_input_tokens_seen": 849752,
      "step": 1495
    },
    {
      "epoch": 26.31858407079646,
      "grad_norm": 0.003572483779862523,
      "learning_rate": 0.298961652788648,
      "loss": 0.0022,
      "num_input_tokens_seen": 852632,
      "step": 1500
    },
    {
      "epoch": 26.4070796460177,
      "grad_norm": 0.0845576599240303,
      "learning_rate": 0.29895472237306986,
      "loss": 0.0479,
      "num_input_tokens_seen": 855640,
      "step": 1505
    },
    {
      "epoch": 26.495575221238937,
      "grad_norm": 0.0020605502650141716,
      "learning_rate": 0.29894776898680164,
      "loss": 0.0065,
      "num_input_tokens_seen": 858728,
      "step": 1510
    },
    {
      "epoch": 26.58407079646018,
      "grad_norm": 0.06568868458271027,
      "learning_rate": 0.29894079263091566,
      "loss": 0.0767,
      "num_input_tokens_seen": 861736,
      "step": 1515
    },
    {
      "epoch": 26.672566371681416,
      "grad_norm": 0.04822291433811188,
      "learning_rate": 0.2989337933064877,
      "loss": 0.0383,
      "num_input_tokens_seen": 864248,
      "step": 1520
    },
    {
      "epoch": 26.761061946902654,
      "grad_norm": 0.11739757657051086,
      "learning_rate": 0.29892677101459725,
      "loss": 0.1148,
      "num_input_tokens_seen": 867288,
      "step": 1525
    },
    {
      "epoch": 26.849557522123895,
      "grad_norm": 0.10341432690620422,
      "learning_rate": 0.2989197257563272,
      "loss": 0.1161,
      "num_input_tokens_seen": 870536,
      "step": 1530
    },
    {
      "epoch": 26.938053097345133,
      "grad_norm": 0.06631104648113251,
      "learning_rate": 0.2989126575327639,
      "loss": 0.0718,
      "num_input_tokens_seen": 873048,
      "step": 1535
    },
    {
      "epoch": 27.01769911504425,
      "grad_norm": 0.007914594374597073,
      "learning_rate": 0.29890556634499754,
      "loss": 0.073,
      "num_input_tokens_seen": 875304,
      "step": 1540
    },
    {
      "epoch": 27.106194690265486,
      "grad_norm": 0.004657643381506205,
      "learning_rate": 0.2988984521941216,
      "loss": 0.0566,
      "num_input_tokens_seen": 878312,
      "step": 1545
    },
    {
      "epoch": 27.194690265486727,
      "grad_norm": 0.04576535150408745,
      "learning_rate": 0.29889131508123307,
      "loss": 0.0811,
      "num_input_tokens_seen": 881192,
      "step": 1550
    },
    {
      "epoch": 27.283185840707965,
      "grad_norm": 0.005946827586740255,
      "learning_rate": 0.2988841550074327,
      "loss": 0.1027,
      "num_input_tokens_seen": 883896,
      "step": 1555
    },
    {
      "epoch": 27.371681415929203,
      "grad_norm": 0.09602929651737213,
      "learning_rate": 0.2988769719738246,
      "loss": 0.1841,
      "num_input_tokens_seen": 886376,
      "step": 1560
    },
    {
      "epoch": 27.460176991150444,
      "grad_norm": 0.005651470273733139,
      "learning_rate": 0.29886976598151666,
      "loss": 0.0983,
      "num_input_tokens_seen": 889384,
      "step": 1565
    },
    {
      "epoch": 27.548672566371682,
      "grad_norm": 0.006867815740406513,
      "learning_rate": 0.29886253703161986,
      "loss": 0.0125,
      "num_input_tokens_seen": 892344,
      "step": 1570
    },
    {
      "epoch": 27.63716814159292,
      "grad_norm": 0.020086800679564476,
      "learning_rate": 0.29885528512524917,
      "loss": 0.0948,
      "num_input_tokens_seen": 895048,
      "step": 1575
    },
    {
      "epoch": 27.72566371681416,
      "grad_norm": 0.005952614359557629,
      "learning_rate": 0.29884801026352287,
      "loss": 0.3111,
      "num_input_tokens_seen": 897992,
      "step": 1580
    },
    {
      "epoch": 27.8141592920354,
      "grad_norm": 0.010969198308885098,
      "learning_rate": 0.2988407124475629,
      "loss": 0.1454,
      "num_input_tokens_seen": 900888,
      "step": 1585
    },
    {
      "epoch": 27.902654867256636,
      "grad_norm": 0.008461329154670238,
      "learning_rate": 0.2988333916784945,
      "loss": 0.1315,
      "num_input_tokens_seen": 904072,
      "step": 1590
    },
    {
      "epoch": 27.991150442477878,
      "grad_norm": 0.09626653045415878,
      "learning_rate": 0.2988260479574468,
      "loss": 0.1699,
      "num_input_tokens_seen": 907016,
      "step": 1595
    },
    {
      "epoch": 28.07079646017699,
      "grad_norm": 0.02973833493888378,
      "learning_rate": 0.2988186812855523,
      "loss": 0.1297,
      "num_input_tokens_seen": 909320,
      "step": 1600
    },
    {
      "epoch": 28.07079646017699,
      "eval_loss": 0.1714715212583542,
      "eval_runtime": 0.9186,
      "eval_samples_per_second": 27.214,
      "eval_steps_per_second": 14.151,
      "num_input_tokens_seen": 909320,
      "step": 1600
    },
    {
      "epoch": 28.15929203539823,
      "grad_norm": 0.013810357078909874,
      "learning_rate": 0.29881129166394693,
      "loss": 0.0483,
      "num_input_tokens_seen": 912536,
      "step": 1605
    },
    {
      "epoch": 28.24778761061947,
      "grad_norm": 0.0178288035094738,
      "learning_rate": 0.29880387909377026,
      "loss": 0.178,
      "num_input_tokens_seen": 915480,
      "step": 1610
    },
    {
      "epoch": 28.336283185840706,
      "grad_norm": 0.10714978724718094,
      "learning_rate": 0.2987964435761655,
      "loss": 0.2568,
      "num_input_tokens_seen": 918344,
      "step": 1615
    },
    {
      "epoch": 28.424778761061948,
      "grad_norm": 0.013234782963991165,
      "learning_rate": 0.29878898511227925,
      "loss": 0.1044,
      "num_input_tokens_seen": 920808,
      "step": 1620
    },
    {
      "epoch": 28.513274336283185,
      "grad_norm": 0.04466245695948601,
      "learning_rate": 0.2987815037032617,
      "loss": 0.1365,
      "num_input_tokens_seen": 923976,
      "step": 1625
    },
    {
      "epoch": 28.601769911504427,
      "grad_norm": 0.014913058839738369,
      "learning_rate": 0.29877399935026655,
      "loss": 0.0472,
      "num_input_tokens_seen": 926568,
      "step": 1630
    },
    {
      "epoch": 28.690265486725664,
      "grad_norm": 0.03534584864974022,
      "learning_rate": 0.2987664720544511,
      "loss": 0.0901,
      "num_input_tokens_seen": 929512,
      "step": 1635
    },
    {
      "epoch": 28.778761061946902,
      "grad_norm": 0.0429089292883873,
      "learning_rate": 0.2987589218169761,
      "loss": 0.0837,
      "num_input_tokens_seen": 932424,
      "step": 1640
    },
    {
      "epoch": 28.86725663716814,
      "grad_norm": 0.04179130867123604,
      "learning_rate": 0.29875134863900604,
      "loss": 0.033,
      "num_input_tokens_seen": 935496,
      "step": 1645
    },
    {
      "epoch": 28.95575221238938,
      "grad_norm": 0.018635129556059837,
      "learning_rate": 0.29874375252170865,
      "loss": 0.0218,
      "num_input_tokens_seen": 938664,
      "step": 1650
    },
    {
      "epoch": 29.035398230088497,
      "grad_norm": 0.008180464617908001,
      "learning_rate": 0.2987361334662553,
      "loss": 0.0077,
      "num_input_tokens_seen": 940776,
      "step": 1655
    },
    {
      "epoch": 29.123893805309734,
      "grad_norm": 0.004545401781797409,
      "learning_rate": 0.29872849147382113,
      "loss": 0.005,
      "num_input_tokens_seen": 943624,
      "step": 1660
    },
    {
      "epoch": 29.212389380530972,
      "grad_norm": 0.04973456263542175,
      "learning_rate": 0.2987208265455845,
      "loss": 0.0639,
      "num_input_tokens_seen": 946728,
      "step": 1665
    },
    {
      "epoch": 29.300884955752213,
      "grad_norm": 0.009610278531908989,
      "learning_rate": 0.29871313868272753,
      "loss": 0.2009,
      "num_input_tokens_seen": 949880,
      "step": 1670
    },
    {
      "epoch": 29.38938053097345,
      "grad_norm": 0.02161199040710926,
      "learning_rate": 0.29870542788643567,
      "loss": 0.1236,
      "num_input_tokens_seen": 952376,
      "step": 1675
    },
    {
      "epoch": 29.47787610619469,
      "grad_norm": 0.00747729791328311,
      "learning_rate": 0.2986976941578981,
      "loss": 0.111,
      "num_input_tokens_seen": 955272,
      "step": 1680
    },
    {
      "epoch": 29.56637168141593,
      "grad_norm": 0.037067610770463943,
      "learning_rate": 0.29868993749830747,
      "loss": 0.0864,
      "num_input_tokens_seen": 958376,
      "step": 1685
    },
    {
      "epoch": 29.654867256637168,
      "grad_norm": 0.010913576930761337,
      "learning_rate": 0.2986821579088598,
      "loss": 0.122,
      "num_input_tokens_seen": 960952,
      "step": 1690
    },
    {
      "epoch": 29.743362831858406,
      "grad_norm": 0.017758822068572044,
      "learning_rate": 0.29867435539075504,
      "loss": 0.0474,
      "num_input_tokens_seen": 963752,
      "step": 1695
    },
    {
      "epoch": 29.831858407079647,
      "grad_norm": 0.002401971723884344,
      "learning_rate": 0.2986665299451963,
      "loss": 0.0371,
      "num_input_tokens_seen": 967032,
      "step": 1700
    },
    {
      "epoch": 29.920353982300885,
      "grad_norm": 0.01882217451930046,
      "learning_rate": 0.29865868157339037,
      "loss": 0.0414,
      "num_input_tokens_seen": 969672,
      "step": 1705
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.00020096651860512793,
      "learning_rate": 0.2986508102765476,
      "loss": 0.1137,
      "num_input_tokens_seen": 972184,
      "step": 1710
    },
    {
      "epoch": 30.088495575221238,
      "grad_norm": 0.030124805867671967,
      "learning_rate": 0.2986429160558818,
      "loss": 0.0351,
      "num_input_tokens_seen": 974888,
      "step": 1715
    },
    {
      "epoch": 30.17699115044248,
      "grad_norm": 0.016149351373314857,
      "learning_rate": 0.2986349989126104,
      "loss": 0.0394,
      "num_input_tokens_seen": 977368,
      "step": 1720
    },
    {
      "epoch": 30.265486725663717,
      "grad_norm": 0.061579059809446335,
      "learning_rate": 0.29862705884795426,
      "loss": 0.0692,
      "num_input_tokens_seen": 980696,
      "step": 1725
    },
    {
      "epoch": 30.353982300884955,
      "grad_norm": 0.003947456367313862,
      "learning_rate": 0.2986190958631379,
      "loss": 0.0296,
      "num_input_tokens_seen": 983288,
      "step": 1730
    },
    {
      "epoch": 30.442477876106196,
      "grad_norm": 0.006744958460330963,
      "learning_rate": 0.29861110995938933,
      "loss": 0.0523,
      "num_input_tokens_seen": 986520,
      "step": 1735
    },
    {
      "epoch": 30.530973451327434,
      "grad_norm": 0.005091800820082426,
      "learning_rate": 0.29860310113794,
      "loss": 0.0182,
      "num_input_tokens_seen": 989496,
      "step": 1740
    },
    {
      "epoch": 30.61946902654867,
      "grad_norm": 0.04587063565850258,
      "learning_rate": 0.29859506940002506,
      "loss": 0.0408,
      "num_input_tokens_seen": 992552,
      "step": 1745
    },
    {
      "epoch": 30.707964601769913,
      "grad_norm": 0.0051162526942789555,
      "learning_rate": 0.298587014746883,
      "loss": 0.036,
      "num_input_tokens_seen": 995448,
      "step": 1750
    },
    {
      "epoch": 30.79646017699115,
      "grad_norm": 0.009663380682468414,
      "learning_rate": 0.298578937179756,
      "loss": 0.0337,
      "num_input_tokens_seen": 998504,
      "step": 1755
    },
    {
      "epoch": 30.884955752212388,
      "grad_norm": 0.002576402621343732,
      "learning_rate": 0.29857083669988976,
      "loss": 0.0551,
      "num_input_tokens_seen": 1001352,
      "step": 1760
    },
    {
      "epoch": 30.97345132743363,
      "grad_norm": 0.07329577207565308,
      "learning_rate": 0.29856271330853346,
      "loss": 0.0463,
      "num_input_tokens_seen": 1003896,
      "step": 1765
    },
    {
      "epoch": 31.053097345132745,
      "grad_norm": 0.001554815680719912,
      "learning_rate": 0.2985545670069398,
      "loss": 0.0156,
      "num_input_tokens_seen": 1006384,
      "step": 1770
    },
    {
      "epoch": 31.141592920353983,
      "grad_norm": 0.06041880324482918,
      "learning_rate": 0.29854639779636505,
      "loss": 0.0525,
      "num_input_tokens_seen": 1009600,
      "step": 1775
    },
    {
      "epoch": 31.23008849557522,
      "grad_norm": 0.03961880877614021,
      "learning_rate": 0.298538205678069,
      "loss": 0.0377,
      "num_input_tokens_seen": 1012736,
      "step": 1780
    },
    {
      "epoch": 31.31858407079646,
      "grad_norm": 0.0006600599735975266,
      "learning_rate": 0.298529990653315,
      "loss": 0.0034,
      "num_input_tokens_seen": 1015632,
      "step": 1785
    },
    {
      "epoch": 31.4070796460177,
      "grad_norm": 0.050408147275447845,
      "learning_rate": 0.29852175272336984,
      "loss": 0.0425,
      "num_input_tokens_seen": 1018624,
      "step": 1790
    },
    {
      "epoch": 31.495575221238937,
      "grad_norm": 0.019935214892029762,
      "learning_rate": 0.29851349188950405,
      "loss": 0.0381,
      "num_input_tokens_seen": 1021152,
      "step": 1795
    },
    {
      "epoch": 31.58407079646018,
      "grad_norm": 0.0744040235877037,
      "learning_rate": 0.2985052081529914,
      "loss": 0.0484,
      "num_input_tokens_seen": 1023696,
      "step": 1800
    },
    {
      "epoch": 31.58407079646018,
      "eval_loss": 0.28920552134513855,
      "eval_runtime": 0.9363,
      "eval_samples_per_second": 26.7,
      "eval_steps_per_second": 13.884,
      "num_input_tokens_seen": 1023696,
      "step": 1800
    },
    {
      "epoch": 31.672566371681416,
      "grad_norm": 0.001106122275814414,
      "learning_rate": 0.29849690151510944,
      "loss": 0.0156,
      "num_input_tokens_seen": 1026400,
      "step": 1805
    },
    {
      "epoch": 31.761061946902654,
      "grad_norm": 0.046785518527030945,
      "learning_rate": 0.2984885719771392,
      "loss": 0.122,
      "num_input_tokens_seen": 1029040,
      "step": 1810
    },
    {
      "epoch": 31.849557522123895,
      "grad_norm": 0.04179898276925087,
      "learning_rate": 0.2984802195403651,
      "loss": 0.0581,
      "num_input_tokens_seen": 1032032,
      "step": 1815
    },
    {
      "epoch": 31.938053097345133,
      "grad_norm": 0.00936127919703722,
      "learning_rate": 0.2984718442060752,
      "loss": 0.0284,
      "num_input_tokens_seen": 1034912,
      "step": 1820
    },
    {
      "epoch": 32.017699115044245,
      "grad_norm": 0.020806491374969482,
      "learning_rate": 0.2984634459755611,
      "loss": 0.0336,
      "num_input_tokens_seen": 1037416,
      "step": 1825
    },
    {
      "epoch": 32.10619469026549,
      "grad_norm": 0.041175976395606995,
      "learning_rate": 0.29845502485011793,
      "loss": 0.0165,
      "num_input_tokens_seen": 1040040,
      "step": 1830
    },
    {
      "epoch": 32.19469026548673,
      "grad_norm": 0.003511570394039154,
      "learning_rate": 0.2984465808310444,
      "loss": 0.0032,
      "num_input_tokens_seen": 1043288,
      "step": 1835
    },
    {
      "epoch": 32.283185840707965,
      "grad_norm": 0.0015427340986207128,
      "learning_rate": 0.29843811391964253,
      "loss": 0.0475,
      "num_input_tokens_seen": 1045896,
      "step": 1840
    },
    {
      "epoch": 32.3716814159292,
      "grad_norm": 0.011623822152614594,
      "learning_rate": 0.2984296241172182,
      "loss": 0.0094,
      "num_input_tokens_seen": 1048792,
      "step": 1845
    },
    {
      "epoch": 32.46017699115044,
      "grad_norm": 0.0001682483562035486,
      "learning_rate": 0.29842111142508043,
      "loss": 0.0068,
      "num_input_tokens_seen": 1051352,
      "step": 1850
    },
    {
      "epoch": 32.54867256637168,
      "grad_norm": 0.0014628011267632246,
      "learning_rate": 0.29841257584454217,
      "loss": 0.0116,
      "num_input_tokens_seen": 1054520,
      "step": 1855
    },
    {
      "epoch": 32.63716814159292,
      "grad_norm": 0.00035647553158923984,
      "learning_rate": 0.29840401737691963,
      "loss": 0.0049,
      "num_input_tokens_seen": 1057768,
      "step": 1860
    },
    {
      "epoch": 32.72566371681416,
      "grad_norm": 0.05992606282234192,
      "learning_rate": 0.29839543602353263,
      "loss": 0.1039,
      "num_input_tokens_seen": 1060936,
      "step": 1865
    },
    {
      "epoch": 32.8141592920354,
      "grad_norm": 0.015231838449835777,
      "learning_rate": 0.2983868317857046,
      "loss": 0.1211,
      "num_input_tokens_seen": 1063880,
      "step": 1870
    },
    {
      "epoch": 32.902654867256636,
      "grad_norm": 0.01562560722231865,
      "learning_rate": 0.2983782046647623,
      "loss": 0.0378,
      "num_input_tokens_seen": 1066616,
      "step": 1875
    },
    {
      "epoch": 32.991150442477874,
      "grad_norm": 0.008026563562452793,
      "learning_rate": 0.2983695546620362,
      "loss": 0.0457,
      "num_input_tokens_seen": 1069080,
      "step": 1880
    },
    {
      "epoch": 33.07079646017699,
      "grad_norm": 0.0028766130562871695,
      "learning_rate": 0.2983608817788603,
      "loss": 0.0156,
      "num_input_tokens_seen": 1071712,
      "step": 1885
    },
    {
      "epoch": 33.15929203539823,
      "grad_norm": 0.03110269270837307,
      "learning_rate": 0.29835218601657193,
      "loss": 0.0176,
      "num_input_tokens_seen": 1074608,
      "step": 1890
    },
    {
      "epoch": 33.24778761061947,
      "grad_norm": 0.0035526317078620195,
      "learning_rate": 0.29834346737651224,
      "loss": 0.0398,
      "num_input_tokens_seen": 1077168,
      "step": 1895
    },
    {
      "epoch": 33.336283185840706,
      "grad_norm": 0.09152396023273468,
      "learning_rate": 0.29833472586002563,
      "loss": 0.1019,
      "num_input_tokens_seen": 1079920,
      "step": 1900
    },
    {
      "epoch": 33.424778761061944,
      "grad_norm": 0.0025378770660609007,
      "learning_rate": 0.29832596146846024,
      "loss": 0.0171,
      "num_input_tokens_seen": 1082816,
      "step": 1905
    },
    {
      "epoch": 33.51327433628319,
      "grad_norm": 0.0006385207525454462,
      "learning_rate": 0.2983171742031676,
      "loss": 0.0375,
      "num_input_tokens_seen": 1085600,
      "step": 1910
    },
    {
      "epoch": 33.60176991150443,
      "grad_norm": 0.020058417692780495,
      "learning_rate": 0.2983083640655028,
      "loss": 0.0725,
      "num_input_tokens_seen": 1088464,
      "step": 1915
    },
    {
      "epoch": 33.690265486725664,
      "grad_norm": 0.03535565361380577,
      "learning_rate": 0.29829953105682455,
      "loss": 0.0217,
      "num_input_tokens_seen": 1091648,
      "step": 1920
    },
    {
      "epoch": 33.7787610619469,
      "grad_norm": 0.0005175524274818599,
      "learning_rate": 0.29829067517849495,
      "loss": 0.0181,
      "num_input_tokens_seen": 1094080,
      "step": 1925
    },
    {
      "epoch": 33.86725663716814,
      "grad_norm": 0.002039116807281971,
      "learning_rate": 0.2982817964318797,
      "loss": 0.0483,
      "num_input_tokens_seen": 1097344,
      "step": 1930
    },
    {
      "epoch": 33.95575221238938,
      "grad_norm": 0.019467122852802277,
      "learning_rate": 0.298272894818348,
      "loss": 0.0238,
      "num_input_tokens_seen": 1100032,
      "step": 1935
    },
    {
      "epoch": 34.0353982300885,
      "grad_norm": 0.01485920324921608,
      "learning_rate": 0.2982639703392726,
      "loss": 0.2296,
      "num_input_tokens_seen": 1102528,
      "step": 1940
    },
    {
      "epoch": 34.123893805309734,
      "grad_norm": 0.045919399708509445,
      "learning_rate": 0.29825502299602974,
      "loss": 0.0699,
      "num_input_tokens_seen": 1105648,
      "step": 1945
    },
    {
      "epoch": 34.21238938053097,
      "grad_norm": 0.053987715393304825,
      "learning_rate": 0.2982460527899993,
      "loss": 0.0452,
      "num_input_tokens_seen": 1108656,
      "step": 1950
    },
    {
      "epoch": 34.30088495575221,
      "grad_norm": 0.11219354718923569,
      "learning_rate": 0.29823705972256453,
      "loss": 0.1482,
      "num_input_tokens_seen": 1111504,
      "step": 1955
    },
    {
      "epoch": 34.389380530973455,
      "grad_norm": 0.008548084646463394,
      "learning_rate": 0.2982280437951123,
      "loss": 0.044,
      "num_input_tokens_seen": 1114448,
      "step": 1960
    },
    {
      "epoch": 34.47787610619469,
      "grad_norm": 0.07235093414783478,
      "learning_rate": 0.298219005009033,
      "loss": 0.2286,
      "num_input_tokens_seen": 1118144,
      "step": 1965
    },
    {
      "epoch": 34.56637168141593,
      "grad_norm": 0.25690484046936035,
      "learning_rate": 0.29820994336572043,
      "loss": 0.5325,
      "num_input_tokens_seen": 1120592,
      "step": 1970
    },
    {
      "epoch": 34.65486725663717,
      "grad_norm": 0.11676662415266037,
      "learning_rate": 0.2982008588665721,
      "loss": 0.3398,
      "num_input_tokens_seen": 1123024,
      "step": 1975
    },
    {
      "epoch": 34.743362831858406,
      "grad_norm": 0.07518859952688217,
      "learning_rate": 0.2981917515129889,
      "loss": 0.2636,
      "num_input_tokens_seen": 1126016,
      "step": 1980
    },
    {
      "epoch": 34.83185840707964,
      "grad_norm": 0.010378862731158733,
      "learning_rate": 0.2981826213063753,
      "loss": 0.0812,
      "num_input_tokens_seen": 1129056,
      "step": 1985
    },
    {
      "epoch": 34.92035398230089,
      "grad_norm": 0.052443962544202805,
      "learning_rate": 0.2981734682481394,
      "loss": 0.2229,
      "num_input_tokens_seen": 1131600,
      "step": 1990
    },
    {
      "epoch": 35.0,
      "grad_norm": 0.04272105544805527,
      "learning_rate": 0.29816429233969255,
      "loss": 0.1478,
      "num_input_tokens_seen": 1134032,
      "step": 1995
    },
    {
      "epoch": 35.08849557522124,
      "grad_norm": 0.06886915117502213,
      "learning_rate": 0.2981550935824499,
      "loss": 0.1897,
      "num_input_tokens_seen": 1137280,
      "step": 2000
    },
    {
      "epoch": 35.08849557522124,
      "eval_loss": 0.3682709038257599,
      "eval_runtime": 0.9332,
      "eval_samples_per_second": 26.79,
      "eval_steps_per_second": 13.931,
      "num_input_tokens_seen": 1137280,
      "step": 2000
    },
    {
      "epoch": 35.176991150442475,
      "grad_norm": 0.010127265006303787,
      "learning_rate": 0.29814587197783,
      "loss": 0.0925,
      "num_input_tokens_seen": 1139856,
      "step": 2005
    },
    {
      "epoch": 35.26548672566372,
      "grad_norm": 0.010259071364998817,
      "learning_rate": 0.29813662752725495,
      "loss": 0.2344,
      "num_input_tokens_seen": 1142800,
      "step": 2010
    },
    {
      "epoch": 35.35398230088496,
      "grad_norm": 0.03356510400772095,
      "learning_rate": 0.29812736023215025,
      "loss": 0.1673,
      "num_input_tokens_seen": 1145824,
      "step": 2015
    },
    {
      "epoch": 35.442477876106196,
      "grad_norm": 0.0005747402901761234,
      "learning_rate": 0.29811807009394514,
      "loss": 0.1207,
      "num_input_tokens_seen": 1148736,
      "step": 2020
    },
    {
      "epoch": 35.530973451327434,
      "grad_norm": 0.00039747063419781625,
      "learning_rate": 0.2981087571140723,
      "loss": 0.0546,
      "num_input_tokens_seen": 1151760,
      "step": 2025
    },
    {
      "epoch": 35.61946902654867,
      "grad_norm": 0.006995708215981722,
      "learning_rate": 0.2980994212939678,
      "loss": 0.285,
      "num_input_tokens_seen": 1154496,
      "step": 2030
    },
    {
      "epoch": 35.70796460176991,
      "grad_norm": 0.03026936948299408,
      "learning_rate": 0.2980900626350715,
      "loss": 0.1518,
      "num_input_tokens_seen": 1157088,
      "step": 2035
    },
    {
      "epoch": 35.796460176991154,
      "grad_norm": 0.012430732138454914,
      "learning_rate": 0.29808068113882646,
      "loss": 0.1007,
      "num_input_tokens_seen": 1159824,
      "step": 2040
    },
    {
      "epoch": 35.88495575221239,
      "grad_norm": 0.02682667039334774,
      "learning_rate": 0.2980712768066795,
      "loss": 0.1987,
      "num_input_tokens_seen": 1162432,
      "step": 2045
    },
    {
      "epoch": 35.97345132743363,
      "grad_norm": 0.005583827383816242,
      "learning_rate": 0.2980618496400809,
      "loss": 0.1475,
      "num_input_tokens_seen": 1165680,
      "step": 2050
    },
    {
      "epoch": 36.05309734513274,
      "grad_norm": 0.006933813448995352,
      "learning_rate": 0.2980523996404844,
      "loss": 0.0823,
      "num_input_tokens_seen": 1168136,
      "step": 2055
    },
    {
      "epoch": 36.14159292035398,
      "grad_norm": 0.012457326054573059,
      "learning_rate": 0.2980429268093473,
      "loss": 0.0938,
      "num_input_tokens_seen": 1171576,
      "step": 2060
    },
    {
      "epoch": 36.230088495575224,
      "grad_norm": 0.006737505551427603,
      "learning_rate": 0.29803343114813047,
      "loss": 0.0864,
      "num_input_tokens_seen": 1174360,
      "step": 2065
    },
    {
      "epoch": 36.31858407079646,
      "grad_norm": 0.006521584931761026,
      "learning_rate": 0.2980239126582983,
      "loss": 0.0472,
      "num_input_tokens_seen": 1177240,
      "step": 2070
    },
    {
      "epoch": 36.4070796460177,
      "grad_norm": 0.002053737174719572,
      "learning_rate": 0.2980143713413186,
      "loss": 0.0901,
      "num_input_tokens_seen": 1179896,
      "step": 2075
    },
    {
      "epoch": 36.49557522123894,
      "grad_norm": 0.0017744585638865829,
      "learning_rate": 0.29800480719866274,
      "loss": 0.0332,
      "num_input_tokens_seen": 1182664,
      "step": 2080
    },
    {
      "epoch": 36.584070796460175,
      "grad_norm": 0.021712226793169975,
      "learning_rate": 0.2979952202318057,
      "loss": 0.0669,
      "num_input_tokens_seen": 1185304,
      "step": 2085
    },
    {
      "epoch": 36.67256637168141,
      "grad_norm": 0.038005631417036057,
      "learning_rate": 0.2979856104422259,
      "loss": 0.1202,
      "num_input_tokens_seen": 1188536,
      "step": 2090
    },
    {
      "epoch": 36.76106194690266,
      "grad_norm": 0.008423271588981152,
      "learning_rate": 0.2979759778314052,
      "loss": 0.0335,
      "num_input_tokens_seen": 1191304,
      "step": 2095
    },
    {
      "epoch": 36.849557522123895,
      "grad_norm": 0.002938485937193036,
      "learning_rate": 0.2979663224008292,
      "loss": 0.0599,
      "num_input_tokens_seen": 1193864,
      "step": 2100
    },
    {
      "epoch": 36.93805309734513,
      "grad_norm": 0.01930067129433155,
      "learning_rate": 0.2979566441519868,
      "loss": 0.0655,
      "num_input_tokens_seen": 1197096,
      "step": 2105
    },
    {
      "epoch": 37.017699115044245,
      "grad_norm": 0.0013358413707464933,
      "learning_rate": 0.29794694308637054,
      "loss": 0.0734,
      "num_input_tokens_seen": 1199704,
      "step": 2110
    },
    {
      "epoch": 37.10619469026549,
      "grad_norm": 0.003562435507774353,
      "learning_rate": 0.2979372192054764,
      "loss": 0.0579,
      "num_input_tokens_seen": 1203080,
      "step": 2115
    },
    {
      "epoch": 37.19469026548673,
      "grad_norm": 0.03605637699365616,
      "learning_rate": 0.297927472510804,
      "loss": 0.0883,
      "num_input_tokens_seen": 1205912,
      "step": 2120
    },
    {
      "epoch": 37.283185840707965,
      "grad_norm": 0.040509190410375595,
      "learning_rate": 0.29791770300385634,
      "loss": 0.1017,
      "num_input_tokens_seen": 1208792,
      "step": 2125
    },
    {
      "epoch": 37.3716814159292,
      "grad_norm": 0.00545346038416028,
      "learning_rate": 0.29790791068614003,
      "loss": 0.0397,
      "num_input_tokens_seen": 1211752,
      "step": 2130
    },
    {
      "epoch": 37.46017699115044,
      "grad_norm": 0.01528411265462637,
      "learning_rate": 0.2978980955591652,
      "loss": 0.017,
      "num_input_tokens_seen": 1214312,
      "step": 2135
    },
    {
      "epoch": 37.54867256637168,
      "grad_norm": 0.034804169088602066,
      "learning_rate": 0.2978882576244454,
      "loss": 0.0881,
      "num_input_tokens_seen": 1216968,
      "step": 2140
    },
    {
      "epoch": 37.63716814159292,
      "grad_norm": 0.0058987257070839405,
      "learning_rate": 0.2978783968834978,
      "loss": 0.0278,
      "num_input_tokens_seen": 1220744,
      "step": 2145
    },
    {
      "epoch": 37.72566371681416,
      "grad_norm": 0.002121626166626811,
      "learning_rate": 0.29786851333784303,
      "loss": 0.0367,
      "num_input_tokens_seen": 1223352,
      "step": 2150
    },
    {
      "epoch": 37.8141592920354,
      "grad_norm": 0.04717940837144852,
      "learning_rate": 0.2978586069890053,
      "loss": 0.0835,
      "num_input_tokens_seen": 1226424,
      "step": 2155
    },
    {
      "epoch": 37.902654867256636,
      "grad_norm": 0.009007456712424755,
      "learning_rate": 0.29784867783851227,
      "loss": 0.0282,
      "num_input_tokens_seen": 1228920,
      "step": 2160
    },
    {
      "epoch": 37.991150442477874,
      "grad_norm": 0.020844202488660812,
      "learning_rate": 0.2978387258878951,
      "loss": 0.1106,
      "num_input_tokens_seen": 1231672,
      "step": 2165
    },
    {
      "epoch": 38.07079646017699,
      "grad_norm": 0.002085199113935232,
      "learning_rate": 0.29782875113868856,
      "loss": 0.0292,
      "num_input_tokens_seen": 1234104,
      "step": 2170
    },
    {
      "epoch": 38.15929203539823,
      "grad_norm": 0.0055747199803590775,
      "learning_rate": 0.2978187535924309,
      "loss": 0.1164,
      "num_input_tokens_seen": 1237080,
      "step": 2175
    },
    {
      "epoch": 38.24778761061947,
      "grad_norm": 0.006691089831292629,
      "learning_rate": 0.29780873325066376,
      "loss": 0.03,
      "num_input_tokens_seen": 1240248,
      "step": 2180
    },
    {
      "epoch": 38.336283185840706,
      "grad_norm": 0.007961523719131947,
      "learning_rate": 0.2977986901149325,
      "loss": 0.128,
      "num_input_tokens_seen": 1242936,
      "step": 2185
    },
    {
      "epoch": 38.424778761061944,
      "grad_norm": 0.01194468792527914,
      "learning_rate": 0.29778862418678587,
      "loss": 0.0533,
      "num_input_tokens_seen": 1245752,
      "step": 2190
    },
    {
      "epoch": 38.51327433628319,
      "grad_norm": 0.0014250905951485038,
      "learning_rate": 0.29777853546777616,
      "loss": 0.0486,
      "num_input_tokens_seen": 1248152,
      "step": 2195
    },
    {
      "epoch": 38.60176991150443,
      "grad_norm": 0.010688142850995064,
      "learning_rate": 0.2977684239594592,
      "loss": 0.0221,
      "num_input_tokens_seen": 1251592,
      "step": 2200
    },
    {
      "epoch": 38.60176991150443,
      "eval_loss": 0.2030499279499054,
      "eval_runtime": 0.9316,
      "eval_samples_per_second": 26.836,
      "eval_steps_per_second": 13.955,
      "num_input_tokens_seen": 1251592,
      "step": 2200
    },
    {
      "epoch": 38.690265486725664,
      "grad_norm": 0.03169664740562439,
      "learning_rate": 0.29775828966339424,
      "loss": 0.0425,
      "num_input_tokens_seen": 1254312,
      "step": 2205
    },
    {
      "epoch": 38.7787610619469,
      "grad_norm": 0.02540615387260914,
      "learning_rate": 0.29774813258114424,
      "loss": 0.0525,
      "num_input_tokens_seen": 1257096,
      "step": 2210
    },
    {
      "epoch": 38.86725663716814,
      "grad_norm": 0.0032327938824892044,
      "learning_rate": 0.29773795271427544,
      "loss": 0.0378,
      "num_input_tokens_seen": 1260184,
      "step": 2215
    },
    {
      "epoch": 38.95575221238938,
      "grad_norm": 0.00040652762982062995,
      "learning_rate": 0.2977277500643577,
      "loss": 0.0125,
      "num_input_tokens_seen": 1262888,
      "step": 2220
    },
    {
      "epoch": 39.0353982300885,
      "grad_norm": 0.00010131843009730801,
      "learning_rate": 0.29771752463296447,
      "loss": 0.0529,
      "num_input_tokens_seen": 1265464,
      "step": 2225
    },
    {
      "epoch": 39.123893805309734,
      "grad_norm": 0.02381172403693199,
      "learning_rate": 0.29770727642167266,
      "loss": 0.0156,
      "num_input_tokens_seen": 1268088,
      "step": 2230
    },
    {
      "epoch": 39.21238938053097,
      "grad_norm": 0.010098197497427464,
      "learning_rate": 0.29769700543206257,
      "loss": 0.0051,
      "num_input_tokens_seen": 1271352,
      "step": 2235
    },
    {
      "epoch": 39.30088495575221,
      "grad_norm": 0.0014061558758839965,
      "learning_rate": 0.2976867116657182,
      "loss": 0.0086,
      "num_input_tokens_seen": 1273736,
      "step": 2240
    },
    {
      "epoch": 39.389380530973455,
      "grad_norm": 0.00041998736560344696,
      "learning_rate": 0.2976763951242269,
      "loss": 0.0293,
      "num_input_tokens_seen": 1276600,
      "step": 2245
    },
    {
      "epoch": 39.47787610619469,
      "grad_norm": 0.0028859747108072042,
      "learning_rate": 0.29766605580917965,
      "loss": 0.002,
      "num_input_tokens_seen": 1279800,
      "step": 2250
    },
    {
      "epoch": 39.56637168141593,
      "grad_norm": 0.005030928645282984,
      "learning_rate": 0.29765569372217093,
      "loss": 0.0078,
      "num_input_tokens_seen": 1283144,
      "step": 2255
    },
    {
      "epoch": 39.65486725663717,
      "grad_norm": 0.02045232057571411,
      "learning_rate": 0.2976453088647987,
      "loss": 0.0579,
      "num_input_tokens_seen": 1285816,
      "step": 2260
    },
    {
      "epoch": 39.743362831858406,
      "grad_norm": 0.00452482420951128,
      "learning_rate": 0.2976349012386644,
      "loss": 0.0073,
      "num_input_tokens_seen": 1288856,
      "step": 2265
    },
    {
      "epoch": 39.83185840707964,
      "grad_norm": 0.040368787944316864,
      "learning_rate": 0.29762447084537297,
      "loss": 0.0859,
      "num_input_tokens_seen": 1291544,
      "step": 2270
    },
    {
      "epoch": 39.92035398230089,
      "grad_norm": 0.042311638593673706,
      "learning_rate": 0.29761401768653306,
      "loss": 0.0589,
      "num_input_tokens_seen": 1294296,
      "step": 2275
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.0013129874132573605,
      "learning_rate": 0.29760354176375653,
      "loss": 0.0073,
      "num_input_tokens_seen": 1296696,
      "step": 2280
    },
    {
      "epoch": 40.08849557522124,
      "grad_norm": 0.04044976457953453,
      "learning_rate": 0.29759304307865897,
      "loss": 0.0196,
      "num_input_tokens_seen": 1299640,
      "step": 2285
    },
    {
      "epoch": 40.176991150442475,
      "grad_norm": 0.0229876097291708,
      "learning_rate": 0.2975825216328594,
      "loss": 0.0287,
      "num_input_tokens_seen": 1302520,
      "step": 2290
    },
    {
      "epoch": 40.26548672566372,
      "grad_norm": 0.1036461815237999,
      "learning_rate": 0.2975719774279804,
      "loss": 0.0405,
      "num_input_tokens_seen": 1305784,
      "step": 2295
    },
    {
      "epoch": 40.35398230088496,
      "grad_norm": 0.0316178984940052,
      "learning_rate": 0.29756141046564794,
      "loss": 0.018,
      "num_input_tokens_seen": 1309128,
      "step": 2300
    },
    {
      "epoch": 40.442477876106196,
      "grad_norm": 0.0034250549506396055,
      "learning_rate": 0.2975508207474916,
      "loss": 0.0103,
      "num_input_tokens_seen": 1311592,
      "step": 2305
    },
    {
      "epoch": 40.530973451327434,
      "grad_norm": 0.00023040662927087396,
      "learning_rate": 0.2975402082751445,
      "loss": 0.0054,
      "num_input_tokens_seen": 1314104,
      "step": 2310
    },
    {
      "epoch": 40.61946902654867,
      "grad_norm": 0.0020521068945527077,
      "learning_rate": 0.29752957305024313,
      "loss": 0.0113,
      "num_input_tokens_seen": 1316776,
      "step": 2315
    },
    {
      "epoch": 40.70796460176991,
      "grad_norm": 0.0021144552156329155,
      "learning_rate": 0.2975189150744277,
      "loss": 0.03,
      "num_input_tokens_seen": 1319400,
      "step": 2320
    },
    {
      "epoch": 40.796460176991154,
      "grad_norm": 0.0022206571884453297,
      "learning_rate": 0.29750823434934165,
      "loss": 0.0068,
      "num_input_tokens_seen": 1322408,
      "step": 2325
    },
    {
      "epoch": 40.88495575221239,
      "grad_norm": 0.002589950803667307,
      "learning_rate": 0.29749753087663217,
      "loss": 0.0053,
      "num_input_tokens_seen": 1325256,
      "step": 2330
    },
    {
      "epoch": 40.97345132743363,
      "grad_norm": 0.03549109399318695,
      "learning_rate": 0.29748680465794985,
      "loss": 0.0352,
      "num_input_tokens_seen": 1327976,
      "step": 2335
    },
    {
      "epoch": 41.05309734513274,
      "grad_norm": 0.005971208680421114,
      "learning_rate": 0.29747605569494884,
      "loss": 0.0062,
      "num_input_tokens_seen": 1330648,
      "step": 2340
    },
    {
      "epoch": 41.14159292035398,
      "grad_norm": 0.00230122241191566,
      "learning_rate": 0.29746528398928673,
      "loss": 0.0111,
      "num_input_tokens_seen": 1334088,
      "step": 2345
    },
    {
      "epoch": 41.230088495575224,
      "grad_norm": 0.0012742439284920692,
      "learning_rate": 0.2974544895426247,
      "loss": 0.0008,
      "num_input_tokens_seen": 1337000,
      "step": 2350
    },
    {
      "epoch": 41.31858407079646,
      "grad_norm": 0.0003652924788184464,
      "learning_rate": 0.29744367235662733,
      "loss": 0.0008,
      "num_input_tokens_seen": 1339976,
      "step": 2355
    },
    {
      "epoch": 41.4070796460177,
      "grad_norm": 0.00021368266607169062,
      "learning_rate": 0.29743283243296276,
      "loss": 0.0046,
      "num_input_tokens_seen": 1342792,
      "step": 2360
    },
    {
      "epoch": 41.49557522123894,
      "grad_norm": 0.0009084056364372373,
      "learning_rate": 0.29742196977330276,
      "loss": 0.0011,
      "num_input_tokens_seen": 1345480,
      "step": 2365
    },
    {
      "epoch": 41.584070796460175,
      "grad_norm": 0.0026350938715040684,
      "learning_rate": 0.2974110843793223,
      "loss": 0.006,
      "num_input_tokens_seen": 1348472,
      "step": 2370
    },
    {
      "epoch": 41.67256637168141,
      "grad_norm": 0.0006701347301714122,
      "learning_rate": 0.2974001762527002,
      "loss": 0.0008,
      "num_input_tokens_seen": 1351432,
      "step": 2375
    },
    {
      "epoch": 41.76106194690266,
      "grad_norm": 0.0020258172880858183,
      "learning_rate": 0.2973892453951186,
      "loss": 0.0083,
      "num_input_tokens_seen": 1354168,
      "step": 2380
    },
    {
      "epoch": 41.849557522123895,
      "grad_norm": 0.050072696059942245,
      "learning_rate": 0.2973782918082631,
      "loss": 0.0236,
      "num_input_tokens_seen": 1356952,
      "step": 2385
    },
    {
      "epoch": 41.93805309734513,
      "grad_norm": 0.00030286749824881554,
      "learning_rate": 0.29736731549382295,
      "loss": 0.0004,
      "num_input_tokens_seen": 1359304,
      "step": 2390
    },
    {
      "epoch": 42.017699115044245,
      "grad_norm": 0.00046257005305960774,
      "learning_rate": 0.2973563164534908,
      "loss": 0.0004,
      "num_input_tokens_seen": 1361752,
      "step": 2395
    },
    {
      "epoch": 42.10619469026549,
      "grad_norm": 0.00022960211208555847,
      "learning_rate": 0.29734529468896287,
      "loss": 0.0254,
      "num_input_tokens_seen": 1364312,
      "step": 2400
    },
    {
      "epoch": 42.10619469026549,
      "eval_loss": 0.32499754428863525,
      "eval_runtime": 1.3048,
      "eval_samples_per_second": 19.16,
      "eval_steps_per_second": 9.963,
      "num_input_tokens_seen": 1364312,
      "step": 2400
    },
    {
      "epoch": 42.19469026548673,
      "grad_norm": 0.000963477767072618,
      "learning_rate": 0.2973342502019388,
      "loss": 0.0009,
      "num_input_tokens_seen": 1367144,
      "step": 2405
    },
    {
      "epoch": 42.283185840707965,
      "grad_norm": 0.0012080012820661068,
      "learning_rate": 0.2973231829941219,
      "loss": 0.0006,
      "num_input_tokens_seen": 1369864,
      "step": 2410
    },
    {
      "epoch": 42.3716814159292,
      "grad_norm": 0.0013240133412182331,
      "learning_rate": 0.2973120930672188,
      "loss": 0.0027,
      "num_input_tokens_seen": 1373304,
      "step": 2415
    },
    {
      "epoch": 42.46017699115044,
      "grad_norm": 0.0076743527315557,
      "learning_rate": 0.2973009804229397,
      "loss": 0.0037,
      "num_input_tokens_seen": 1376168,
      "step": 2420
    },
    {
      "epoch": 42.54867256637168,
      "grad_norm": 0.0006759791867807508,
      "learning_rate": 0.29728984506299827,
      "loss": 0.0044,
      "num_input_tokens_seen": 1379080,
      "step": 2425
    },
    {
      "epoch": 42.63716814159292,
      "grad_norm": 0.005893242545425892,
      "learning_rate": 0.2972786869891118,
      "loss": 0.0034,
      "num_input_tokens_seen": 1382088,
      "step": 2430
    },
    {
      "epoch": 42.72566371681416,
      "grad_norm": 0.001658368855714798,
      "learning_rate": 0.29726750620300096,
      "loss": 0.0007,
      "num_input_tokens_seen": 1384600,
      "step": 2435
    },
    {
      "epoch": 42.8141592920354,
      "grad_norm": 0.009218432009220123,
      "learning_rate": 0.29725630270639003,
      "loss": 0.003,
      "num_input_tokens_seen": 1387128,
      "step": 2440
    },
    {
      "epoch": 42.902654867256636,
      "grad_norm": 0.024031465873122215,
      "learning_rate": 0.2972450765010067,
      "loss": 0.0053,
      "num_input_tokens_seen": 1389896,
      "step": 2445
    },
    {
      "epoch": 42.991150442477874,
      "grad_norm": 0.00016472725837957114,
      "learning_rate": 0.29723382758858213,
      "loss": 0.0039,
      "num_input_tokens_seen": 1393192,
      "step": 2450
    },
    {
      "epoch": 43.07079646017699,
      "grad_norm": 0.0019065671367570758,
      "learning_rate": 0.29722255597085107,
      "loss": 0.0033,
      "num_input_tokens_seen": 1395600,
      "step": 2455
    },
    {
      "epoch": 43.15929203539823,
      "grad_norm": 0.0070867761969566345,
      "learning_rate": 0.2972112616495518,
      "loss": 0.0059,
      "num_input_tokens_seen": 1398944,
      "step": 2460
    },
    {
      "epoch": 43.24778761061947,
      "grad_norm": 0.0006388761103153229,
      "learning_rate": 0.297199944626426,
      "loss": 0.0009,
      "num_input_tokens_seen": 1402160,
      "step": 2465
    },
    {
      "epoch": 43.336283185840706,
      "grad_norm": 1.3827519978804048e-05,
      "learning_rate": 0.2971886049032189,
      "loss": 0.0003,
      "num_input_tokens_seen": 1404784,
      "step": 2470
    },
    {
      "epoch": 43.424778761061944,
      "grad_norm": 7.785346679156646e-05,
      "learning_rate": 0.29717724248167926,
      "loss": 0.0001,
      "num_input_tokens_seen": 1407344,
      "step": 2475
    },
    {
      "epoch": 43.51327433628319,
      "grad_norm": 5.201085878070444e-05,
      "learning_rate": 0.29716585736355927,
      "loss": 0.0005,
      "num_input_tokens_seen": 1410656,
      "step": 2480
    },
    {
      "epoch": 43.60176991150443,
      "grad_norm": 0.007106730714440346,
      "learning_rate": 0.2971544495506147,
      "loss": 0.0017,
      "num_input_tokens_seen": 1413344,
      "step": 2485
    },
    {
      "epoch": 43.690265486725664,
      "grad_norm": 4.0393471863353625e-05,
      "learning_rate": 0.2971430190446048,
      "loss": 0.0007,
      "num_input_tokens_seen": 1415904,
      "step": 2490
    },
    {
      "epoch": 43.7787610619469,
      "grad_norm": 5.077809328213334e-05,
      "learning_rate": 0.2971315658472921,
      "loss": 0.0002,
      "num_input_tokens_seen": 1418384,
      "step": 2495
    },
    {
      "epoch": 43.86725663716814,
      "grad_norm": 0.0006421466241590679,
      "learning_rate": 0.2971200899604431,
      "loss": 0.0005,
      "num_input_tokens_seen": 1421328,
      "step": 2500
    },
    {
      "epoch": 43.95575221238938,
      "grad_norm": 0.0005047202575951815,
      "learning_rate": 0.29710859138582735,
      "loss": 0.0003,
      "num_input_tokens_seen": 1424304,
      "step": 2505
    },
    {
      "epoch": 44.0353982300885,
      "grad_norm": 0.00023062039690557867,
      "learning_rate": 0.29709707012521813,
      "loss": 0.0009,
      "num_input_tokens_seen": 1426608,
      "step": 2510
    },
    {
      "epoch": 44.123893805309734,
      "grad_norm": 3.449825817369856e-05,
      "learning_rate": 0.29708552618039213,
      "loss": 0.0002,
      "num_input_tokens_seen": 1429392,
      "step": 2515
    },
    {
      "epoch": 44.21238938053097,
      "grad_norm": 0.00013140607916284353,
      "learning_rate": 0.2970739595531296,
      "loss": 0.0002,
      "num_input_tokens_seen": 1432464,
      "step": 2520
    },
    {
      "epoch": 44.30088495575221,
      "grad_norm": 0.00016725629393476993,
      "learning_rate": 0.2970623702452143,
      "loss": 0.0002,
      "num_input_tokens_seen": 1435200,
      "step": 2525
    },
    {
      "epoch": 44.389380530973455,
      "grad_norm": 0.00014046099386177957,
      "learning_rate": 0.2970507582584334,
      "loss": 0.0001,
      "num_input_tokens_seen": 1438448,
      "step": 2530
    },
    {
      "epoch": 44.47787610619469,
      "grad_norm": 0.00039028681931085885,
      "learning_rate": 0.2970391235945776,
      "loss": 0.0002,
      "num_input_tokens_seen": 1440976,
      "step": 2535
    },
    {
      "epoch": 44.56637168141593,
      "grad_norm": 0.0005338029586710036,
      "learning_rate": 0.2970274662554412,
      "loss": 0.0002,
      "num_input_tokens_seen": 1444160,
      "step": 2540
    },
    {
      "epoch": 44.65486725663717,
      "grad_norm": 3.023855788342189e-05,
      "learning_rate": 0.2970157862428218,
      "loss": 0.0004,
      "num_input_tokens_seen": 1446944,
      "step": 2545
    },
    {
      "epoch": 44.743362831858406,
      "grad_norm": 0.00011754921433748677,
      "learning_rate": 0.2970040835585206,
      "loss": 0.0002,
      "num_input_tokens_seen": 1450336,
      "step": 2550
    },
    {
      "epoch": 44.83185840707964,
      "grad_norm": 0.0002539013512432575,
      "learning_rate": 0.2969923582043424,
      "loss": 0.0002,
      "num_input_tokens_seen": 1453328,
      "step": 2555
    },
    {
      "epoch": 44.92035398230089,
      "grad_norm": 2.270279401273001e-05,
      "learning_rate": 0.2969806101820953,
      "loss": 0.0001,
      "num_input_tokens_seen": 1456208,
      "step": 2560
    },
    {
      "epoch": 45.0,
      "grad_norm": 1.9186840290785767e-05,
      "learning_rate": 0.2969688394935911,
      "loss": 0.0003,
      "num_input_tokens_seen": 1458448,
      "step": 2565
    },
    {
      "epoch": 45.08849557522124,
      "grad_norm": 2.004735688387882e-05,
      "learning_rate": 0.2969570461406449,
      "loss": 0.0001,
      "num_input_tokens_seen": 1461616,
      "step": 2570
    },
    {
      "epoch": 45.176991150442475,
      "grad_norm": 0.00011274288408458233,
      "learning_rate": 0.29694523012507534,
      "loss": 0.0001,
      "num_input_tokens_seen": 1464352,
      "step": 2575
    },
    {
      "epoch": 45.26548672566372,
      "grad_norm": 0.00018203615036327392,
      "learning_rate": 0.2969333914487048,
      "loss": 0.0002,
      "num_input_tokens_seen": 1467232,
      "step": 2580
    },
    {
      "epoch": 45.35398230088496,
      "grad_norm": 0.00010410078539280221,
      "learning_rate": 0.2969215301133587,
      "loss": 0.0002,
      "num_input_tokens_seen": 1470144,
      "step": 2585
    },
    {
      "epoch": 45.442477876106196,
      "grad_norm": 3.477710561128333e-05,
      "learning_rate": 0.29690964612086634,
      "loss": 0.0001,
      "num_input_tokens_seen": 1473120,
      "step": 2590
    },
    {
      "epoch": 45.530973451327434,
      "grad_norm": 0.0003340657567605376,
      "learning_rate": 0.2968977394730604,
      "loss": 0.0001,
      "num_input_tokens_seen": 1475856,
      "step": 2595
    },
    {
      "epoch": 45.61946902654867,
      "grad_norm": 2.1355777789722197e-05,
      "learning_rate": 0.296885810171777,
      "loss": 0.0001,
      "num_input_tokens_seen": 1478704,
      "step": 2600
    },
    {
      "epoch": 45.61946902654867,
      "eval_loss": 0.3398699462413788,
      "eval_runtime": 0.9123,
      "eval_samples_per_second": 27.404,
      "eval_steps_per_second": 14.25,
      "num_input_tokens_seen": 1478704,
      "step": 2600
    },
    {
      "epoch": 45.70796460176991,
      "grad_norm": 3.7565023376373574e-06,
      "learning_rate": 0.2968738582188558,
      "loss": 0.0001,
      "num_input_tokens_seen": 1481120,
      "step": 2605
    },
    {
      "epoch": 45.796460176991154,
      "grad_norm": 8.239444287028164e-05,
      "learning_rate": 0.2968618836161399,
      "loss": 0.0001,
      "num_input_tokens_seen": 1484320,
      "step": 2610
    },
    {
      "epoch": 45.88495575221239,
      "grad_norm": 4.265425377525389e-05,
      "learning_rate": 0.296849886365476,
      "loss": 0.0001,
      "num_input_tokens_seen": 1487568,
      "step": 2615
    },
    {
      "epoch": 45.97345132743363,
      "grad_norm": 0.00030638196039944887,
      "learning_rate": 0.2968378664687142,
      "loss": 0.0002,
      "num_input_tokens_seen": 1490304,
      "step": 2620
    },
    {
      "epoch": 46.05309734513274,
      "grad_norm": 5.462783155962825e-05,
      "learning_rate": 0.296825823927708,
      "loss": 0.0001,
      "num_input_tokens_seen": 1492544,
      "step": 2625
    },
    {
      "epoch": 46.14159292035398,
      "grad_norm": 7.614289643242955e-05,
      "learning_rate": 0.29681375874431476,
      "loss": 0.0,
      "num_input_tokens_seen": 1495632,
      "step": 2630
    },
    {
      "epoch": 46.230088495575224,
      "grad_norm": 0.0001503842358943075,
      "learning_rate": 0.29680167092039483,
      "loss": 0.0002,
      "num_input_tokens_seen": 1498528,
      "step": 2635
    },
    {
      "epoch": 46.31858407079646,
      "grad_norm": 0.0001590990723343566,
      "learning_rate": 0.2967895604578125,
      "loss": 0.0001,
      "num_input_tokens_seen": 1501648,
      "step": 2640
    },
    {
      "epoch": 46.4070796460177,
      "grad_norm": 0.00012324508861638606,
      "learning_rate": 0.2967774273584352,
      "loss": 0.0001,
      "num_input_tokens_seen": 1504352,
      "step": 2645
    },
    {
      "epoch": 46.49557522123894,
      "grad_norm": 0.0002609076036605984,
      "learning_rate": 0.2967652716241342,
      "loss": 0.0001,
      "num_input_tokens_seen": 1507360,
      "step": 2650
    },
    {
      "epoch": 46.584070796460175,
      "grad_norm": 0.00014193008246365935,
      "learning_rate": 0.29675309325678384,
      "loss": 0.0001,
      "num_input_tokens_seen": 1510032,
      "step": 2655
    },
    {
      "epoch": 46.67256637168141,
      "grad_norm": 0.00013018070603720844,
      "learning_rate": 0.29674089225826233,
      "loss": 0.0001,
      "num_input_tokens_seen": 1513136,
      "step": 2660
    },
    {
      "epoch": 46.76106194690266,
      "grad_norm": 1.0219133400823921e-05,
      "learning_rate": 0.29672866863045116,
      "loss": 0.0001,
      "num_input_tokens_seen": 1515520,
      "step": 2665
    },
    {
      "epoch": 46.849557522123895,
      "grad_norm": 0.0002818087232299149,
      "learning_rate": 0.2967164223752354,
      "loss": 0.0002,
      "num_input_tokens_seen": 1518544,
      "step": 2670
    },
    {
      "epoch": 46.93805309734513,
      "grad_norm": 1.9716044334927574e-05,
      "learning_rate": 0.2967041534945035,
      "loss": 0.0001,
      "num_input_tokens_seen": 1521280,
      "step": 2675
    },
    {
      "epoch": 47.017699115044245,
      "grad_norm": 0.00020203158783260733,
      "learning_rate": 0.2966918619901476,
      "loss": 0.0001,
      "num_input_tokens_seen": 1523616,
      "step": 2680
    },
    {
      "epoch": 47.10619469026549,
      "grad_norm": 5.987444819766097e-05,
      "learning_rate": 0.2966795478640631,
      "loss": 0.0001,
      "num_input_tokens_seen": 1526816,
      "step": 2685
    },
    {
      "epoch": 47.19469026548673,
      "grad_norm": 4.1342242184327915e-05,
      "learning_rate": 0.29666721111814903,
      "loss": 0.0001,
      "num_input_tokens_seen": 1529776,
      "step": 2690
    },
    {
      "epoch": 47.283185840707965,
      "grad_norm": 0.00016613007755950093,
      "learning_rate": 0.2966548517543079,
      "loss": 0.0001,
      "num_input_tokens_seen": 1532624,
      "step": 2695
    },
    {
      "epoch": 47.3716814159292,
      "grad_norm": 1.1030122550437227e-05,
      "learning_rate": 0.29664246977444564,
      "loss": 0.0001,
      "num_input_tokens_seen": 1535440,
      "step": 2700
    },
    {
      "epoch": 47.46017699115044,
      "grad_norm": 1.6608530131634325e-05,
      "learning_rate": 0.2966300651804717,
      "loss": 0.0001,
      "num_input_tokens_seen": 1538336,
      "step": 2705
    },
    {
      "epoch": 47.54867256637168,
      "grad_norm": 8.151827933033928e-05,
      "learning_rate": 0.296617637974299,
      "loss": 0.0001,
      "num_input_tokens_seen": 1541024,
      "step": 2710
    },
    {
      "epoch": 47.63716814159292,
      "grad_norm": 8.813589374767616e-05,
      "learning_rate": 0.2966051881578441,
      "loss": 0.0001,
      "num_input_tokens_seen": 1544160,
      "step": 2715
    },
    {
      "epoch": 47.72566371681416,
      "grad_norm": 0.00020420108921825886,
      "learning_rate": 0.29659271573302676,
      "loss": 0.0001,
      "num_input_tokens_seen": 1546880,
      "step": 2720
    },
    {
      "epoch": 47.8141592920354,
      "grad_norm": 7.070146239129826e-05,
      "learning_rate": 0.2965802207017705,
      "loss": 0.0001,
      "num_input_tokens_seen": 1549216,
      "step": 2725
    },
    {
      "epoch": 47.902654867256636,
      "grad_norm": 8.066048030741513e-05,
      "learning_rate": 0.2965677030660021,
      "loss": 0.0001,
      "num_input_tokens_seen": 1551648,
      "step": 2730
    },
    {
      "epoch": 47.991150442477874,
      "grad_norm": 1.5294517652364448e-05,
      "learning_rate": 0.2965551628276521,
      "loss": 0.0001,
      "num_input_tokens_seen": 1554752,
      "step": 2735
    },
    {
      "epoch": 48.07079646017699,
      "grad_norm": 5.138834694662364e-06,
      "learning_rate": 0.29654259998865423,
      "loss": 0.0,
      "num_input_tokens_seen": 1557288,
      "step": 2740
    },
    {
      "epoch": 48.15929203539823,
      "grad_norm": 0.0001015152593026869,
      "learning_rate": 0.2965300145509458,
      "loss": 0.0001,
      "num_input_tokens_seen": 1560408,
      "step": 2745
    },
    {
      "epoch": 48.24778761061947,
      "grad_norm": 2.979358941956889e-05,
      "learning_rate": 0.2965174065164678,
      "loss": 0.0001,
      "num_input_tokens_seen": 1563160,
      "step": 2750
    },
    {
      "epoch": 48.336283185840706,
      "grad_norm": 0.00016515686002094299,
      "learning_rate": 0.2965047758871644,
      "loss": 0.0001,
      "num_input_tokens_seen": 1565896,
      "step": 2755
    },
    {
      "epoch": 48.424778761061944,
      "grad_norm": 2.683965431060642e-05,
      "learning_rate": 0.2964921226649835,
      "loss": 0.0001,
      "num_input_tokens_seen": 1568776,
      "step": 2760
    },
    {
      "epoch": 48.51327433628319,
      "grad_norm": 3.990148616139777e-05,
      "learning_rate": 0.2964794468518763,
      "loss": 0.0001,
      "num_input_tokens_seen": 1571736,
      "step": 2765
    },
    {
      "epoch": 48.60176991150443,
      "grad_norm": 2.5928427930921316e-05,
      "learning_rate": 0.2964667484497977,
      "loss": 0.0001,
      "num_input_tokens_seen": 1575016,
      "step": 2770
    },
    {
      "epoch": 48.690265486725664,
      "grad_norm": 7.753532554488629e-05,
      "learning_rate": 0.29645402746070587,
      "loss": 0.0001,
      "num_input_tokens_seen": 1577448,
      "step": 2775
    },
    {
      "epoch": 48.7787610619469,
      "grad_norm": 0.00011889307643286884,
      "learning_rate": 0.2964412838865625,
      "loss": 0.0001,
      "num_input_tokens_seen": 1580056,
      "step": 2780
    },
    {
      "epoch": 48.86725663716814,
      "grad_norm": 4.798822192242369e-05,
      "learning_rate": 0.29642851772933293,
      "loss": 0.0001,
      "num_input_tokens_seen": 1583096,
      "step": 2785
    },
    {
      "epoch": 48.95575221238938,
      "grad_norm": 3.1255076464731246e-05,
      "learning_rate": 0.29641572899098567,
      "loss": 0.0001,
      "num_input_tokens_seen": 1585864,
      "step": 2790
    },
    {
      "epoch": 49.0353982300885,
      "grad_norm": 3.93828668165952e-05,
      "learning_rate": 0.29640291767349314,
      "loss": 0.0001,
      "num_input_tokens_seen": 1588512,
      "step": 2795
    },
    {
      "epoch": 49.123893805309734,
      "grad_norm": 9.597477037459612e-05,
      "learning_rate": 0.2963900837788308,
      "loss": 0.0001,
      "num_input_tokens_seen": 1591424,
      "step": 2800
    },
    {
      "epoch": 49.123893805309734,
      "eval_loss": 0.34531939029693604,
      "eval_runtime": 0.927,
      "eval_samples_per_second": 26.968,
      "eval_steps_per_second": 14.023,
      "num_input_tokens_seen": 1591424,
      "step": 2800
    },
    {
      "epoch": 49.21238938053097,
      "grad_norm": 1.5117789189389441e-05,
      "learning_rate": 0.2963772273089779,
      "loss": 0.0001,
      "num_input_tokens_seen": 1593840,
      "step": 2805
    },
    {
      "epoch": 49.30088495575221,
      "grad_norm": 5.635102934320457e-05,
      "learning_rate": 0.2963643482659171,
      "loss": 0.0001,
      "num_input_tokens_seen": 1597232,
      "step": 2810
    },
    {
      "epoch": 49.389380530973455,
      "grad_norm": 0.0001008128238026984,
      "learning_rate": 0.2963514466516345,
      "loss": 0.0001,
      "num_input_tokens_seen": 1600144,
      "step": 2815
    },
    {
      "epoch": 49.47787610619469,
      "grad_norm": 5.571169458562508e-05,
      "learning_rate": 0.2963385224681196,
      "loss": 0.0,
      "num_input_tokens_seen": 1602800,
      "step": 2820
    },
    {
      "epoch": 49.56637168141593,
      "grad_norm": 1.1481000001367647e-05,
      "learning_rate": 0.29632557571736556,
      "loss": 0.0001,
      "num_input_tokens_seen": 1605872,
      "step": 2825
    },
    {
      "epoch": 49.65486725663717,
      "grad_norm": 5.6025917729130015e-05,
      "learning_rate": 0.2963126064013689,
      "loss": 0.0,
      "num_input_tokens_seen": 1608768,
      "step": 2830
    },
    {
      "epoch": 49.743362831858406,
      "grad_norm": 9.012265945784748e-05,
      "learning_rate": 0.29629961452212966,
      "loss": 0.0001,
      "num_input_tokens_seen": 1611488,
      "step": 2835
    },
    {
      "epoch": 49.83185840707964,
      "grad_norm": 6.681102968286723e-05,
      "learning_rate": 0.2962866000816513,
      "loss": 0.0001,
      "num_input_tokens_seen": 1614448,
      "step": 2840
    },
    {
      "epoch": 49.92035398230089,
      "grad_norm": 4.147098297835328e-05,
      "learning_rate": 0.2962735630819409,
      "loss": 0.0,
      "num_input_tokens_seen": 1616784,
      "step": 2845
    },
    {
      "epoch": 50.0,
      "grad_norm": 2.157991502826917e-06,
      "learning_rate": 0.2962605035250089,
      "loss": 0.0001,
      "num_input_tokens_seen": 1619360,
      "step": 2850
    },
    {
      "epoch": 50.08849557522124,
      "grad_norm": 3.0686205718666315e-05,
      "learning_rate": 0.29624742141286914,
      "loss": 0.0001,
      "num_input_tokens_seen": 1622448,
      "step": 2855
    },
    {
      "epoch": 50.176991150442475,
      "grad_norm": 0.0002142949670087546,
      "learning_rate": 0.29623431674753925,
      "loss": 0.0001,
      "num_input_tokens_seen": 1625536,
      "step": 2860
    },
    {
      "epoch": 50.26548672566372,
      "grad_norm": 3.992580332123907e-06,
      "learning_rate": 0.29622118953103993,
      "loss": 0.0001,
      "num_input_tokens_seen": 1628176,
      "step": 2865
    },
    {
      "epoch": 50.35398230088496,
      "grad_norm": 1.9366199921933003e-05,
      "learning_rate": 0.2962080397653957,
      "loss": 0.0001,
      "num_input_tokens_seen": 1630640,
      "step": 2870
    },
    {
      "epoch": 50.442477876106196,
      "grad_norm": 1.979381158889737e-05,
      "learning_rate": 0.29619486745263435,
      "loss": 0.0,
      "num_input_tokens_seen": 1633072,
      "step": 2875
    },
    {
      "epoch": 50.530973451327434,
      "grad_norm": 2.276863051520195e-05,
      "learning_rate": 0.2961816725947873,
      "loss": 0.0,
      "num_input_tokens_seen": 1635664,
      "step": 2880
    },
    {
      "epoch": 50.61946902654867,
      "grad_norm": 2.0678015062003396e-05,
      "learning_rate": 0.29616845519388924,
      "loss": 0.0001,
      "num_input_tokens_seen": 1638816,
      "step": 2885
    },
    {
      "epoch": 50.70796460176991,
      "grad_norm": 8.507293387083337e-05,
      "learning_rate": 0.2961552152519785,
      "loss": 0.0001,
      "num_input_tokens_seen": 1641968,
      "step": 2890
    },
    {
      "epoch": 50.796460176991154,
      "grad_norm": 9.362563105241861e-06,
      "learning_rate": 0.29614195277109695,
      "loss": 0.0,
      "num_input_tokens_seen": 1644960,
      "step": 2895
    },
    {
      "epoch": 50.88495575221239,
      "grad_norm": 0.00014587417535949498,
      "learning_rate": 0.2961286677532897,
      "loss": 0.0001,
      "num_input_tokens_seen": 1648272,
      "step": 2900
    },
    {
      "epoch": 50.97345132743363,
      "grad_norm": 1.0639868378348183e-05,
      "learning_rate": 0.2961153602006055,
      "loss": 0.0,
      "num_input_tokens_seen": 1651152,
      "step": 2905
    },
    {
      "epoch": 51.05309734513274,
      "grad_norm": 8.087149035418406e-05,
      "learning_rate": 0.29610203011509656,
      "loss": 0.0001,
      "num_input_tokens_seen": 1653504,
      "step": 2910
    },
    {
      "epoch": 51.14159292035398,
      "grad_norm": 2.990651955769863e-05,
      "learning_rate": 0.29608867749881856,
      "loss": 0.0001,
      "num_input_tokens_seen": 1656624,
      "step": 2915
    },
    {
      "epoch": 51.230088495575224,
      "grad_norm": 5.3029732953291386e-05,
      "learning_rate": 0.29607530235383067,
      "loss": 0.0,
      "num_input_tokens_seen": 1659424,
      "step": 2920
    },
    {
      "epoch": 51.31858407079646,
      "grad_norm": 5.0947433919645846e-05,
      "learning_rate": 0.2960619046821954,
      "loss": 0.0001,
      "num_input_tokens_seen": 1662528,
      "step": 2925
    },
    {
      "epoch": 51.4070796460177,
      "grad_norm": 7.634783105459064e-05,
      "learning_rate": 0.2960484844859789,
      "loss": 0.0,
      "num_input_tokens_seen": 1665632,
      "step": 2930
    },
    {
      "epoch": 51.49557522123894,
      "grad_norm": 0.00011019138764822856,
      "learning_rate": 0.29603504176725076,
      "loss": 0.0001,
      "num_input_tokens_seen": 1668544,
      "step": 2935
    },
    {
      "epoch": 51.584070796460175,
      "grad_norm": 5.928683094680309e-05,
      "learning_rate": 0.296021576528084,
      "loss": 0.0001,
      "num_input_tokens_seen": 1671184,
      "step": 2940
    },
    {
      "epoch": 51.67256637168141,
      "grad_norm": 9.8450500445324e-06,
      "learning_rate": 0.29600808877055507,
      "loss": 0.0001,
      "num_input_tokens_seen": 1674320,
      "step": 2945
    },
    {
      "epoch": 51.76106194690266,
      "grad_norm": 2.676615258678794e-05,
      "learning_rate": 0.29599457849674404,
      "loss": 0.0,
      "num_input_tokens_seen": 1677216,
      "step": 2950
    },
    {
      "epoch": 51.849557522123895,
      "grad_norm": 2.0211069568176754e-05,
      "learning_rate": 0.2959810457087343,
      "loss": 0.0,
      "num_input_tokens_seen": 1679968,
      "step": 2955
    },
    {
      "epoch": 51.93805309734513,
      "grad_norm": 0.00011889731831615791,
      "learning_rate": 0.2959674904086128,
      "loss": 0.0001,
      "num_input_tokens_seen": 1682480,
      "step": 2960
    },
    {
      "epoch": 52.017699115044245,
      "grad_norm": 4.294277459848672e-05,
      "learning_rate": 0.2959539125984699,
      "loss": 0.0,
      "num_input_tokens_seen": 1685352,
      "step": 2965
    },
    {
      "epoch": 52.10619469026549,
      "grad_norm": 4.894227095064707e-05,
      "learning_rate": 0.2959403122803996,
      "loss": 0.0001,
      "num_input_tokens_seen": 1687784,
      "step": 2970
    },
    {
      "epoch": 52.19469026548673,
      "grad_norm": 5.468092058436014e-05,
      "learning_rate": 0.2959266894564991,
      "loss": 0.0001,
      "num_input_tokens_seen": 1691416,
      "step": 2975
    },
    {
      "epoch": 52.283185840707965,
      "grad_norm": 0.00013957191549707204,
      "learning_rate": 0.2959130441288692,
      "loss": 0.0001,
      "num_input_tokens_seen": 1694344,
      "step": 2980
    },
    {
      "epoch": 52.3716814159292,
      "grad_norm": 6.133555143605918e-05,
      "learning_rate": 0.2958993762996143,
      "loss": 0.0001,
      "num_input_tokens_seen": 1696840,
      "step": 2985
    },
    {
      "epoch": 52.46017699115044,
      "grad_norm": 4.692682341556065e-05,
      "learning_rate": 0.2958856859708421,
      "loss": 0.0,
      "num_input_tokens_seen": 1699304,
      "step": 2990
    },
    {
      "epoch": 52.54867256637168,
      "grad_norm": 5.648274964187294e-06,
      "learning_rate": 0.2958719731446638,
      "loss": 0.0,
      "num_input_tokens_seen": 1701880,
      "step": 2995
    },
    {
      "epoch": 52.63716814159292,
      "grad_norm": 2.9879267458454706e-05,
      "learning_rate": 0.29585823782319404,
      "loss": 0.0,
      "num_input_tokens_seen": 1705000,
      "step": 3000
    },
    {
      "epoch": 52.63716814159292,
      "eval_loss": 0.3432482182979584,
      "eval_runtime": 0.9289,
      "eval_samples_per_second": 26.912,
      "eval_steps_per_second": 13.994,
      "num_input_tokens_seen": 1705000,
      "step": 3000
    },
    {
      "epoch": 52.72566371681416,
      "grad_norm": 2.4884155209292658e-05,
      "learning_rate": 0.2958444800085511,
      "loss": 0.0,
      "num_input_tokens_seen": 1707896,
      "step": 3005
    },
    {
      "epoch": 52.8141592920354,
      "grad_norm": 1.7843229215941392e-05,
      "learning_rate": 0.2958306997028565,
      "loss": 0.0,
      "num_input_tokens_seen": 1710840,
      "step": 3010
    },
    {
      "epoch": 52.902654867256636,
      "grad_norm": 0.00010329360520699993,
      "learning_rate": 0.2958168969082354,
      "loss": 0.0,
      "num_input_tokens_seen": 1713480,
      "step": 3015
    },
    {
      "epoch": 52.991150442477874,
      "grad_norm": 5.095766027807258e-05,
      "learning_rate": 0.2958030716268164,
      "loss": 0.0001,
      "num_input_tokens_seen": 1716696,
      "step": 3020
    },
    {
      "epoch": 53.07079646017699,
      "grad_norm": 3.9759801438776776e-05,
      "learning_rate": 0.2957892238607314,
      "loss": 0.0,
      "num_input_tokens_seen": 1719344,
      "step": 3025
    },
    {
      "epoch": 53.15929203539823,
      "grad_norm": 9.61593832471408e-05,
      "learning_rate": 0.2957753536121161,
      "loss": 0.0,
      "num_input_tokens_seen": 1722352,
      "step": 3030
    },
    {
      "epoch": 53.24778761061947,
      "grad_norm": 5.18625856784638e-05,
      "learning_rate": 0.29576146088310923,
      "loss": 0.0,
      "num_input_tokens_seen": 1725312,
      "step": 3035
    },
    {
      "epoch": 53.336283185840706,
      "grad_norm": 0.00011492300109239295,
      "learning_rate": 0.2957475456758533,
      "loss": 0.0001,
      "num_input_tokens_seen": 1728208,
      "step": 3040
    },
    {
      "epoch": 53.424778761061944,
      "grad_norm": 7.969270518515259e-05,
      "learning_rate": 0.2957336079924944,
      "loss": 0.0,
      "num_input_tokens_seen": 1731056,
      "step": 3045
    },
    {
      "epoch": 53.51327433628319,
      "grad_norm": 2.0095483705517836e-05,
      "learning_rate": 0.2957196478351816,
      "loss": 0.0,
      "num_input_tokens_seen": 1734304,
      "step": 3050
    },
    {
      "epoch": 53.60176991150443,
      "grad_norm": 5.5109999266278464e-06,
      "learning_rate": 0.295705665206068,
      "loss": 0.0,
      "num_input_tokens_seen": 1737168,
      "step": 3055
    },
    {
      "epoch": 53.690265486725664,
      "grad_norm": 2.2567943233298138e-05,
      "learning_rate": 0.2956916601073097,
      "loss": 0.0,
      "num_input_tokens_seen": 1740048,
      "step": 3060
    },
    {
      "epoch": 53.7787610619469,
      "grad_norm": 3.392180587979965e-05,
      "learning_rate": 0.29567763254106655,
      "loss": 0.0001,
      "num_input_tokens_seen": 1742960,
      "step": 3065
    },
    {
      "epoch": 53.86725663716814,
      "grad_norm": 0.00010556568304309621,
      "learning_rate": 0.29566358250950175,
      "loss": 0.0001,
      "num_input_tokens_seen": 1745376,
      "step": 3070
    },
    {
      "epoch": 53.95575221238938,
      "grad_norm": 3.916039349860512e-05,
      "learning_rate": 0.295649510014782,
      "loss": 0.0001,
      "num_input_tokens_seen": 1748160,
      "step": 3075
    },
    {
      "epoch": 54.0353982300885,
      "grad_norm": 2.8387836209731176e-05,
      "learning_rate": 0.2956354150590775,
      "loss": 0.0,
      "num_input_tokens_seen": 1750480,
      "step": 3080
    },
    {
      "epoch": 54.123893805309734,
      "grad_norm": 3.4690929169300944e-05,
      "learning_rate": 0.2956212976445618,
      "loss": 0.0,
      "num_input_tokens_seen": 1753504,
      "step": 3085
    },
    {
      "epoch": 54.21238938053097,
      "grad_norm": 6.044669135008007e-06,
      "learning_rate": 0.295607157773412,
      "loss": 0.0001,
      "num_input_tokens_seen": 1756128,
      "step": 3090
    },
    {
      "epoch": 54.30088495575221,
      "grad_norm": 6.607595423702151e-06,
      "learning_rate": 0.2955929954478087,
      "loss": 0.0,
      "num_input_tokens_seen": 1758800,
      "step": 3095
    },
    {
      "epoch": 54.389380530973455,
      "grad_norm": 4.2982086597476155e-05,
      "learning_rate": 0.29557881066993585,
      "loss": 0.0,
      "num_input_tokens_seen": 1761888,
      "step": 3100
    },
    {
      "epoch": 54.47787610619469,
      "grad_norm": 3.6731697036884725e-05,
      "learning_rate": 0.29556460344198093,
      "loss": 0.0,
      "num_input_tokens_seen": 1765200,
      "step": 3105
    },
    {
      "epoch": 54.56637168141593,
      "grad_norm": 2.8716565793729387e-05,
      "learning_rate": 0.29555037376613486,
      "loss": 0.0001,
      "num_input_tokens_seen": 1767712,
      "step": 3110
    },
    {
      "epoch": 54.65486725663717,
      "grad_norm": 2.0178968043182977e-05,
      "learning_rate": 0.29553612164459203,
      "loss": 0.0,
      "num_input_tokens_seen": 1769920,
      "step": 3115
    },
    {
      "epoch": 54.743362831858406,
      "grad_norm": 1.1766589523176663e-05,
      "learning_rate": 0.29552184707955037,
      "loss": 0.0,
      "num_input_tokens_seen": 1773312,
      "step": 3120
    },
    {
      "epoch": 54.83185840707964,
      "grad_norm": 3.3964772683248157e-06,
      "learning_rate": 0.29550755007321117,
      "loss": 0.0001,
      "num_input_tokens_seen": 1776128,
      "step": 3125
    },
    {
      "epoch": 54.92035398230089,
      "grad_norm": 8.562380389776081e-05,
      "learning_rate": 0.29549323062777916,
      "loss": 0.0001,
      "num_input_tokens_seen": 1779088,
      "step": 3130
    },
    {
      "epoch": 55.0,
      "grad_norm": 2.1120120436535217e-05,
      "learning_rate": 0.29547888874546263,
      "loss": 0.0,
      "num_input_tokens_seen": 1781944,
      "step": 3135
    },
    {
      "epoch": 55.08849557522124,
      "grad_norm": 7.86690870882012e-05,
      "learning_rate": 0.2954645244284732,
      "loss": 0.0,
      "num_input_tokens_seen": 1784568,
      "step": 3140
    },
    {
      "epoch": 55.176991150442475,
      "grad_norm": 4.128727232455276e-05,
      "learning_rate": 0.2954501376790261,
      "loss": 0.0001,
      "num_input_tokens_seen": 1787560,
      "step": 3145
    },
    {
      "epoch": 55.26548672566372,
      "grad_norm": 3.9662732888245955e-05,
      "learning_rate": 0.29543572849933997,
      "loss": 0.0,
      "num_input_tokens_seen": 1790824,
      "step": 3150
    },
    {
      "epoch": 55.35398230088496,
      "grad_norm": 5.2851195505354553e-05,
      "learning_rate": 0.2954212968916368,
      "loss": 0.0,
      "num_input_tokens_seen": 1793176,
      "step": 3155
    },
    {
      "epoch": 55.442477876106196,
      "grad_norm": 1.7206702978000976e-05,
      "learning_rate": 0.29540684285814217,
      "loss": 0.0,
      "num_input_tokens_seen": 1796008,
      "step": 3160
    },
    {
      "epoch": 55.530973451327434,
      "grad_norm": 2.7548794605536386e-05,
      "learning_rate": 0.2953923664010851,
      "loss": 0.0001,
      "num_input_tokens_seen": 1799096,
      "step": 3165
    },
    {
      "epoch": 55.61946902654867,
      "grad_norm": 5.025412974646315e-05,
      "learning_rate": 0.295377867522698,
      "loss": 0.0,
      "num_input_tokens_seen": 1802104,
      "step": 3170
    },
    {
      "epoch": 55.70796460176991,
      "grad_norm": 3.946249307773542e-06,
      "learning_rate": 0.2953633462252168,
      "loss": 0.0,
      "num_input_tokens_seen": 1804808,
      "step": 3175
    },
    {
      "epoch": 55.796460176991154,
      "grad_norm": 5.955912001809338e-06,
      "learning_rate": 0.2953488025108809,
      "loss": 0.0,
      "num_input_tokens_seen": 1807704,
      "step": 3180
    },
    {
      "epoch": 55.88495575221239,
      "grad_norm": 5.4806539992569014e-05,
      "learning_rate": 0.295334236381933,
      "loss": 0.0,
      "num_input_tokens_seen": 1810440,
      "step": 3185
    },
    {
      "epoch": 55.97345132743363,
      "grad_norm": 4.975424235453829e-05,
      "learning_rate": 0.29531964784061954,
      "loss": 0.0001,
      "num_input_tokens_seen": 1813272,
      "step": 3190
    },
    {
      "epoch": 56.05309734513274,
      "grad_norm": 2.4640861738589592e-05,
      "learning_rate": 0.2953050368891902,
      "loss": 0.0,
      "num_input_tokens_seen": 1815616,
      "step": 3195
    },
    {
      "epoch": 56.14159292035398,
      "grad_norm": 7.316016854019836e-05,
      "learning_rate": 0.29529040352989805,
      "loss": 0.0,
      "num_input_tokens_seen": 1818688,
      "step": 3200
    },
    {
      "epoch": 56.14159292035398,
      "eval_loss": 0.367011696100235,
      "eval_runtime": 0.9212,
      "eval_samples_per_second": 27.138,
      "eval_steps_per_second": 14.112,
      "num_input_tokens_seen": 1818688,
      "step": 3200
    },
    {
      "epoch": 56.230088495575224,
      "grad_norm": 1.605027682671789e-05,
      "learning_rate": 0.29527574776499993,
      "loss": 0.0,
      "num_input_tokens_seen": 1821104,
      "step": 3205
    },
    {
      "epoch": 56.31858407079646,
      "grad_norm": 5.4767086112406105e-05,
      "learning_rate": 0.2952610695967558,
      "loss": 0.0,
      "num_input_tokens_seen": 1824144,
      "step": 3210
    },
    {
      "epoch": 56.4070796460177,
      "grad_norm": 8.414339390583336e-05,
      "learning_rate": 0.29524636902742935,
      "loss": 0.0,
      "num_input_tokens_seen": 1827536,
      "step": 3215
    },
    {
      "epoch": 56.49557522123894,
      "grad_norm": 8.184802572941408e-05,
      "learning_rate": 0.2952316460592875,
      "loss": 0.0001,
      "num_input_tokens_seen": 1830624,
      "step": 3220
    },
    {
      "epoch": 56.584070796460175,
      "grad_norm": 4.687665932578966e-05,
      "learning_rate": 0.29521690069460066,
      "loss": 0.0,
      "num_input_tokens_seen": 1832992,
      "step": 3225
    },
    {
      "epoch": 56.67256637168141,
      "grad_norm": 2.7223957658861764e-05,
      "learning_rate": 0.29520213293564285,
      "loss": 0.0001,
      "num_input_tokens_seen": 1835936,
      "step": 3230
    },
    {
      "epoch": 56.76106194690266,
      "grad_norm": 2.9260703740874305e-05,
      "learning_rate": 0.29518734278469144,
      "loss": 0.0,
      "num_input_tokens_seen": 1838768,
      "step": 3235
    },
    {
      "epoch": 56.849557522123895,
      "grad_norm": 1.3085154932923615e-05,
      "learning_rate": 0.29517253024402723,
      "loss": 0.0,
      "num_input_tokens_seen": 1841488,
      "step": 3240
    },
    {
      "epoch": 56.93805309734513,
      "grad_norm": 3.1526549719274044e-05,
      "learning_rate": 0.2951576953159345,
      "loss": 0.0,
      "num_input_tokens_seen": 1844688,
      "step": 3245
    },
    {
      "epoch": 57.017699115044245,
      "grad_norm": 2.665295687620528e-05,
      "learning_rate": 0.29514283800270097,
      "loss": 0.0,
      "num_input_tokens_seen": 1847216,
      "step": 3250
    },
    {
      "epoch": 57.10619469026549,
      "grad_norm": 5.847301508765668e-05,
      "learning_rate": 0.2951279583066179,
      "loss": 0.0001,
      "num_input_tokens_seen": 1849856,
      "step": 3255
    },
    {
      "epoch": 57.19469026548673,
      "grad_norm": 6.715930067002773e-05,
      "learning_rate": 0.2951130562299798,
      "loss": 0.0001,
      "num_input_tokens_seen": 1852576,
      "step": 3260
    },
    {
      "epoch": 57.283185840707965,
      "grad_norm": 3.9144859329098836e-05,
      "learning_rate": 0.29509813177508487,
      "loss": 0.0,
      "num_input_tokens_seen": 1855872,
      "step": 3265
    },
    {
      "epoch": 57.3716814159292,
      "grad_norm": 4.464506673684809e-06,
      "learning_rate": 0.2950831849442346,
      "loss": 0.0,
      "num_input_tokens_seen": 1858416,
      "step": 3270
    },
    {
      "epoch": 57.46017699115044,
      "grad_norm": 3.8103535189293325e-05,
      "learning_rate": 0.2950682157397339,
      "loss": 0.0,
      "num_input_tokens_seen": 1861024,
      "step": 3275
    },
    {
      "epoch": 57.54867256637168,
      "grad_norm": 2.776885412458796e-05,
      "learning_rate": 0.2950532241638914,
      "loss": 0.0,
      "num_input_tokens_seen": 1863616,
      "step": 3280
    },
    {
      "epoch": 57.63716814159292,
      "grad_norm": 2.5324638045276515e-05,
      "learning_rate": 0.2950382102190188,
      "loss": 0.0,
      "num_input_tokens_seen": 1866672,
      "step": 3285
    },
    {
      "epoch": 57.72566371681416,
      "grad_norm": 4.904646266368218e-05,
      "learning_rate": 0.2950231739074316,
      "loss": 0.0,
      "num_input_tokens_seen": 1869456,
      "step": 3290
    },
    {
      "epoch": 57.8141592920354,
      "grad_norm": 1.1954310139117297e-05,
      "learning_rate": 0.29500811523144843,
      "loss": 0.0,
      "num_input_tokens_seen": 1872032,
      "step": 3295
    },
    {
      "epoch": 57.902654867256636,
      "grad_norm": 2.068079811579082e-05,
      "learning_rate": 0.2949930341933917,
      "loss": 0.0,
      "num_input_tokens_seen": 1875104,
      "step": 3300
    },
    {
      "epoch": 57.991150442477874,
      "grad_norm": 1.9430868633207865e-05,
      "learning_rate": 0.29497793079558693,
      "loss": 0.0,
      "num_input_tokens_seen": 1878672,
      "step": 3305
    },
    {
      "epoch": 58.07079646017699,
      "grad_norm": 1.1421545423218049e-05,
      "learning_rate": 0.2949628050403633,
      "loss": 0.0,
      "num_input_tokens_seen": 1881032,
      "step": 3310
    },
    {
      "epoch": 58.15929203539823,
      "grad_norm": 5.9051759308204055e-05,
      "learning_rate": 0.2949476569300535,
      "loss": 0.0,
      "num_input_tokens_seen": 1883480,
      "step": 3315
    },
    {
      "epoch": 58.24778761061947,
      "grad_norm": 2.4441131245112047e-05,
      "learning_rate": 0.29493248646699344,
      "loss": 0.0,
      "num_input_tokens_seen": 1886648,
      "step": 3320
    },
    {
      "epoch": 58.336283185840706,
      "grad_norm": 4.138241638429463e-05,
      "learning_rate": 0.29491729365352265,
      "loss": 0.0,
      "num_input_tokens_seen": 1890216,
      "step": 3325
    },
    {
      "epoch": 58.424778761061944,
      "grad_norm": 1.017398426483851e-05,
      "learning_rate": 0.29490207849198397,
      "loss": 0.0,
      "num_input_tokens_seen": 1892888,
      "step": 3330
    },
    {
      "epoch": 58.51327433628319,
      "grad_norm": 4.3009556975448504e-05,
      "learning_rate": 0.29488684098472384,
      "loss": 0.0,
      "num_input_tokens_seen": 1895944,
      "step": 3335
    },
    {
      "epoch": 58.60176991150443,
      "grad_norm": 2.5724417355377227e-05,
      "learning_rate": 0.2948715811340921,
      "loss": 0.0,
      "num_input_tokens_seen": 1898840,
      "step": 3340
    },
    {
      "epoch": 58.690265486725664,
      "grad_norm": 1.1311556590953842e-05,
      "learning_rate": 0.294856298942442,
      "loss": 0.0,
      "num_input_tokens_seen": 1901560,
      "step": 3345
    },
    {
      "epoch": 58.7787610619469,
      "grad_norm": 2.4466533432132564e-05,
      "learning_rate": 0.2948409944121302,
      "loss": 0.0,
      "num_input_tokens_seen": 1904248,
      "step": 3350
    },
    {
      "epoch": 58.86725663716814,
      "grad_norm": 3.420401844778098e-05,
      "learning_rate": 0.29482566754551687,
      "loss": 0.0,
      "num_input_tokens_seen": 1906904,
      "step": 3355
    },
    {
      "epoch": 58.95575221238938,
      "grad_norm": 9.09080117708072e-05,
      "learning_rate": 0.2948103183449656,
      "loss": 0.0001,
      "num_input_tokens_seen": 1909800,
      "step": 3360
    },
    {
      "epoch": 59.0353982300885,
      "grad_norm": 2.430028689559549e-05,
      "learning_rate": 0.2947949468128435,
      "loss": 0.0,
      "num_input_tokens_seen": 1912184,
      "step": 3365
    },
    {
      "epoch": 59.123893805309734,
      "grad_norm": 7.5930161074211355e-06,
      "learning_rate": 0.2947795529515209,
      "loss": 0.0,
      "num_input_tokens_seen": 1914504,
      "step": 3370
    },
    {
      "epoch": 59.21238938053097,
      "grad_norm": 2.669124114618171e-05,
      "learning_rate": 0.29476413676337193,
      "loss": 0.0,
      "num_input_tokens_seen": 1917192,
      "step": 3375
    },
    {
      "epoch": 59.30088495575221,
      "grad_norm": 3.4852207591029583e-06,
      "learning_rate": 0.2947486982507738,
      "loss": 0.0,
      "num_input_tokens_seen": 1920056,
      "step": 3380
    },
    {
      "epoch": 59.389380530973455,
      "grad_norm": 2.071005837933626e-05,
      "learning_rate": 0.29473323741610735,
      "loss": 0.0,
      "num_input_tokens_seen": 1923048,
      "step": 3385
    },
    {
      "epoch": 59.47787610619469,
      "grad_norm": 9.639786185289267e-06,
      "learning_rate": 0.2947177542617569,
      "loss": 0.0,
      "num_input_tokens_seen": 1926328,
      "step": 3390
    },
    {
      "epoch": 59.56637168141593,
      "grad_norm": 6.62473394186236e-05,
      "learning_rate": 0.2947022487901101,
      "loss": 0.0,
      "num_input_tokens_seen": 1929640,
      "step": 3395
    },
    {
      "epoch": 59.65486725663717,
      "grad_norm": 4.277366315363906e-05,
      "learning_rate": 0.2946867210035581,
      "loss": 0.0,
      "num_input_tokens_seen": 1932248,
      "step": 3400
    },
    {
      "epoch": 59.65486725663717,
      "eval_loss": 0.3640179932117462,
      "eval_runtime": 0.9173,
      "eval_samples_per_second": 27.253,
      "eval_steps_per_second": 14.172,
      "num_input_tokens_seen": 1932248,
      "step": 3400
    },
    {
      "epoch": 59.743362831858406,
      "grad_norm": 3.996547820861451e-05,
      "learning_rate": 0.2946711709044954,
      "loss": 0.0,
      "num_input_tokens_seen": 1935208,
      "step": 3405
    },
    {
      "epoch": 59.83185840707964,
      "grad_norm": 5.84391182201216e-06,
      "learning_rate": 0.2946555984953202,
      "loss": 0.0,
      "num_input_tokens_seen": 1938216,
      "step": 3410
    },
    {
      "epoch": 59.92035398230089,
      "grad_norm": 7.713926606811583e-06,
      "learning_rate": 0.2946400037784338,
      "loss": 0.0,
      "num_input_tokens_seen": 1941000,
      "step": 3415
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.00011831547453766689,
      "learning_rate": 0.29462438675624114,
      "loss": 0.0001,
      "num_input_tokens_seen": 1943544,
      "step": 3420
    },
    {
      "epoch": 60.08849557522124,
      "grad_norm": 3.3156073186546564e-05,
      "learning_rate": 0.2946087474311506,
      "loss": 0.0,
      "num_input_tokens_seen": 1946152,
      "step": 3425
    },
    {
      "epoch": 60.176991150442475,
      "grad_norm": 1.1541232197487261e-05,
      "learning_rate": 0.294593085805574,
      "loss": 0.0,
      "num_input_tokens_seen": 1949240,
      "step": 3430
    },
    {
      "epoch": 60.26548672566372,
      "grad_norm": 7.341849413933232e-05,
      "learning_rate": 0.2945774018819264,
      "loss": 0.0,
      "num_input_tokens_seen": 1952248,
      "step": 3435
    },
    {
      "epoch": 60.35398230088496,
      "grad_norm": 4.056173929711804e-05,
      "learning_rate": 0.2945616956626266,
      "loss": 0.0,
      "num_input_tokens_seen": 1955064,
      "step": 3440
    },
    {
      "epoch": 60.442477876106196,
      "grad_norm": 1.5421097486978397e-05,
      "learning_rate": 0.2945459671500966,
      "loss": 0.0,
      "num_input_tokens_seen": 1957896,
      "step": 3445
    },
    {
      "epoch": 60.530973451327434,
      "grad_norm": 1.6032827261369675e-06,
      "learning_rate": 0.2945302163467621,
      "loss": 0.0,
      "num_input_tokens_seen": 1960632,
      "step": 3450
    },
    {
      "epoch": 60.61946902654867,
      "grad_norm": 4.69860497105401e-05,
      "learning_rate": 0.2945144432550519,
      "loss": 0.0,
      "num_input_tokens_seen": 1963416,
      "step": 3455
    },
    {
      "epoch": 60.70796460176991,
      "grad_norm": 1.8185739463660866e-05,
      "learning_rate": 0.29449864787739843,
      "loss": 0.0,
      "num_input_tokens_seen": 1966376,
      "step": 3460
    },
    {
      "epoch": 60.796460176991154,
      "grad_norm": 5.031113687437028e-05,
      "learning_rate": 0.2944828302162376,
      "loss": 0.0,
      "num_input_tokens_seen": 1969272,
      "step": 3465
    },
    {
      "epoch": 60.88495575221239,
      "grad_norm": 1.8644868760020472e-05,
      "learning_rate": 0.2944669902740087,
      "loss": 0.0,
      "num_input_tokens_seen": 1972168,
      "step": 3470
    },
    {
      "epoch": 60.97345132743363,
      "grad_norm": 1.5105496459000278e-05,
      "learning_rate": 0.2944511280531544,
      "loss": 0.0,
      "num_input_tokens_seen": 1975432,
      "step": 3475
    },
    {
      "epoch": 61.05309734513274,
      "grad_norm": 4.591782271745615e-06,
      "learning_rate": 0.29443524355612083,
      "loss": 0.0001,
      "num_input_tokens_seen": 1978104,
      "step": 3480
    },
    {
      "epoch": 61.14159292035398,
      "grad_norm": 3.4458971640560776e-05,
      "learning_rate": 0.29441933678535764,
      "loss": 0.0,
      "num_input_tokens_seen": 1980872,
      "step": 3485
    },
    {
      "epoch": 61.230088495575224,
      "grad_norm": 1.9792671082541347e-05,
      "learning_rate": 0.29440340774331786,
      "loss": 0.0,
      "num_input_tokens_seen": 1983864,
      "step": 3490
    },
    {
      "epoch": 61.31858407079646,
      "grad_norm": 2.2356160116032697e-05,
      "learning_rate": 0.2943874564324579,
      "loss": 0.0,
      "num_input_tokens_seen": 1986440,
      "step": 3495
    },
    {
      "epoch": 61.4070796460177,
      "grad_norm": 1.9416729628574103e-06,
      "learning_rate": 0.2943714828552376,
      "loss": 0.0,
      "num_input_tokens_seen": 1989336,
      "step": 3500
    },
    {
      "epoch": 61.49557522123894,
      "grad_norm": 2.1110237867105752e-05,
      "learning_rate": 0.29435548701412045,
      "loss": 0.0,
      "num_input_tokens_seen": 1991624,
      "step": 3505
    },
    {
      "epoch": 61.584070796460175,
      "grad_norm": 1.0600576388242189e-05,
      "learning_rate": 0.2943394689115731,
      "loss": 0.0,
      "num_input_tokens_seen": 1994440,
      "step": 3510
    },
    {
      "epoch": 61.67256637168141,
      "grad_norm": 1.8550312233855948e-05,
      "learning_rate": 0.29432342855006577,
      "loss": 0.0,
      "num_input_tokens_seen": 1997704,
      "step": 3515
    },
    {
      "epoch": 61.76106194690266,
      "grad_norm": 2.7752890673582442e-05,
      "learning_rate": 0.294307365932072,
      "loss": 0.0,
      "num_input_tokens_seen": 2000312,
      "step": 3520
    },
    {
      "epoch": 61.849557522123895,
      "grad_norm": 1.898999653349165e-05,
      "learning_rate": 0.294291281060069,
      "loss": 0.0,
      "num_input_tokens_seen": 2003096,
      "step": 3525
    },
    {
      "epoch": 61.93805309734513,
      "grad_norm": 2.5115748940152116e-05,
      "learning_rate": 0.29427517393653724,
      "loss": 0.0,
      "num_input_tokens_seen": 2006040,
      "step": 3530
    },
    {
      "epoch": 62.017699115044245,
      "grad_norm": 1.0225969163002446e-05,
      "learning_rate": 0.29425904456396046,
      "loss": 0.0,
      "num_input_tokens_seen": 2008712,
      "step": 3535
    },
    {
      "epoch": 62.10619469026549,
      "grad_norm": 8.17652380646905e-06,
      "learning_rate": 0.2942428929448262,
      "loss": 0.0,
      "num_input_tokens_seen": 2011672,
      "step": 3540
    },
    {
      "epoch": 62.19469026548673,
      "grad_norm": 3.286463106633164e-05,
      "learning_rate": 0.2942267190816252,
      "loss": 0.0,
      "num_input_tokens_seen": 2014808,
      "step": 3545
    },
    {
      "epoch": 62.283185840707965,
      "grad_norm": 1.4190652109391522e-05,
      "learning_rate": 0.2942105229768516,
      "loss": 0.0,
      "num_input_tokens_seen": 2018024,
      "step": 3550
    },
    {
      "epoch": 62.3716814159292,
      "grad_norm": 1.3379671145230532e-05,
      "learning_rate": 0.29419430463300306,
      "loss": 0.0,
      "num_input_tokens_seen": 2020360,
      "step": 3555
    },
    {
      "epoch": 62.46017699115044,
      "grad_norm": 7.918322808109224e-05,
      "learning_rate": 0.2941780640525808,
      "loss": 0.0,
      "num_input_tokens_seen": 2023064,
      "step": 3560
    },
    {
      "epoch": 62.54867256637168,
      "grad_norm": 2.0234196199453436e-05,
      "learning_rate": 0.2941618012380891,
      "loss": 0.0,
      "num_input_tokens_seen": 2026040,
      "step": 3565
    },
    {
      "epoch": 62.63716814159292,
      "grad_norm": 7.978651410667226e-06,
      "learning_rate": 0.29414551619203605,
      "loss": 0.0,
      "num_input_tokens_seen": 2028696,
      "step": 3570
    },
    {
      "epoch": 62.72566371681416,
      "grad_norm": 2.4655011657159775e-05,
      "learning_rate": 0.29412920891693295,
      "loss": 0.0,
      "num_input_tokens_seen": 2031720,
      "step": 3575
    },
    {
      "epoch": 62.8141592920354,
      "grad_norm": 1.907130535983015e-05,
      "learning_rate": 0.2941128794152946,
      "loss": 0.0,
      "num_input_tokens_seen": 2034760,
      "step": 3580
    },
    {
      "epoch": 62.902654867256636,
      "grad_norm": 6.197288894327357e-05,
      "learning_rate": 0.2940965276896392,
      "loss": 0.0,
      "num_input_tokens_seen": 2037736,
      "step": 3585
    },
    {
      "epoch": 62.991150442477874,
      "grad_norm": 3.361006383784115e-05,
      "learning_rate": 0.2940801537424884,
      "loss": 0.0,
      "num_input_tokens_seen": 2040008,
      "step": 3590
    },
    {
      "epoch": 63.07079646017699,
      "grad_norm": 6.9220732257235795e-06,
      "learning_rate": 0.2940637575763673,
      "loss": 0.0,
      "num_input_tokens_seen": 2042440,
      "step": 3595
    },
    {
      "epoch": 63.15929203539823,
      "grad_norm": 1.1588084817049094e-05,
      "learning_rate": 0.2940473391938043,
      "loss": 0.0,
      "num_input_tokens_seen": 2045464,
      "step": 3600
    },
    {
      "epoch": 63.15929203539823,
      "eval_loss": 0.36257556080818176,
      "eval_runtime": 0.9329,
      "eval_samples_per_second": 26.799,
      "eval_steps_per_second": 13.935,
      "num_input_tokens_seen": 2045464,
      "step": 3600
    },
    {
      "epoch": 63.24778761061947,
      "grad_norm": 1.743879693094641e-05,
      "learning_rate": 0.29403089859733145,
      "loss": 0.0,
      "num_input_tokens_seen": 2048200,
      "step": 3605
    },
    {
      "epoch": 63.336283185840706,
      "grad_norm": 2.00543745449977e-05,
      "learning_rate": 0.294014435789484,
      "loss": 0.0,
      "num_input_tokens_seen": 2051080,
      "step": 3610
    },
    {
      "epoch": 63.424778761061944,
      "grad_norm": 2.1424470105557702e-05,
      "learning_rate": 0.2939979507728007,
      "loss": 0.0,
      "num_input_tokens_seen": 2054040,
      "step": 3615
    },
    {
      "epoch": 63.51327433628319,
      "grad_norm": 3.447451308602467e-05,
      "learning_rate": 0.2939814435498239,
      "loss": 0.0,
      "num_input_tokens_seen": 2056968,
      "step": 3620
    },
    {
      "epoch": 63.60176991150443,
      "grad_norm": 1.3003301319258753e-05,
      "learning_rate": 0.29396491412309905,
      "loss": 0.0,
      "num_input_tokens_seen": 2059704,
      "step": 3625
    },
    {
      "epoch": 63.690265486725664,
      "grad_norm": 3.005195139849093e-06,
      "learning_rate": 0.2939483624951753,
      "loss": 0.0,
      "num_input_tokens_seen": 2062520,
      "step": 3630
    },
    {
      "epoch": 63.7787610619469,
      "grad_norm": 1.224400784849422e-05,
      "learning_rate": 0.2939317886686051,
      "loss": 0.0,
      "num_input_tokens_seen": 2065688,
      "step": 3635
    },
    {
      "epoch": 63.86725663716814,
      "grad_norm": 1.872412394732237e-05,
      "learning_rate": 0.2939151926459443,
      "loss": 0.0,
      "num_input_tokens_seen": 2068472,
      "step": 3640
    },
    {
      "epoch": 63.95575221238938,
      "grad_norm": 3.355348962941207e-05,
      "learning_rate": 0.2938985744297522,
      "loss": 0.0,
      "num_input_tokens_seen": 2071192,
      "step": 3645
    },
    {
      "epoch": 64.03539823008849,
      "grad_norm": 2.813566788972821e-05,
      "learning_rate": 0.29388193402259166,
      "loss": 0.0,
      "num_input_tokens_seen": 2073424,
      "step": 3650
    },
    {
      "epoch": 64.12389380530973,
      "grad_norm": 3.366770033608191e-05,
      "learning_rate": 0.29386527142702873,
      "loss": 0.0,
      "num_input_tokens_seen": 2076352,
      "step": 3655
    },
    {
      "epoch": 64.21238938053098,
      "grad_norm": 2.9681104933843017e-05,
      "learning_rate": 0.293848586645633,
      "loss": 0.0,
      "num_input_tokens_seen": 2078912,
      "step": 3660
    },
    {
      "epoch": 64.30088495575221,
      "grad_norm": 5.300247357808985e-05,
      "learning_rate": 0.2938318796809775,
      "loss": 0.0,
      "num_input_tokens_seen": 2081904,
      "step": 3665
    },
    {
      "epoch": 64.38938053097345,
      "grad_norm": 1.645704105612822e-05,
      "learning_rate": 0.29381515053563867,
      "loss": 0.0,
      "num_input_tokens_seen": 2084768,
      "step": 3670
    },
    {
      "epoch": 64.47787610619469,
      "grad_norm": 1.1793941666837782e-05,
      "learning_rate": 0.29379839921219636,
      "loss": 0.0,
      "num_input_tokens_seen": 2087792,
      "step": 3675
    },
    {
      "epoch": 64.56637168141593,
      "grad_norm": 2.8017246222589165e-05,
      "learning_rate": 0.2937816257132338,
      "loss": 0.0,
      "num_input_tokens_seen": 2090336,
      "step": 3680
    },
    {
      "epoch": 64.65486725663717,
      "grad_norm": 3.0241208150982857e-06,
      "learning_rate": 0.2937648300413376,
      "loss": 0.0,
      "num_input_tokens_seen": 2093264,
      "step": 3685
    },
    {
      "epoch": 64.7433628318584,
      "grad_norm": 5.729058102588169e-05,
      "learning_rate": 0.293748012199098,
      "loss": 0.0,
      "num_input_tokens_seen": 2095856,
      "step": 3690
    },
    {
      "epoch": 64.83185840707965,
      "grad_norm": 2.1401046979008242e-06,
      "learning_rate": 0.29373117218910844,
      "loss": 0.0,
      "num_input_tokens_seen": 2099568,
      "step": 3695
    },
    {
      "epoch": 64.92035398230088,
      "grad_norm": 1.6656838852213696e-05,
      "learning_rate": 0.2937143100139659,
      "loss": 0.0,
      "num_input_tokens_seen": 2102688,
      "step": 3700
    },
    {
      "epoch": 65.0,
      "grad_norm": 6.4887540247582365e-06,
      "learning_rate": 0.29369742567627083,
      "loss": 0.0,
      "num_input_tokens_seen": 2104960,
      "step": 3705
    },
    {
      "epoch": 65.08849557522124,
      "grad_norm": 9.477848834649194e-06,
      "learning_rate": 0.29368051917862675,
      "loss": 0.0,
      "num_input_tokens_seen": 2107680,
      "step": 3710
    },
    {
      "epoch": 65.17699115044248,
      "grad_norm": 1.1240840649406891e-05,
      "learning_rate": 0.2936635905236411,
      "loss": 0.0,
      "num_input_tokens_seen": 2110384,
      "step": 3715
    },
    {
      "epoch": 65.26548672566372,
      "grad_norm": 3.6127050407230854e-05,
      "learning_rate": 0.2936466397139244,
      "loss": 0.0,
      "num_input_tokens_seen": 2113184,
      "step": 3720
    },
    {
      "epoch": 65.35398230088495,
      "grad_norm": 2.2430753233493306e-05,
      "learning_rate": 0.2936296667520907,
      "loss": 0.0,
      "num_input_tokens_seen": 2116048,
      "step": 3725
    },
    {
      "epoch": 65.4424778761062,
      "grad_norm": 1.661307214817498e-05,
      "learning_rate": 0.2936126716407574,
      "loss": 0.0,
      "num_input_tokens_seen": 2119344,
      "step": 3730
    },
    {
      "epoch": 65.53097345132744,
      "grad_norm": 8.324688678840175e-06,
      "learning_rate": 0.29359565438254537,
      "loss": 0.0,
      "num_input_tokens_seen": 2122240,
      "step": 3735
    },
    {
      "epoch": 65.61946902654867,
      "grad_norm": 1.6947155017987825e-05,
      "learning_rate": 0.29357861498007887,
      "loss": 0.0,
      "num_input_tokens_seen": 2124912,
      "step": 3740
    },
    {
      "epoch": 65.70796460176992,
      "grad_norm": 1.5571436961181462e-05,
      "learning_rate": 0.29356155343598567,
      "loss": 0.0,
      "num_input_tokens_seen": 2128208,
      "step": 3745
    },
    {
      "epoch": 65.79646017699115,
      "grad_norm": 2.6291845642845146e-05,
      "learning_rate": 0.2935444697528968,
      "loss": 0.0,
      "num_input_tokens_seen": 2130848,
      "step": 3750
    },
    {
      "epoch": 65.88495575221239,
      "grad_norm": 2.0684970877482556e-05,
      "learning_rate": 0.2935273639334468,
      "loss": 0.0,
      "num_input_tokens_seen": 2133728,
      "step": 3755
    },
    {
      "epoch": 65.97345132743362,
      "grad_norm": 2.110229615936987e-05,
      "learning_rate": 0.29351023598027365,
      "loss": 0.0,
      "num_input_tokens_seen": 2136624,
      "step": 3760
    },
    {
      "epoch": 66.05309734513274,
      "grad_norm": 1.0571146049187519e-05,
      "learning_rate": 0.2934930858960186,
      "loss": 0.0,
      "num_input_tokens_seen": 2139000,
      "step": 3765
    },
    {
      "epoch": 66.14159292035399,
      "grad_norm": 1.583739503985271e-05,
      "learning_rate": 0.29347591368332643,
      "loss": 0.0,
      "num_input_tokens_seen": 2142296,
      "step": 3770
    },
    {
      "epoch": 66.23008849557522,
      "grad_norm": 7.381248451565625e-06,
      "learning_rate": 0.2934587193448454,
      "loss": 0.0,
      "num_input_tokens_seen": 2145432,
      "step": 3775
    },
    {
      "epoch": 66.31858407079646,
      "grad_norm": 4.2322859371779487e-05,
      "learning_rate": 0.29344150288322696,
      "loss": 0.0,
      "num_input_tokens_seen": 2147960,
      "step": 3780
    },
    {
      "epoch": 66.40707964601769,
      "grad_norm": 1.9186632925993763e-05,
      "learning_rate": 0.2934242643011263,
      "loss": 0.0,
      "num_input_tokens_seen": 2151112,
      "step": 3785
    },
    {
      "epoch": 66.49557522123894,
      "grad_norm": 1.4251646689444897e-06,
      "learning_rate": 0.2934070036012016,
      "loss": 0.0,
      "num_input_tokens_seen": 2153448,
      "step": 3790
    },
    {
      "epoch": 66.58407079646018,
      "grad_norm": 3.9978836866794154e-05,
      "learning_rate": 0.29338972078611475,
      "loss": 0.0,
      "num_input_tokens_seen": 2156280,
      "step": 3795
    },
    {
      "epoch": 66.67256637168141,
      "grad_norm": 3.4388358471915126e-05,
      "learning_rate": 0.2933724158585311,
      "loss": 0.0,
      "num_input_tokens_seen": 2159128,
      "step": 3800
    },
    {
      "epoch": 66.67256637168141,
      "eval_loss": 0.3677736520767212,
      "eval_runtime": 0.9122,
      "eval_samples_per_second": 27.405,
      "eval_steps_per_second": 14.251,
      "num_input_tokens_seen": 2159128,
      "step": 3800
    },
    {
      "epoch": 66.76106194690266,
      "grad_norm": 1.5423656805069186e-05,
      "learning_rate": 0.29335508882111916,
      "loss": 0.0,
      "num_input_tokens_seen": 2162344,
      "step": 3805
    },
    {
      "epoch": 66.84955752212389,
      "grad_norm": 3.076591156059294e-06,
      "learning_rate": 0.29333773967655097,
      "loss": 0.0,
      "num_input_tokens_seen": 2164920,
      "step": 3810
    },
    {
      "epoch": 66.93805309734513,
      "grad_norm": 4.15208050981164e-05,
      "learning_rate": 0.2933203684275021,
      "loss": 0.0,
      "num_input_tokens_seen": 2168024,
      "step": 3815
    },
    {
      "epoch": 67.01769911504425,
      "grad_norm": 1.0525621291890275e-05,
      "learning_rate": 0.2933029750766513,
      "loss": 0.0,
      "num_input_tokens_seen": 2170720,
      "step": 3820
    },
    {
      "epoch": 67.10619469026548,
      "grad_norm": 3.237776400055736e-05,
      "learning_rate": 0.2932855596266809,
      "loss": 0.0,
      "num_input_tokens_seen": 2173312,
      "step": 3825
    },
    {
      "epoch": 67.19469026548673,
      "grad_norm": 2.807303189911181e-06,
      "learning_rate": 0.2932681220802765,
      "loss": 0.0,
      "num_input_tokens_seen": 2175984,
      "step": 3830
    },
    {
      "epoch": 67.28318584070796,
      "grad_norm": 1.4826597180217505e-05,
      "learning_rate": 0.2932506624401274,
      "loss": 0.0,
      "num_input_tokens_seen": 2178624,
      "step": 3835
    },
    {
      "epoch": 67.3716814159292,
      "grad_norm": 2.9049484510323964e-05,
      "learning_rate": 0.29323318070892584,
      "loss": 0.0,
      "num_input_tokens_seen": 2181088,
      "step": 3840
    },
    {
      "epoch": 67.46017699115045,
      "grad_norm": 1.7954704162548296e-05,
      "learning_rate": 0.29321567688936784,
      "loss": 0.0,
      "num_input_tokens_seen": 2184736,
      "step": 3845
    },
    {
      "epoch": 67.54867256637168,
      "grad_norm": 3.859784919768572e-05,
      "learning_rate": 0.29319815098415275,
      "loss": 0.0,
      "num_input_tokens_seen": 2187296,
      "step": 3850
    },
    {
      "epoch": 67.63716814159292,
      "grad_norm": 8.108509064186364e-06,
      "learning_rate": 0.2931806029959832,
      "loss": 0.0,
      "num_input_tokens_seen": 2190432,
      "step": 3855
    },
    {
      "epoch": 67.72566371681415,
      "grad_norm": 2.2443511625169776e-05,
      "learning_rate": 0.29316303292756535,
      "loss": 0.0,
      "num_input_tokens_seen": 2193568,
      "step": 3860
    },
    {
      "epoch": 67.8141592920354,
      "grad_norm": 2.516521817597095e-05,
      "learning_rate": 0.29314544078160876,
      "loss": 0.0,
      "num_input_tokens_seen": 2196464,
      "step": 3865
    },
    {
      "epoch": 67.90265486725664,
      "grad_norm": 2.00249287445331e-05,
      "learning_rate": 0.2931278265608263,
      "loss": 0.0,
      "num_input_tokens_seen": 2199584,
      "step": 3870
    },
    {
      "epoch": 67.99115044247787,
      "grad_norm": 3.619547715061344e-05,
      "learning_rate": 0.29311019026793433,
      "loss": 0.0,
      "num_input_tokens_seen": 2202432,
      "step": 3875
    },
    {
      "epoch": 68.070796460177,
      "grad_norm": 2.2741040083928965e-05,
      "learning_rate": 0.29309253190565254,
      "loss": 0.0,
      "num_input_tokens_seen": 2205616,
      "step": 3880
    },
    {
      "epoch": 68.15929203539822,
      "grad_norm": 1.1188014468643814e-05,
      "learning_rate": 0.2930748514767042,
      "loss": 0.0,
      "num_input_tokens_seen": 2208608,
      "step": 3885
    },
    {
      "epoch": 68.24778761061947,
      "grad_norm": 1.85035023605451e-05,
      "learning_rate": 0.29305714898381574,
      "loss": 0.0,
      "num_input_tokens_seen": 2210912,
      "step": 3890
    },
    {
      "epoch": 68.33628318584071,
      "grad_norm": 2.842759931809269e-05,
      "learning_rate": 0.29303942442971714,
      "loss": 0.0,
      "num_input_tokens_seen": 2213952,
      "step": 3895
    },
    {
      "epoch": 68.42477876106194,
      "grad_norm": 2.0531195332296193e-05,
      "learning_rate": 0.2930216778171417,
      "loss": 0.0,
      "num_input_tokens_seen": 2217024,
      "step": 3900
    },
    {
      "epoch": 68.51327433628319,
      "grad_norm": 1.0467572792549618e-05,
      "learning_rate": 0.2930039091488263,
      "loss": 0.0,
      "num_input_tokens_seen": 2219680,
      "step": 3905
    },
    {
      "epoch": 68.60176991150442,
      "grad_norm": 6.6619077188079245e-06,
      "learning_rate": 0.29298611842751093,
      "loss": 0.0,
      "num_input_tokens_seen": 2222384,
      "step": 3910
    },
    {
      "epoch": 68.69026548672566,
      "grad_norm": 1.021869138639886e-05,
      "learning_rate": 0.29296830565593923,
      "loss": 0.0,
      "num_input_tokens_seen": 2224864,
      "step": 3915
    },
    {
      "epoch": 68.77876106194691,
      "grad_norm": 2.8807148737541866e-06,
      "learning_rate": 0.2929504708368582,
      "loss": 0.0,
      "num_input_tokens_seen": 2228192,
      "step": 3920
    },
    {
      "epoch": 68.86725663716814,
      "grad_norm": 1.1849306247313507e-05,
      "learning_rate": 0.29293261397301806,
      "loss": 0.0,
      "num_input_tokens_seen": 2230976,
      "step": 3925
    },
    {
      "epoch": 68.95575221238938,
      "grad_norm": 1.0004808245867025e-05,
      "learning_rate": 0.29291473506717275,
      "loss": 0.0,
      "num_input_tokens_seen": 2233968,
      "step": 3930
    },
    {
      "epoch": 69.03539823008849,
      "grad_norm": 3.3840062769741053e-06,
      "learning_rate": 0.29289683412207923,
      "loss": 0.0,
      "num_input_tokens_seen": 2236392,
      "step": 3935
    },
    {
      "epoch": 69.12389380530973,
      "grad_norm": 5.918141141592059e-06,
      "learning_rate": 0.29287891114049813,
      "loss": 0.0,
      "num_input_tokens_seen": 2239032,
      "step": 3940
    },
    {
      "epoch": 69.21238938053098,
      "grad_norm": 3.469240982667543e-05,
      "learning_rate": 0.29286096612519347,
      "loss": 0.0,
      "num_input_tokens_seen": 2241912,
      "step": 3945
    },
    {
      "epoch": 69.30088495575221,
      "grad_norm": 7.144365099520655e-06,
      "learning_rate": 0.2928429990789325,
      "loss": 0.0,
      "num_input_tokens_seen": 2245016,
      "step": 3950
    },
    {
      "epoch": 69.38938053097345,
      "grad_norm": 1.3272005162434652e-05,
      "learning_rate": 0.29282501000448596,
      "loss": 0.0,
      "num_input_tokens_seen": 2247720,
      "step": 3955
    },
    {
      "epoch": 69.47787610619469,
      "grad_norm": 6.599826519959606e-06,
      "learning_rate": 0.2928069989046281,
      "loss": 0.0,
      "num_input_tokens_seen": 2250360,
      "step": 3960
    },
    {
      "epoch": 69.56637168141593,
      "grad_norm": 9.144034265773371e-06,
      "learning_rate": 0.2927889657821363,
      "loss": 0.0,
      "num_input_tokens_seen": 2253272,
      "step": 3965
    },
    {
      "epoch": 69.65486725663717,
      "grad_norm": 2.5255953914893325e-06,
      "learning_rate": 0.2927709106397916,
      "loss": 0.0,
      "num_input_tokens_seen": 2256120,
      "step": 3970
    },
    {
      "epoch": 69.7433628318584,
      "grad_norm": 1.5825173250050284e-05,
      "learning_rate": 0.29275283348037834,
      "loss": 0.0,
      "num_input_tokens_seen": 2259400,
      "step": 3975
    },
    {
      "epoch": 69.83185840707965,
      "grad_norm": 2.662315637280699e-05,
      "learning_rate": 0.29273473430668423,
      "loss": 0.0,
      "num_input_tokens_seen": 2262344,
      "step": 3980
    },
    {
      "epoch": 69.92035398230088,
      "grad_norm": 2.2529902707901783e-05,
      "learning_rate": 0.2927166131215003,
      "loss": 0.0,
      "num_input_tokens_seen": 2265160,
      "step": 3985
    },
    {
      "epoch": 70.0,
      "grad_norm": 4.212867224850925e-06,
      "learning_rate": 0.2926984699276212,
      "loss": 0.0,
      "num_input_tokens_seen": 2267528,
      "step": 3990
    },
    {
      "epoch": 70.08849557522124,
      "grad_norm": 3.805399683187716e-05,
      "learning_rate": 0.29268030472784473,
      "loss": 0.0,
      "num_input_tokens_seen": 2270056,
      "step": 3995
    },
    {
      "epoch": 70.17699115044248,
      "grad_norm": 9.422032235306688e-06,
      "learning_rate": 0.2926621175249723,
      "loss": 0.0,
      "num_input_tokens_seen": 2272792,
      "step": 4000
    },
    {
      "epoch": 70.17699115044248,
      "eval_loss": 0.36759963631629944,
      "eval_runtime": 0.9198,
      "eval_samples_per_second": 27.179,
      "eval_steps_per_second": 14.133,
      "num_input_tokens_seen": 2272792,
      "step": 4000
    },
    {
      "epoch": 70.26548672566372,
      "grad_norm": 2.9090702810208313e-05,
      "learning_rate": 0.29264390832180853,
      "loss": 0.0,
      "num_input_tokens_seen": 2275432,
      "step": 4005
    },
    {
      "epoch": 70.35398230088495,
      "grad_norm": 2.2283124053501524e-05,
      "learning_rate": 0.29262567712116144,
      "loss": 0.0,
      "num_input_tokens_seen": 2278280,
      "step": 4010
    },
    {
      "epoch": 70.4424778761062,
      "grad_norm": 6.148783086246112e-06,
      "learning_rate": 0.29260742392584266,
      "loss": 0.0,
      "num_input_tokens_seen": 2281128,
      "step": 4015
    },
    {
      "epoch": 70.53097345132744,
      "grad_norm": 1.527516360511072e-05,
      "learning_rate": 0.292589148738667,
      "loss": 0.0,
      "num_input_tokens_seen": 2284376,
      "step": 4020
    },
    {
      "epoch": 70.61946902654867,
      "grad_norm": 1.203413830808131e-05,
      "learning_rate": 0.2925708515624527,
      "loss": 0.0,
      "num_input_tokens_seen": 2287208,
      "step": 4025
    },
    {
      "epoch": 70.70796460176992,
      "grad_norm": 1.9319243165227817e-06,
      "learning_rate": 0.29255253240002144,
      "loss": 0.0,
      "num_input_tokens_seen": 2290024,
      "step": 4030
    },
    {
      "epoch": 70.79646017699115,
      "grad_norm": 8.436039934167638e-06,
      "learning_rate": 0.2925341912541983,
      "loss": 0.0,
      "num_input_tokens_seen": 2292616,
      "step": 4035
    },
    {
      "epoch": 70.88495575221239,
      "grad_norm": 7.0400778895418625e-06,
      "learning_rate": 0.2925158281278116,
      "loss": 0.0,
      "num_input_tokens_seen": 2296104,
      "step": 4040
    },
    {
      "epoch": 70.97345132743362,
      "grad_norm": 1.4069966709939763e-05,
      "learning_rate": 0.29249744302369324,
      "loss": 0.0,
      "num_input_tokens_seen": 2298888,
      "step": 4045
    },
    {
      "epoch": 71.05309734513274,
      "grad_norm": 2.107543514284771e-05,
      "learning_rate": 0.29247903594467844,
      "loss": 0.0,
      "num_input_tokens_seen": 2301192,
      "step": 4050
    },
    {
      "epoch": 71.14159292035399,
      "grad_norm": 2.4853881768649444e-05,
      "learning_rate": 0.2924606068936058,
      "loss": 0.0,
      "num_input_tokens_seen": 2304312,
      "step": 4055
    },
    {
      "epoch": 71.23008849557522,
      "grad_norm": 1.627755591471214e-05,
      "learning_rate": 0.2924421558733173,
      "loss": 0.0,
      "num_input_tokens_seen": 2306824,
      "step": 4060
    },
    {
      "epoch": 71.31858407079646,
      "grad_norm": 2.586418304417748e-05,
      "learning_rate": 0.2924236828866583,
      "loss": 0.0,
      "num_input_tokens_seen": 2309800,
      "step": 4065
    },
    {
      "epoch": 71.40707964601769,
      "grad_norm": 2.6714513296610676e-05,
      "learning_rate": 0.29240518793647763,
      "loss": 0.0,
      "num_input_tokens_seen": 2312728,
      "step": 4070
    },
    {
      "epoch": 71.49557522123894,
      "grad_norm": 2.024080094997771e-05,
      "learning_rate": 0.29238667102562743,
      "loss": 0.0,
      "num_input_tokens_seen": 2315608,
      "step": 4075
    },
    {
      "epoch": 71.58407079646018,
      "grad_norm": 4.3234398617642e-05,
      "learning_rate": 0.29236813215696317,
      "loss": 0.0,
      "num_input_tokens_seen": 2318264,
      "step": 4080
    },
    {
      "epoch": 71.67256637168141,
      "grad_norm": 1.0371915095674922e-06,
      "learning_rate": 0.2923495713333439,
      "loss": 0.0,
      "num_input_tokens_seen": 2321032,
      "step": 4085
    },
    {
      "epoch": 71.76106194690266,
      "grad_norm": 2.0506815417320468e-05,
      "learning_rate": 0.29233098855763173,
      "loss": 0.0,
      "num_input_tokens_seen": 2324568,
      "step": 4090
    },
    {
      "epoch": 71.84955752212389,
      "grad_norm": 1.4703994565934408e-05,
      "learning_rate": 0.29231238383269254,
      "loss": 0.0,
      "num_input_tokens_seen": 2327144,
      "step": 4095
    },
    {
      "epoch": 71.93805309734513,
      "grad_norm": 8.721726771909744e-06,
      "learning_rate": 0.2922937571613954,
      "loss": 0.0,
      "num_input_tokens_seen": 2330104,
      "step": 4100
    },
    {
      "epoch": 72.01769911504425,
      "grad_norm": 1.4340456800709944e-05,
      "learning_rate": 0.29227510854661265,
      "loss": 0.0,
      "num_input_tokens_seen": 2332656,
      "step": 4105
    },
    {
      "epoch": 72.10619469026548,
      "grad_norm": 3.811505848716479e-06,
      "learning_rate": 0.29225643799122025,
      "loss": 0.0,
      "num_input_tokens_seen": 2335584,
      "step": 4110
    },
    {
      "epoch": 72.19469026548673,
      "grad_norm": 2.8085227313567884e-05,
      "learning_rate": 0.2922377454980974,
      "loss": 0.0,
      "num_input_tokens_seen": 2338048,
      "step": 4115
    },
    {
      "epoch": 72.28318584070796,
      "grad_norm": 1.8555034330347553e-05,
      "learning_rate": 0.29221903107012676,
      "loss": 0.0,
      "num_input_tokens_seen": 2340848,
      "step": 4120
    },
    {
      "epoch": 72.3716814159292,
      "grad_norm": 8.461062861897517e-06,
      "learning_rate": 0.29220029471019426,
      "loss": 0.0,
      "num_input_tokens_seen": 2344112,
      "step": 4125
    },
    {
      "epoch": 72.46017699115045,
      "grad_norm": 1.403624628437683e-05,
      "learning_rate": 0.2921815364211893,
      "loss": 0.0,
      "num_input_tokens_seen": 2346544,
      "step": 4130
    },
    {
      "epoch": 72.54867256637168,
      "grad_norm": 3.411239231354557e-05,
      "learning_rate": 0.29216275620600474,
      "loss": 0.0,
      "num_input_tokens_seen": 2349456,
      "step": 4135
    },
    {
      "epoch": 72.63716814159292,
      "grad_norm": 3.712247007570113e-06,
      "learning_rate": 0.29214395406753657,
      "loss": 0.0,
      "num_input_tokens_seen": 2352160,
      "step": 4140
    },
    {
      "epoch": 72.72566371681415,
      "grad_norm": 3.315218236821238e-06,
      "learning_rate": 0.2921251300086844,
      "loss": 0.0,
      "num_input_tokens_seen": 2355104,
      "step": 4145
    },
    {
      "epoch": 72.8141592920354,
      "grad_norm": 2.315730853297282e-05,
      "learning_rate": 0.2921062840323511,
      "loss": 0.0,
      "num_input_tokens_seen": 2358288,
      "step": 4150
    },
    {
      "epoch": 72.90265486725664,
      "grad_norm": 1.399136999680195e-05,
      "learning_rate": 0.29208741614144307,
      "loss": 0.0,
      "num_input_tokens_seen": 2361056,
      "step": 4155
    },
    {
      "epoch": 72.99115044247787,
      "grad_norm": 8.795711437414866e-06,
      "learning_rate": 0.2920685263388698,
      "loss": 0.0,
      "num_input_tokens_seen": 2364160,
      "step": 4160
    },
    {
      "epoch": 73.070796460177,
      "grad_norm": 9.402758223586716e-06,
      "learning_rate": 0.2920496146275445,
      "loss": 0.0,
      "num_input_tokens_seen": 2366384,
      "step": 4165
    },
    {
      "epoch": 73.15929203539822,
      "grad_norm": 1.2161111044406425e-05,
      "learning_rate": 0.29203068101038343,
      "loss": 0.0,
      "num_input_tokens_seen": 2369632,
      "step": 4170
    },
    {
      "epoch": 73.24778761061947,
      "grad_norm": 1.0335465958632994e-05,
      "learning_rate": 0.2920117254903065,
      "loss": 0.0,
      "num_input_tokens_seen": 2372880,
      "step": 4175
    },
    {
      "epoch": 73.33628318584071,
      "grad_norm": 1.2378063729556743e-05,
      "learning_rate": 0.29199274807023695,
      "loss": 0.0,
      "num_input_tokens_seen": 2375344,
      "step": 4180
    },
    {
      "epoch": 73.42477876106194,
      "grad_norm": 2.986155595863238e-05,
      "learning_rate": 0.29197374875310117,
      "loss": 0.0,
      "num_input_tokens_seen": 2378288,
      "step": 4185
    },
    {
      "epoch": 73.51327433628319,
      "grad_norm": 1.3800038232147926e-06,
      "learning_rate": 0.2919547275418292,
      "loss": 0.0,
      "num_input_tokens_seen": 2381600,
      "step": 4190
    },
    {
      "epoch": 73.60176991150442,
      "grad_norm": 2.637521356518846e-05,
      "learning_rate": 0.29193568443935436,
      "loss": 0.0,
      "num_input_tokens_seen": 2384096,
      "step": 4195
    },
    {
      "epoch": 73.69026548672566,
      "grad_norm": 2.1674657546100207e-05,
      "learning_rate": 0.2919166194486133,
      "loss": 0.0,
      "num_input_tokens_seen": 2387344,
      "step": 4200
    },
    {
      "epoch": 73.69026548672566,
      "eval_loss": 0.36723047494888306,
      "eval_runtime": 0.9354,
      "eval_samples_per_second": 26.726,
      "eval_steps_per_second": 13.898,
      "num_input_tokens_seen": 2387344,
      "step": 4200
    },
    {
      "epoch": 73.77876106194691,
      "grad_norm": 1.824320133891888e-05,
      "learning_rate": 0.2918975325725461,
      "loss": 0.0,
      "num_input_tokens_seen": 2390256,
      "step": 4205
    },
    {
      "epoch": 73.86725663716814,
      "grad_norm": 4.217048171994975e-06,
      "learning_rate": 0.29187842381409607,
      "loss": 0.0,
      "num_input_tokens_seen": 2393056,
      "step": 4210
    },
    {
      "epoch": 73.95575221238938,
      "grad_norm": 5.750944637838984e-06,
      "learning_rate": 0.29185929317621023,
      "loss": 0.0,
      "num_input_tokens_seen": 2395600,
      "step": 4215
    },
    {
      "epoch": 74.03539823008849,
      "grad_norm": 1.1150009413540829e-05,
      "learning_rate": 0.29184014066183867,
      "loss": 0.0,
      "num_input_tokens_seen": 2397832,
      "step": 4220
    },
    {
      "epoch": 74.12389380530973,
      "grad_norm": 3.069782223974471e-06,
      "learning_rate": 0.2918209662739349,
      "loss": 0.0,
      "num_input_tokens_seen": 2400872,
      "step": 4225
    },
    {
      "epoch": 74.21238938053098,
      "grad_norm": 3.798725811066106e-05,
      "learning_rate": 0.29180177001545593,
      "loss": 0.0,
      "num_input_tokens_seen": 2403544,
      "step": 4230
    },
    {
      "epoch": 74.30088495575221,
      "grad_norm": 2.520867383282166e-05,
      "learning_rate": 0.29178255188936203,
      "loss": 0.0,
      "num_input_tokens_seen": 2406744,
      "step": 4235
    },
    {
      "epoch": 74.38938053097345,
      "grad_norm": 1.0224712241324596e-05,
      "learning_rate": 0.2917633118986169,
      "loss": 0.0,
      "num_input_tokens_seen": 2409880,
      "step": 4240
    },
    {
      "epoch": 74.47787610619469,
      "grad_norm": 9.725671588967089e-06,
      "learning_rate": 0.2917440500461875,
      "loss": 0.0,
      "num_input_tokens_seen": 2412408,
      "step": 4245
    },
    {
      "epoch": 74.56637168141593,
      "grad_norm": 2.4388580186496256e-06,
      "learning_rate": 0.29172476633504435,
      "loss": 0.0,
      "num_input_tokens_seen": 2415304,
      "step": 4250
    },
    {
      "epoch": 74.65486725663717,
      "grad_norm": 3.5032303458137903e-06,
      "learning_rate": 0.2917054607681612,
      "loss": 0.0,
      "num_input_tokens_seen": 2418088,
      "step": 4255
    },
    {
      "epoch": 74.7433628318584,
      "grad_norm": 9.095188943319954e-06,
      "learning_rate": 0.29168613334851523,
      "loss": 0.0,
      "num_input_tokens_seen": 2420872,
      "step": 4260
    },
    {
      "epoch": 74.83185840707965,
      "grad_norm": 1.1440006346674636e-05,
      "learning_rate": 0.2916667840790869,
      "loss": 0.0,
      "num_input_tokens_seen": 2423592,
      "step": 4265
    },
    {
      "epoch": 74.92035398230088,
      "grad_norm": 2.935229076683754e-06,
      "learning_rate": 0.2916474129628603,
      "loss": 0.0,
      "num_input_tokens_seen": 2426584,
      "step": 4270
    },
    {
      "epoch": 75.0,
      "grad_norm": 2.804813084367197e-05,
      "learning_rate": 0.29162802000282245,
      "loss": 0.0,
      "num_input_tokens_seen": 2428728,
      "step": 4275
    },
    {
      "epoch": 75.08849557522124,
      "grad_norm": 2.6733125196187757e-05,
      "learning_rate": 0.2916086052019642,
      "loss": 0.0,
      "num_input_tokens_seen": 2431576,
      "step": 4280
    },
    {
      "epoch": 75.17699115044248,
      "grad_norm": 2.111234061885625e-05,
      "learning_rate": 0.2915891685632794,
      "loss": 0.0,
      "num_input_tokens_seen": 2434488,
      "step": 4285
    },
    {
      "epoch": 75.26548672566372,
      "grad_norm": 1.1898192497028504e-05,
      "learning_rate": 0.29156971008976545,
      "loss": 0.0,
      "num_input_tokens_seen": 2437208,
      "step": 4290
    },
    {
      "epoch": 75.35398230088495,
      "grad_norm": 1.543630969536025e-05,
      "learning_rate": 0.2915502297844232,
      "loss": 0.0,
      "num_input_tokens_seen": 2440120,
      "step": 4295
    },
    {
      "epoch": 75.4424778761062,
      "grad_norm": 3.0471141144516878e-05,
      "learning_rate": 0.2915307276502566,
      "loss": 0.0,
      "num_input_tokens_seen": 2442856,
      "step": 4300
    },
    {
      "epoch": 75.53097345132744,
      "grad_norm": 9.20421280170558e-06,
      "learning_rate": 0.29151120369027334,
      "loss": 0.0,
      "num_input_tokens_seen": 2445816,
      "step": 4305
    },
    {
      "epoch": 75.61946902654867,
      "grad_norm": 5.196699930820614e-06,
      "learning_rate": 0.29149165790748405,
      "loss": 0.0,
      "num_input_tokens_seen": 2448520,
      "step": 4310
    },
    {
      "epoch": 75.70796460176992,
      "grad_norm": 3.7334625631046947e-06,
      "learning_rate": 0.291472090304903,
      "loss": 0.0,
      "num_input_tokens_seen": 2451704,
      "step": 4315
    },
    {
      "epoch": 75.79646017699115,
      "grad_norm": 7.74161071603885e-06,
      "learning_rate": 0.2914525008855478,
      "loss": 0.0,
      "num_input_tokens_seen": 2454392,
      "step": 4320
    },
    {
      "epoch": 75.88495575221239,
      "grad_norm": 1.5376670489786193e-05,
      "learning_rate": 0.2914328896524394,
      "loss": 0.0,
      "num_input_tokens_seen": 2457304,
      "step": 4325
    },
    {
      "epoch": 75.97345132743362,
      "grad_norm": 6.465412297984585e-06,
      "learning_rate": 0.291413256608602,
      "loss": 0.0,
      "num_input_tokens_seen": 2460728,
      "step": 4330
    },
    {
      "epoch": 76.05309734513274,
      "grad_norm": 2.3227035853778943e-05,
      "learning_rate": 0.29139360175706336,
      "loss": 0.0,
      "num_input_tokens_seen": 2463304,
      "step": 4335
    },
    {
      "epoch": 76.14159292035399,
      "grad_norm": 8.571220860176254e-06,
      "learning_rate": 0.2913739251008544,
      "loss": 0.0,
      "num_input_tokens_seen": 2465720,
      "step": 4340
    },
    {
      "epoch": 76.23008849557522,
      "grad_norm": 2.317671533091925e-05,
      "learning_rate": 0.29135422664300964,
      "loss": 0.0,
      "num_input_tokens_seen": 2468824,
      "step": 4345
    },
    {
      "epoch": 76.31858407079646,
      "grad_norm": 1.6231848348979838e-05,
      "learning_rate": 0.29133450638656677,
      "loss": 0.0,
      "num_input_tokens_seen": 2472040,
      "step": 4350
    },
    {
      "epoch": 76.40707964601769,
      "grad_norm": 9.262331332138274e-06,
      "learning_rate": 0.2913147643345669,
      "loss": 0.0,
      "num_input_tokens_seen": 2474936,
      "step": 4355
    },
    {
      "epoch": 76.49557522123894,
      "grad_norm": 1.877340764622204e-05,
      "learning_rate": 0.29129500049005447,
      "loss": 0.0,
      "num_input_tokens_seen": 2477736,
      "step": 4360
    },
    {
      "epoch": 76.58407079646018,
      "grad_norm": 1.683931623119861e-05,
      "learning_rate": 0.2912752148560773,
      "loss": 0.0,
      "num_input_tokens_seen": 2480296,
      "step": 4365
    },
    {
      "epoch": 76.67256637168141,
      "grad_norm": 7.220799489005003e-06,
      "learning_rate": 0.2912554074356866,
      "loss": 0.0,
      "num_input_tokens_seen": 2483256,
      "step": 4370
    },
    {
      "epoch": 76.76106194690266,
      "grad_norm": 1.2846136087318882e-05,
      "learning_rate": 0.2912355782319371,
      "loss": 0.0,
      "num_input_tokens_seen": 2486360,
      "step": 4375
    },
    {
      "epoch": 76.84955752212389,
      "grad_norm": 1.734439138090238e-05,
      "learning_rate": 0.2912157272478864,
      "loss": 0.0,
      "num_input_tokens_seen": 2489352,
      "step": 4380
    },
    {
      "epoch": 76.93805309734513,
      "grad_norm": 4.010015800304245e-06,
      "learning_rate": 0.291195854486596,
      "loss": 0.0,
      "num_input_tokens_seen": 2492136,
      "step": 4385
    },
    {
      "epoch": 77.01769911504425,
      "grad_norm": 5.7586989896663e-06,
      "learning_rate": 0.2911759599511305,
      "loss": 0.0,
      "num_input_tokens_seen": 2494240,
      "step": 4390
    },
    {
      "epoch": 77.10619469026548,
      "grad_norm": 1.9729763153009117e-05,
      "learning_rate": 0.29115604364455777,
      "loss": 0.0,
      "num_input_tokens_seen": 2497360,
      "step": 4395
    },
    {
      "epoch": 77.19469026548673,
      "grad_norm": 1.7830365322879516e-05,
      "learning_rate": 0.2911361055699493,
      "loss": 0.0,
      "num_input_tokens_seen": 2500160,
      "step": 4400
    },
    {
      "epoch": 77.19469026548673,
      "eval_loss": 0.38368889689445496,
      "eval_runtime": 0.9318,
      "eval_samples_per_second": 26.83,
      "eval_steps_per_second": 13.952,
      "num_input_tokens_seen": 2500160,
      "step": 4400
    },
    {
      "epoch": 77.28318584070796,
      "grad_norm": 1.655864980421029e-05,
      "learning_rate": 0.2911161457303797,
      "loss": 0.0,
      "num_input_tokens_seen": 2503184,
      "step": 4405
    },
    {
      "epoch": 77.3716814159292,
      "grad_norm": 9.265312655770686e-06,
      "learning_rate": 0.291096164128927,
      "loss": 0.0,
      "num_input_tokens_seen": 2505808,
      "step": 4410
    },
    {
      "epoch": 77.46017699115045,
      "grad_norm": 4.439972599357134e-06,
      "learning_rate": 0.2910761607686727,
      "loss": 0.0,
      "num_input_tokens_seen": 2508128,
      "step": 4415
    },
    {
      "epoch": 77.54867256637168,
      "grad_norm": 1.818980649659352e-06,
      "learning_rate": 0.2910561356527016,
      "loss": 0.0,
      "num_input_tokens_seen": 2510896,
      "step": 4420
    },
    {
      "epoch": 77.63716814159292,
      "grad_norm": 2.7093818061985075e-05,
      "learning_rate": 0.2910360887841017,
      "loss": 0.0,
      "num_input_tokens_seen": 2513792,
      "step": 4425
    },
    {
      "epoch": 77.72566371681415,
      "grad_norm": 5.347759724827483e-06,
      "learning_rate": 0.2910160201659645,
      "loss": 0.0,
      "num_input_tokens_seen": 2516512,
      "step": 4430
    },
    {
      "epoch": 77.8141592920354,
      "grad_norm": 1.3840855899616145e-05,
      "learning_rate": 0.29099592980138494,
      "loss": 0.0,
      "num_input_tokens_seen": 2520448,
      "step": 4435
    },
    {
      "epoch": 77.90265486725664,
      "grad_norm": 1.4718340025865473e-05,
      "learning_rate": 0.29097581769346115,
      "loss": 0.0,
      "num_input_tokens_seen": 2523264,
      "step": 4440
    },
    {
      "epoch": 77.99115044247787,
      "grad_norm": 1.7104386643040925e-05,
      "learning_rate": 0.29095568384529463,
      "loss": 0.0,
      "num_input_tokens_seen": 2525888,
      "step": 4445
    },
    {
      "epoch": 78.070796460177,
      "grad_norm": 4.389239165902836e-06,
      "learning_rate": 0.2909355282599903,
      "loss": 0.0,
      "num_input_tokens_seen": 2528640,
      "step": 4450
    },
    {
      "epoch": 78.15929203539822,
      "grad_norm": 8.524692930222955e-06,
      "learning_rate": 0.29091535094065635,
      "loss": 0.0,
      "num_input_tokens_seen": 2531472,
      "step": 4455
    },
    {
      "epoch": 78.24778761061947,
      "grad_norm": 7.799997547408566e-06,
      "learning_rate": 0.2908951518904045,
      "loss": 0.0,
      "num_input_tokens_seen": 2533968,
      "step": 4460
    },
    {
      "epoch": 78.33628318584071,
      "grad_norm": 1.5775445717736147e-05,
      "learning_rate": 0.29087493111234963,
      "loss": 0.0,
      "num_input_tokens_seen": 2536464,
      "step": 4465
    },
    {
      "epoch": 78.42477876106194,
      "grad_norm": 1.965599221875891e-05,
      "learning_rate": 0.29085468860961,
      "loss": 0.0,
      "num_input_tokens_seen": 2539008,
      "step": 4470
    },
    {
      "epoch": 78.51327433628319,
      "grad_norm": 7.914882189652417e-06,
      "learning_rate": 0.2908344243853073,
      "loss": 0.0,
      "num_input_tokens_seen": 2542368,
      "step": 4475
    },
    {
      "epoch": 78.60176991150442,
      "grad_norm": 1.04861337604234e-05,
      "learning_rate": 0.2908141384425666,
      "loss": 0.0,
      "num_input_tokens_seen": 2545184,
      "step": 4480
    },
    {
      "epoch": 78.69026548672566,
      "grad_norm": 1.8185164663009346e-05,
      "learning_rate": 0.2907938307845161,
      "loss": 0.0,
      "num_input_tokens_seen": 2548224,
      "step": 4485
    },
    {
      "epoch": 78.77876106194691,
      "grad_norm": 3.674972049338976e-06,
      "learning_rate": 0.2907735014142876,
      "loss": 0.0,
      "num_input_tokens_seen": 2551072,
      "step": 4490
    },
    {
      "epoch": 78.86725663716814,
      "grad_norm": 1.638784488022793e-05,
      "learning_rate": 0.2907531503350161,
      "loss": 0.0,
      "num_input_tokens_seen": 2554016,
      "step": 4495
    },
    {
      "epoch": 78.95575221238938,
      "grad_norm": 5.37524920218857e-06,
      "learning_rate": 0.29073277754983995,
      "loss": 0.0,
      "num_input_tokens_seen": 2556960,
      "step": 4500
    },
    {
      "epoch": 79.03539823008849,
      "grad_norm": 7.64646210882347e-06,
      "learning_rate": 0.290712383061901,
      "loss": 0.0,
      "num_input_tokens_seen": 2559240,
      "step": 4505
    },
    {
      "epoch": 79.12389380530973,
      "grad_norm": 1.1027284926967695e-05,
      "learning_rate": 0.2906919668743443,
      "loss": 0.0,
      "num_input_tokens_seen": 2562056,
      "step": 4510
    },
    {
      "epoch": 79.21238938053098,
      "grad_norm": 1.8373077182332054e-05,
      "learning_rate": 0.29067152899031823,
      "loss": 0.0,
      "num_input_tokens_seen": 2564744,
      "step": 4515
    },
    {
      "epoch": 79.30088495575221,
      "grad_norm": 9.340361430076882e-06,
      "learning_rate": 0.2906510694129746,
      "loss": 0.0,
      "num_input_tokens_seen": 2568040,
      "step": 4520
    },
    {
      "epoch": 79.38938053097345,
      "grad_norm": 2.304910412931349e-05,
      "learning_rate": 0.2906305881454685,
      "loss": 0.0,
      "num_input_tokens_seen": 2570808,
      "step": 4525
    },
    {
      "epoch": 79.47787610619469,
      "grad_norm": 2.4579037926741876e-05,
      "learning_rate": 0.2906100851909585,
      "loss": 0.0,
      "num_input_tokens_seen": 2573992,
      "step": 4530
    },
    {
      "epoch": 79.56637168141593,
      "grad_norm": 2.1346629637264414e-06,
      "learning_rate": 0.29058956055260626,
      "loss": 0.0,
      "num_input_tokens_seen": 2576744,
      "step": 4535
    },
    {
      "epoch": 79.65486725663717,
      "grad_norm": 1.560763303132262e-05,
      "learning_rate": 0.2905690142335771,
      "loss": 0.0,
      "num_input_tokens_seen": 2579864,
      "step": 4540
    },
    {
      "epoch": 79.7433628318584,
      "grad_norm": 8.336866812896915e-06,
      "learning_rate": 0.29054844623703946,
      "loss": 0.0,
      "num_input_tokens_seen": 2582632,
      "step": 4545
    },
    {
      "epoch": 79.83185840707965,
      "grad_norm": 6.094447599025443e-06,
      "learning_rate": 0.2905278565661651,
      "loss": 0.0,
      "num_input_tokens_seen": 2585688,
      "step": 4550
    },
    {
      "epoch": 79.92035398230088,
      "grad_norm": 2.2186550268088467e-05,
      "learning_rate": 0.2905072452241293,
      "loss": 0.0,
      "num_input_tokens_seen": 2588504,
      "step": 4555
    },
    {
      "epoch": 80.0,
      "grad_norm": 2.4563826173107373e-06,
      "learning_rate": 0.2904866122141106,
      "loss": 0.0,
      "num_input_tokens_seen": 2590976,
      "step": 4560
    },
    {
      "epoch": 80.08849557522124,
      "grad_norm": 1.119999069487676e-05,
      "learning_rate": 0.2904659575392908,
      "loss": 0.0,
      "num_input_tokens_seen": 2593808,
      "step": 4565
    },
    {
      "epoch": 80.17699115044248,
      "grad_norm": 8.269725526588445e-07,
      "learning_rate": 0.2904452812028551,
      "loss": 0.0,
      "num_input_tokens_seen": 2596800,
      "step": 4570
    },
    {
      "epoch": 80.26548672566372,
      "grad_norm": 3.2291966363118263e-06,
      "learning_rate": 0.2904245832079922,
      "loss": 0.0,
      "num_input_tokens_seen": 2599616,
      "step": 4575
    },
    {
      "epoch": 80.35398230088495,
      "grad_norm": 3.3825422178779263e-06,
      "learning_rate": 0.29040386355789377,
      "loss": 0.0,
      "num_input_tokens_seen": 2602432,
      "step": 4580
    },
    {
      "epoch": 80.4424778761062,
      "grad_norm": 2.996683633682551e-06,
      "learning_rate": 0.29038312225575524,
      "loss": 0.0,
      "num_input_tokens_seen": 2605104,
      "step": 4585
    },
    {
      "epoch": 80.53097345132744,
      "grad_norm": 2.6377067115390673e-05,
      "learning_rate": 0.29036235930477505,
      "loss": 0.0,
      "num_input_tokens_seen": 2607984,
      "step": 4590
    },
    {
      "epoch": 80.61946902654867,
      "grad_norm": 1.1928204003197607e-05,
      "learning_rate": 0.29034157470815514,
      "loss": 0.0,
      "num_input_tokens_seen": 2610832,
      "step": 4595
    },
    {
      "epoch": 80.70796460176992,
      "grad_norm": 3.496194494800875e-06,
      "learning_rate": 0.2903207684691008,
      "loss": 0.0,
      "num_input_tokens_seen": 2614032,
      "step": 4600
    },
    {
      "epoch": 80.70796460176992,
      "eval_loss": 0.3870992958545685,
      "eval_runtime": 0.9399,
      "eval_samples_per_second": 26.598,
      "eval_steps_per_second": 13.831,
      "num_input_tokens_seen": 2614032,
      "step": 4600
    },
    {
      "epoch": 80.79646017699115,
      "grad_norm": 1.9724067897186615e-05,
      "learning_rate": 0.29029994059082054,
      "loss": 0.0,
      "num_input_tokens_seen": 2616800,
      "step": 4605
    },
    {
      "epoch": 80.88495575221239,
      "grad_norm": 1.8426884480504668e-06,
      "learning_rate": 0.2902790910765264,
      "loss": 0.0,
      "num_input_tokens_seen": 2619648,
      "step": 4610
    },
    {
      "epoch": 80.97345132743362,
      "grad_norm": 7.508423095714534e-06,
      "learning_rate": 0.29025821992943346,
      "loss": 0.0,
      "num_input_tokens_seen": 2622864,
      "step": 4615
    },
    {
      "epoch": 81.05309734513274,
      "grad_norm": 9.51443871599622e-06,
      "learning_rate": 0.29023732715276046,
      "loss": 0.0,
      "num_input_tokens_seen": 2625232,
      "step": 4620
    },
    {
      "epoch": 81.14159292035399,
      "grad_norm": 1.909748971229419e-05,
      "learning_rate": 0.2902164127497293,
      "loss": 0.0,
      "num_input_tokens_seen": 2628320,
      "step": 4625
    },
    {
      "epoch": 81.23008849557522,
      "grad_norm": 1.7563539586262777e-05,
      "learning_rate": 0.2901954767235652,
      "loss": 0.0,
      "num_input_tokens_seen": 2631056,
      "step": 4630
    },
    {
      "epoch": 81.31858407079646,
      "grad_norm": 6.830956863268511e-06,
      "learning_rate": 0.2901745190774968,
      "loss": 0.0,
      "num_input_tokens_seen": 2633792,
      "step": 4635
    },
    {
      "epoch": 81.40707964601769,
      "grad_norm": 1.0477417163201608e-05,
      "learning_rate": 0.290153539814756,
      "loss": 0.0,
      "num_input_tokens_seen": 2636288,
      "step": 4640
    },
    {
      "epoch": 81.49557522123894,
      "grad_norm": 9.02857027540449e-06,
      "learning_rate": 0.2901325389385781,
      "loss": 0.0,
      "num_input_tokens_seen": 2639536,
      "step": 4645
    },
    {
      "epoch": 81.58407079646018,
      "grad_norm": 7.994470252015162e-06,
      "learning_rate": 0.2901115164522016,
      "loss": 0.0,
      "num_input_tokens_seen": 2642256,
      "step": 4650
    },
    {
      "epoch": 81.67256637168141,
      "grad_norm": 2.8310262223385507e-06,
      "learning_rate": 0.29009047235886865,
      "loss": 0.0,
      "num_input_tokens_seen": 2645280,
      "step": 4655
    },
    {
      "epoch": 81.76106194690266,
      "grad_norm": 8.874430932337418e-06,
      "learning_rate": 0.2900694066618243,
      "loss": 0.0,
      "num_input_tokens_seen": 2648624,
      "step": 4660
    },
    {
      "epoch": 81.84955752212389,
      "grad_norm": 8.299872206407599e-06,
      "learning_rate": 0.2900483193643172,
      "loss": 0.0,
      "num_input_tokens_seen": 2651824,
      "step": 4665
    },
    {
      "epoch": 81.93805309734513,
      "grad_norm": 1.915339453262277e-05,
      "learning_rate": 0.29002721046959934,
      "loss": 0.0,
      "num_input_tokens_seen": 2654352,
      "step": 4670
    },
    {
      "epoch": 82.01769911504425,
      "grad_norm": 1.232035288012412e-06,
      "learning_rate": 0.29000607998092587,
      "loss": 0.0,
      "num_input_tokens_seen": 2656984,
      "step": 4675
    },
    {
      "epoch": 82.10619469026548,
      "grad_norm": 2.1881428438064177e-06,
      "learning_rate": 0.2899849279015555,
      "loss": 0.0,
      "num_input_tokens_seen": 2659736,
      "step": 4680
    },
    {
      "epoch": 82.19469026548673,
      "grad_norm": 6.808013040426886e-06,
      "learning_rate": 0.28996375423475007,
      "loss": 0.0,
      "num_input_tokens_seen": 2662392,
      "step": 4685
    },
    {
      "epoch": 82.28318584070796,
      "grad_norm": 5.052491815149551e-06,
      "learning_rate": 0.28994255898377486,
      "loss": 0.0,
      "num_input_tokens_seen": 2664808,
      "step": 4690
    },
    {
      "epoch": 82.3716814159292,
      "grad_norm": 2.081787897623144e-05,
      "learning_rate": 0.2899213421518984,
      "loss": 0.0,
      "num_input_tokens_seen": 2667560,
      "step": 4695
    },
    {
      "epoch": 82.46017699115045,
      "grad_norm": 1.5330432461269083e-06,
      "learning_rate": 0.2899001037423926,
      "loss": 0.0,
      "num_input_tokens_seen": 2670232,
      "step": 4700
    },
    {
      "epoch": 82.54867256637168,
      "grad_norm": 1.9790188162005506e-05,
      "learning_rate": 0.28987884375853273,
      "loss": 0.0,
      "num_input_tokens_seen": 2673512,
      "step": 4705
    },
    {
      "epoch": 82.63716814159292,
      "grad_norm": 1.2768457054335158e-05,
      "learning_rate": 0.2898575622035974,
      "loss": 0.0,
      "num_input_tokens_seen": 2676472,
      "step": 4710
    },
    {
      "epoch": 82.72566371681415,
      "grad_norm": 4.696543328464031e-06,
      "learning_rate": 0.2898362590808683,
      "loss": 0.0,
      "num_input_tokens_seen": 2679736,
      "step": 4715
    },
    {
      "epoch": 82.8141592920354,
      "grad_norm": 1.9302939108456485e-05,
      "learning_rate": 0.2898149343936308,
      "loss": 0.0,
      "num_input_tokens_seen": 2682456,
      "step": 4720
    },
    {
      "epoch": 82.90265486725664,
      "grad_norm": 1.3511328688764479e-05,
      "learning_rate": 0.2897935881451734,
      "loss": 0.0,
      "num_input_tokens_seen": 2685880,
      "step": 4725
    },
    {
      "epoch": 82.99115044247787,
      "grad_norm": 3.0456401873379946e-05,
      "learning_rate": 0.28977222033878797,
      "loss": 0.0,
      "num_input_tokens_seen": 2688776,
      "step": 4730
    },
    {
      "epoch": 83.070796460177,
      "grad_norm": 1.2985710782231763e-05,
      "learning_rate": 0.28975083097776966,
      "loss": 0.0,
      "num_input_tokens_seen": 2691032,
      "step": 4735
    },
    {
      "epoch": 83.15929203539822,
      "grad_norm": 1.567654180689715e-05,
      "learning_rate": 0.28972942006541696,
      "loss": 0.0,
      "num_input_tokens_seen": 2693656,
      "step": 4740
    },
    {
      "epoch": 83.24778761061947,
      "grad_norm": 9.73894839262357e-06,
      "learning_rate": 0.2897079876050318,
      "loss": 0.0,
      "num_input_tokens_seen": 2696248,
      "step": 4745
    },
    {
      "epoch": 83.33628318584071,
      "grad_norm": 2.284490165038733e-06,
      "learning_rate": 0.2896865335999192,
      "loss": 0.0,
      "num_input_tokens_seen": 2699672,
      "step": 4750
    },
    {
      "epoch": 83.42477876106194,
      "grad_norm": 5.151137429493247e-06,
      "learning_rate": 0.28966505805338777,
      "loss": 0.0,
      "num_input_tokens_seen": 2702344,
      "step": 4755
    },
    {
      "epoch": 83.51327433628319,
      "grad_norm": 8.255156899394933e-06,
      "learning_rate": 0.2896435609687492,
      "loss": 0.0,
      "num_input_tokens_seen": 2705304,
      "step": 4760
    },
    {
      "epoch": 83.60176991150442,
      "grad_norm": 8.191512279154267e-06,
      "learning_rate": 0.2896220423493187,
      "loss": 0.0,
      "num_input_tokens_seen": 2708328,
      "step": 4765
    },
    {
      "epoch": 83.69026548672566,
      "grad_norm": 8.54578502185177e-06,
      "learning_rate": 0.28960050219841466,
      "loss": 0.0,
      "num_input_tokens_seen": 2711352,
      "step": 4770
    },
    {
      "epoch": 83.77876106194691,
      "grad_norm": 5.412281552708009e-06,
      "learning_rate": 0.28957894051935884,
      "loss": 0.0,
      "num_input_tokens_seen": 2714312,
      "step": 4775
    },
    {
      "epoch": 83.86725663716814,
      "grad_norm": 1.6235857401625253e-05,
      "learning_rate": 0.2895573573154764,
      "loss": 0.0,
      "num_input_tokens_seen": 2717544,
      "step": 4780
    },
    {
      "epoch": 83.95575221238938,
      "grad_norm": 1.0383406333858147e-05,
      "learning_rate": 0.28953575259009556,
      "loss": 0.0,
      "num_input_tokens_seen": 2720344,
      "step": 4785
    },
    {
      "epoch": 84.03539823008849,
      "grad_norm": 1.0638526646289392e-06,
      "learning_rate": 0.2895141263465482,
      "loss": 0.0,
      "num_input_tokens_seen": 2722840,
      "step": 4790
    },
    {
      "epoch": 84.12389380530973,
      "grad_norm": 3.017914878000738e-06,
      "learning_rate": 0.28949247858816934,
      "loss": 0.0,
      "num_input_tokens_seen": 2725656,
      "step": 4795
    },
    {
      "epoch": 84.21238938053098,
      "grad_norm": 4.111861471756129e-06,
      "learning_rate": 0.2894708093182973,
      "loss": 0.0,
      "num_input_tokens_seen": 2728488,
      "step": 4800
    },
    {
      "epoch": 84.21238938053098,
      "eval_loss": 0.38885733485221863,
      "eval_runtime": 0.9304,
      "eval_samples_per_second": 26.87,
      "eval_steps_per_second": 13.973,
      "num_input_tokens_seen": 2728488,
      "step": 4800
    },
    {
      "epoch": 84.30088495575221,
      "grad_norm": 1.6306403267662972e-05,
      "learning_rate": 0.2894491185402737,
      "loss": 0.0,
      "num_input_tokens_seen": 2731112,
      "step": 4805
    },
    {
      "epoch": 84.38938053097345,
      "grad_norm": 1.1271943549218122e-05,
      "learning_rate": 0.2894274062574437,
      "loss": 0.0,
      "num_input_tokens_seen": 2734328,
      "step": 4810
    },
    {
      "epoch": 84.47787610619469,
      "grad_norm": 1.7814594457377098e-06,
      "learning_rate": 0.2894056724731554,
      "loss": 0.0,
      "num_input_tokens_seen": 2737272,
      "step": 4815
    },
    {
      "epoch": 84.56637168141593,
      "grad_norm": 4.007998086308362e-06,
      "learning_rate": 0.28938391719076056,
      "loss": 0.0,
      "num_input_tokens_seen": 2740040,
      "step": 4820
    },
    {
      "epoch": 84.65486725663717,
      "grad_norm": 5.874401267647045e-06,
      "learning_rate": 0.28936214041361413,
      "loss": 0.0,
      "num_input_tokens_seen": 2743128,
      "step": 4825
    },
    {
      "epoch": 84.7433628318584,
      "grad_norm": 1.2655121281568427e-05,
      "learning_rate": 0.2893403421450743,
      "loss": 0.0,
      "num_input_tokens_seen": 2745656,
      "step": 4830
    },
    {
      "epoch": 84.83185840707965,
      "grad_norm": 1.8253658708999865e-05,
      "learning_rate": 0.2893185223885026,
      "loss": 0.0,
      "num_input_tokens_seen": 2748120,
      "step": 4835
    },
    {
      "epoch": 84.92035398230088,
      "grad_norm": 1.1924939826712944e-06,
      "learning_rate": 0.289296681147264,
      "loss": 0.0,
      "num_input_tokens_seen": 2751528,
      "step": 4840
    },
    {
      "epoch": 85.0,
      "grad_norm": 3.4109798434656113e-06,
      "learning_rate": 0.28927481842472663,
      "loss": 0.0,
      "num_input_tokens_seen": 2754000,
      "step": 4845
    },
    {
      "epoch": 85.08849557522124,
      "grad_norm": 2.948901237687096e-05,
      "learning_rate": 0.28925293422426207,
      "loss": 0.0,
      "num_input_tokens_seen": 2756992,
      "step": 4850
    },
    {
      "epoch": 85.17699115044248,
      "grad_norm": 1.2588737945407047e-06,
      "learning_rate": 0.28923102854924504,
      "loss": 0.0,
      "num_input_tokens_seen": 2760000,
      "step": 4855
    },
    {
      "epoch": 85.26548672566372,
      "grad_norm": 8.239495400630403e-06,
      "learning_rate": 0.2892091014030537,
      "loss": 0.0,
      "num_input_tokens_seen": 2762704,
      "step": 4860
    },
    {
      "epoch": 85.35398230088495,
      "grad_norm": 1.9073662770097144e-05,
      "learning_rate": 0.2891871527890696,
      "loss": 0.0,
      "num_input_tokens_seen": 2765568,
      "step": 4865
    },
    {
      "epoch": 85.4424778761062,
      "grad_norm": 8.761867320572492e-07,
      "learning_rate": 0.2891651827106773,
      "loss": 0.0,
      "num_input_tokens_seen": 2768592,
      "step": 4870
    },
    {
      "epoch": 85.53097345132744,
      "grad_norm": 1.3471468491843552e-06,
      "learning_rate": 0.2891431911712651,
      "loss": 0.0,
      "num_input_tokens_seen": 2771504,
      "step": 4875
    },
    {
      "epoch": 85.61946902654867,
      "grad_norm": 8.556247848900966e-06,
      "learning_rate": 0.2891211781742241,
      "loss": 0.0,
      "num_input_tokens_seen": 2774592,
      "step": 4880
    },
    {
      "epoch": 85.70796460176992,
      "grad_norm": 7.468584044545423e-06,
      "learning_rate": 0.2890991437229492,
      "loss": 0.0,
      "num_input_tokens_seen": 2777552,
      "step": 4885
    },
    {
      "epoch": 85.79646017699115,
      "grad_norm": 1.0246650163026061e-05,
      "learning_rate": 0.2890770878208383,
      "loss": 0.0,
      "num_input_tokens_seen": 2780272,
      "step": 4890
    },
    {
      "epoch": 85.88495575221239,
      "grad_norm": 3.511073373374529e-06,
      "learning_rate": 0.28905501047129273,
      "loss": 0.0,
      "num_input_tokens_seen": 2783008,
      "step": 4895
    },
    {
      "epoch": 85.97345132743362,
      "grad_norm": 1.733628596412018e-05,
      "learning_rate": 0.289032911677717,
      "loss": 0.0,
      "num_input_tokens_seen": 2785936,
      "step": 4900
    },
    {
      "epoch": 86.05309734513274,
      "grad_norm": 7.542731509602163e-06,
      "learning_rate": 0.28901079144351915,
      "loss": 0.0,
      "num_input_tokens_seen": 2788192,
      "step": 4905
    },
    {
      "epoch": 86.14159292035399,
      "grad_norm": 2.0946015865774825e-05,
      "learning_rate": 0.2889886497721103,
      "loss": 0.0,
      "num_input_tokens_seen": 2791344,
      "step": 4910
    },
    {
      "epoch": 86.23008849557522,
      "grad_norm": 7.863848622946534e-06,
      "learning_rate": 0.28896648666690505,
      "loss": 0.0,
      "num_input_tokens_seen": 2793792,
      "step": 4915
    },
    {
      "epoch": 86.31858407079646,
      "grad_norm": 9.639821655582637e-06,
      "learning_rate": 0.2889443021313212,
      "loss": 0.0,
      "num_input_tokens_seen": 2796320,
      "step": 4920
    },
    {
      "epoch": 86.40707964601769,
      "grad_norm": 5.281523954181466e-06,
      "learning_rate": 0.28892209616877984,
      "loss": 0.0,
      "num_input_tokens_seen": 2799360,
      "step": 4925
    },
    {
      "epoch": 86.49557522123894,
      "grad_norm": 9.040928489412181e-06,
      "learning_rate": 0.28889986878270546,
      "loss": 0.0,
      "num_input_tokens_seen": 2801824,
      "step": 4930
    },
    {
      "epoch": 86.58407079646018,
      "grad_norm": 1.5492018974327948e-06,
      "learning_rate": 0.28887761997652583,
      "loss": 0.0,
      "num_input_tokens_seen": 2804704,
      "step": 4935
    },
    {
      "epoch": 86.67256637168141,
      "grad_norm": 1.0514958376006689e-05,
      "learning_rate": 0.2888553497536719,
      "loss": 0.0,
      "num_input_tokens_seen": 2807424,
      "step": 4940
    },
    {
      "epoch": 86.76106194690266,
      "grad_norm": 9.266422239306848e-06,
      "learning_rate": 0.2888330581175781,
      "loss": 0.0,
      "num_input_tokens_seen": 2810304,
      "step": 4945
    },
    {
      "epoch": 86.84955752212389,
      "grad_norm": 6.9604320742655545e-06,
      "learning_rate": 0.28881074507168203,
      "loss": 0.0,
      "num_input_tokens_seen": 2813328,
      "step": 4950
    },
    {
      "epoch": 86.93805309734513,
      "grad_norm": 9.27461042010691e-06,
      "learning_rate": 0.2887884106194247,
      "loss": 0.0,
      "num_input_tokens_seen": 2816432,
      "step": 4955
    },
    {
      "epoch": 87.01769911504425,
      "grad_norm": 5.5976183830352966e-06,
      "learning_rate": 0.28876605476425027,
      "loss": 0.0,
      "num_input_tokens_seen": 2819264,
      "step": 4960
    },
    {
      "epoch": 87.10619469026548,
      "grad_norm": 8.884831913746893e-06,
      "learning_rate": 0.2887436775096064,
      "loss": 0.0,
      "num_input_tokens_seen": 2821984,
      "step": 4965
    },
    {
      "epoch": 87.19469026548673,
      "grad_norm": 1.0755811672424898e-05,
      "learning_rate": 0.2887212788589439,
      "loss": 0.0,
      "num_input_tokens_seen": 2824784,
      "step": 4970
    },
    {
      "epoch": 87.28318584070796,
      "grad_norm": 1.2911949852423277e-05,
      "learning_rate": 0.2886988588157169,
      "loss": 0.0,
      "num_input_tokens_seen": 2827456,
      "step": 4975
    },
    {
      "epoch": 87.3716814159292,
      "grad_norm": 1.3602093531517312e-05,
      "learning_rate": 0.28867641738338284,
      "loss": 0.0,
      "num_input_tokens_seen": 2830592,
      "step": 4980
    },
    {
      "epoch": 87.46017699115045,
      "grad_norm": 1.392661488353042e-06,
      "learning_rate": 0.2886539545654026,
      "loss": 0.0,
      "num_input_tokens_seen": 2833328,
      "step": 4985
    },
    {
      "epoch": 87.54867256637168,
      "grad_norm": 2.1592147732008016e-06,
      "learning_rate": 0.28863147036524006,
      "loss": 0.0,
      "num_input_tokens_seen": 2836752,
      "step": 4990
    },
    {
      "epoch": 87.63716814159292,
      "grad_norm": 3.7613256154145347e-06,
      "learning_rate": 0.2886089647863626,
      "loss": 0.0,
      "num_input_tokens_seen": 2840000,
      "step": 4995
    },
    {
      "epoch": 87.72566371681415,
      "grad_norm": 3.5086122807115316e-06,
      "learning_rate": 0.288586437832241,
      "loss": 0.0,
      "num_input_tokens_seen": 2842656,
      "step": 5000
    },
    {
      "epoch": 87.72566371681415,
      "eval_loss": 0.3831348419189453,
      "eval_runtime": 0.9129,
      "eval_samples_per_second": 27.385,
      "eval_steps_per_second": 14.24,
      "num_input_tokens_seen": 2842656,
      "step": 5000
    },
    {
      "epoch": 87.8141592920354,
      "grad_norm": 9.461453373660333e-06,
      "learning_rate": 0.28856388950634904,
      "loss": 0.0,
      "num_input_tokens_seen": 2845328,
      "step": 5005
    },
    {
      "epoch": 87.90265486725664,
      "grad_norm": 9.121586117544211e-06,
      "learning_rate": 0.288541319812164,
      "loss": 0.0,
      "num_input_tokens_seen": 2848560,
      "step": 5010
    },
    {
      "epoch": 87.99115044247787,
      "grad_norm": 6.533849045808893e-06,
      "learning_rate": 0.2885187287531665,
      "loss": 0.0,
      "num_input_tokens_seen": 2851200,
      "step": 5015
    },
    {
      "epoch": 88.070796460177,
      "grad_norm": 1.781542414391879e-05,
      "learning_rate": 0.2884961163328402,
      "loss": 0.0,
      "num_input_tokens_seen": 2853392,
      "step": 5020
    },
    {
      "epoch": 88.15929203539822,
      "grad_norm": 9.960565876099281e-06,
      "learning_rate": 0.28847348255467237,
      "loss": 0.0,
      "num_input_tokens_seen": 2856016,
      "step": 5025
    },
    {
      "epoch": 88.24778761061947,
      "grad_norm": 1.1412565072532743e-05,
      "learning_rate": 0.28845082742215333,
      "loss": 0.0,
      "num_input_tokens_seen": 2859104,
      "step": 5030
    },
    {
      "epoch": 88.33628318584071,
      "grad_norm": 5.683822109858738e-06,
      "learning_rate": 0.2884281509387769,
      "loss": 0.0,
      "num_input_tokens_seen": 2862016,
      "step": 5035
    },
    {
      "epoch": 88.42477876106194,
      "grad_norm": 1.148941919382196e-06,
      "learning_rate": 0.2884054531080399,
      "loss": 0.0,
      "num_input_tokens_seen": 2864800,
      "step": 5040
    },
    {
      "epoch": 88.51327433628319,
      "grad_norm": 1.4054267012397759e-05,
      "learning_rate": 0.28838273393344277,
      "loss": 0.0,
      "num_input_tokens_seen": 2867296,
      "step": 5045
    },
    {
      "epoch": 88.60176991150442,
      "grad_norm": 1.3280301573104225e-05,
      "learning_rate": 0.288359993418489,
      "loss": 0.0,
      "num_input_tokens_seen": 2870608,
      "step": 5050
    },
    {
      "epoch": 88.69026548672566,
      "grad_norm": 1.16450300993165e-05,
      "learning_rate": 0.28833723156668556,
      "loss": 0.0,
      "num_input_tokens_seen": 2873472,
      "step": 5055
    },
    {
      "epoch": 88.77876106194691,
      "grad_norm": 3.0814367164566647e-06,
      "learning_rate": 0.2883144483815425,
      "loss": 0.0,
      "num_input_tokens_seen": 2876592,
      "step": 5060
    },
    {
      "epoch": 88.86725663716814,
      "grad_norm": 4.007888946944149e-06,
      "learning_rate": 0.28829164386657335,
      "loss": 0.0,
      "num_input_tokens_seen": 2879440,
      "step": 5065
    },
    {
      "epoch": 88.95575221238938,
      "grad_norm": 5.850240540894447e-06,
      "learning_rate": 0.28826881802529486,
      "loss": 0.0,
      "num_input_tokens_seen": 2882560,
      "step": 5070
    },
    {
      "epoch": 89.03539823008849,
      "grad_norm": 6.122223226157075e-07,
      "learning_rate": 0.28824597086122705,
      "loss": 0.0,
      "num_input_tokens_seen": 2884808,
      "step": 5075
    },
    {
      "epoch": 89.12389380530973,
      "grad_norm": 9.078589755517896e-06,
      "learning_rate": 0.28822310237789317,
      "loss": 0.0,
      "num_input_tokens_seen": 2887656,
      "step": 5080
    },
    {
      "epoch": 89.21238938053098,
      "grad_norm": 1.4294032553152647e-06,
      "learning_rate": 0.2882002125788199,
      "loss": 0.0,
      "num_input_tokens_seen": 2890552,
      "step": 5085
    },
    {
      "epoch": 89.30088495575221,
      "grad_norm": 2.5968711270252243e-06,
      "learning_rate": 0.2881773014675371,
      "loss": 0.0,
      "num_input_tokens_seen": 2893288,
      "step": 5090
    },
    {
      "epoch": 89.38938053097345,
      "grad_norm": 5.55835595150711e-06,
      "learning_rate": 0.288154369047578,
      "loss": 0.0,
      "num_input_tokens_seen": 2896008,
      "step": 5095
    },
    {
      "epoch": 89.47787610619469,
      "grad_norm": 4.663425443141023e-06,
      "learning_rate": 0.28813141532247905,
      "loss": 0.0,
      "num_input_tokens_seen": 2899096,
      "step": 5100
    },
    {
      "epoch": 89.56637168141593,
      "grad_norm": 4.8926585805020295e-06,
      "learning_rate": 0.28810844029578,
      "loss": 0.0,
      "num_input_tokens_seen": 2901544,
      "step": 5105
    },
    {
      "epoch": 89.65486725663717,
      "grad_norm": 2.1385108993854374e-05,
      "learning_rate": 0.2880854439710238,
      "loss": 0.0,
      "num_input_tokens_seen": 2904632,
      "step": 5110
    },
    {
      "epoch": 89.7433628318584,
      "grad_norm": 1.2534471352410037e-05,
      "learning_rate": 0.28806242635175694,
      "loss": 0.0,
      "num_input_tokens_seen": 2907992,
      "step": 5115
    },
    {
      "epoch": 89.83185840707965,
      "grad_norm": 4.232079390931176e-06,
      "learning_rate": 0.2880393874415289,
      "loss": 0.0,
      "num_input_tokens_seen": 2911192,
      "step": 5120
    },
    {
      "epoch": 89.92035398230088,
      "grad_norm": 1.2675844118348323e-05,
      "learning_rate": 0.2880163272438926,
      "loss": 0.0,
      "num_input_tokens_seen": 2913912,
      "step": 5125
    },
    {
      "epoch": 90.0,
      "grad_norm": 1.3723955817113165e-05,
      "learning_rate": 0.2879932457624042,
      "loss": 0.0,
      "num_input_tokens_seen": 2916440,
      "step": 5130
    },
    {
      "epoch": 90.08849557522124,
      "grad_norm": 2.251704609079752e-05,
      "learning_rate": 0.2879701430006232,
      "loss": 0.0,
      "num_input_tokens_seen": 2919000,
      "step": 5135
    },
    {
      "epoch": 90.17699115044248,
      "grad_norm": 5.5813093240431044e-06,
      "learning_rate": 0.28794701896211233,
      "loss": 0.0,
      "num_input_tokens_seen": 2922520,
      "step": 5140
    },
    {
      "epoch": 90.26548672566372,
      "grad_norm": 1.4057060070626903e-05,
      "learning_rate": 0.28792387365043753,
      "loss": 0.0,
      "num_input_tokens_seen": 2925144,
      "step": 5145
    },
    {
      "epoch": 90.35398230088495,
      "grad_norm": 1.7582716509423335e-06,
      "learning_rate": 0.28790070706916815,
      "loss": 0.0,
      "num_input_tokens_seen": 2928216,
      "step": 5150
    },
    {
      "epoch": 90.4424778761062,
      "grad_norm": 7.362992164416937e-06,
      "learning_rate": 0.2878775192218768,
      "loss": 0.0,
      "num_input_tokens_seen": 2931416,
      "step": 5155
    },
    {
      "epoch": 90.53097345132744,
      "grad_norm": 1.3449930520437192e-05,
      "learning_rate": 0.2878543101121393,
      "loss": 0.0,
      "num_input_tokens_seen": 2934360,
      "step": 5160
    },
    {
      "epoch": 90.61946902654867,
      "grad_norm": 8.30828048492549e-06,
      "learning_rate": 0.28783107974353483,
      "loss": 0.0,
      "num_input_tokens_seen": 2937096,
      "step": 5165
    },
    {
      "epoch": 90.70796460176992,
      "grad_norm": 5.301880264596548e-06,
      "learning_rate": 0.2878078281196457,
      "loss": 0.0,
      "num_input_tokens_seen": 2939720,
      "step": 5170
    },
    {
      "epoch": 90.79646017699115,
      "grad_norm": 9.05451724975137e-06,
      "learning_rate": 0.28778455524405777,
      "loss": 0.0,
      "num_input_tokens_seen": 2942648,
      "step": 5175
    },
    {
      "epoch": 90.88495575221239,
      "grad_norm": 1.5162173440330662e-05,
      "learning_rate": 0.2877612611203598,
      "loss": 0.0,
      "num_input_tokens_seen": 2945032,
      "step": 5180
    },
    {
      "epoch": 90.97345132743362,
      "grad_norm": 7.1386130002792925e-06,
      "learning_rate": 0.28773794575214423,
      "loss": 0.0,
      "num_input_tokens_seen": 2948184,
      "step": 5185
    },
    {
      "epoch": 91.05309734513274,
      "grad_norm": 1.4461415958066937e-05,
      "learning_rate": 0.28771460914300645,
      "loss": 0.0,
      "num_input_tokens_seen": 2950648,
      "step": 5190
    },
    {
      "epoch": 91.14159292035399,
      "grad_norm": 3.3484454888821347e-06,
      "learning_rate": 0.2876912512965454,
      "loss": 0.0,
      "num_input_tokens_seen": 2954088,
      "step": 5195
    },
    {
      "epoch": 91.23008849557522,
      "grad_norm": 5.230941496847663e-06,
      "learning_rate": 0.287667872216363,
      "loss": 0.0,
      "num_input_tokens_seen": 2956824,
      "step": 5200
    },
    {
      "epoch": 91.23008849557522,
      "eval_loss": 0.38440048694610596,
      "eval_runtime": 0.9397,
      "eval_samples_per_second": 26.603,
      "eval_steps_per_second": 13.834,
      "num_input_tokens_seen": 2956824,
      "step": 5200
    },
    {
      "epoch": 91.31858407079646,
      "grad_norm": 2.0295258309488418e-06,
      "learning_rate": 0.2876444719060647,
      "loss": 0.0,
      "num_input_tokens_seen": 2959496,
      "step": 5205
    },
    {
      "epoch": 91.40707964601769,
      "grad_norm": 3.1040040084917564e-06,
      "learning_rate": 0.287621050369259,
      "loss": 0.0,
      "num_input_tokens_seen": 2962744,
      "step": 5210
    },
    {
      "epoch": 91.49557522123894,
      "grad_norm": 3.946323431591736e-06,
      "learning_rate": 0.28759760760955794,
      "loss": 0.0,
      "num_input_tokens_seen": 2965576,
      "step": 5215
    },
    {
      "epoch": 91.58407079646018,
      "grad_norm": 1.8091521951646428e-06,
      "learning_rate": 0.2875741436305766,
      "loss": 0.0,
      "num_input_tokens_seen": 2968264,
      "step": 5220
    },
    {
      "epoch": 91.67256637168141,
      "grad_norm": 5.5147042985481676e-06,
      "learning_rate": 0.28755065843593347,
      "loss": 0.0,
      "num_input_tokens_seen": 2971000,
      "step": 5225
    },
    {
      "epoch": 91.76106194690266,
      "grad_norm": 6.247760211408604e-06,
      "learning_rate": 0.2875271520292502,
      "loss": 0.0,
      "num_input_tokens_seen": 2973608,
      "step": 5230
    },
    {
      "epoch": 91.84955752212389,
      "grad_norm": 1.1990829989372287e-05,
      "learning_rate": 0.28750362441415184,
      "loss": 0.0,
      "num_input_tokens_seen": 2976184,
      "step": 5235
    },
    {
      "epoch": 91.93805309734513,
      "grad_norm": 8.481175427732524e-06,
      "learning_rate": 0.28748007559426664,
      "loss": 0.0,
      "num_input_tokens_seen": 2979208,
      "step": 5240
    },
    {
      "epoch": 92.01769911504425,
      "grad_norm": 5.959813279332593e-06,
      "learning_rate": 0.2874565055732261,
      "loss": 0.0,
      "num_input_tokens_seen": 2981840,
      "step": 5245
    },
    {
      "epoch": 92.10619469026548,
      "grad_norm": 1.2250218787812628e-05,
      "learning_rate": 0.28743291435466495,
      "loss": 0.0,
      "num_input_tokens_seen": 2984448,
      "step": 5250
    },
    {
      "epoch": 92.19469026548673,
      "grad_norm": 3.890971584041836e-06,
      "learning_rate": 0.2874093019422214,
      "loss": 0.0,
      "num_input_tokens_seen": 2987360,
      "step": 5255
    },
    {
      "epoch": 92.28318584070796,
      "grad_norm": 4.438588348421035e-06,
      "learning_rate": 0.28738566833953666,
      "loss": 0.0,
      "num_input_tokens_seen": 2990048,
      "step": 5260
    },
    {
      "epoch": 92.3716814159292,
      "grad_norm": 3.5161633604730014e-06,
      "learning_rate": 0.28736201355025537,
      "loss": 0.0,
      "num_input_tokens_seen": 2992992,
      "step": 5265
    },
    {
      "epoch": 92.46017699115045,
      "grad_norm": 8.457532203465234e-07,
      "learning_rate": 0.28733833757802535,
      "loss": 0.0,
      "num_input_tokens_seen": 2995824,
      "step": 5270
    },
    {
      "epoch": 92.54867256637168,
      "grad_norm": 7.232480129459873e-06,
      "learning_rate": 0.28731464042649785,
      "loss": 0.0,
      "num_input_tokens_seen": 2998688,
      "step": 5275
    },
    {
      "epoch": 92.63716814159292,
      "grad_norm": 4.593528501573019e-06,
      "learning_rate": 0.2872909220993271,
      "loss": 0.0,
      "num_input_tokens_seen": 3001248,
      "step": 5280
    },
    {
      "epoch": 92.72566371681415,
      "grad_norm": 6.342128472169861e-06,
      "learning_rate": 0.287267182600171,
      "loss": 0.0,
      "num_input_tokens_seen": 3003856,
      "step": 5285
    },
    {
      "epoch": 92.8141592920354,
      "grad_norm": 9.76822502707364e-06,
      "learning_rate": 0.2872434219326902,
      "loss": 0.0,
      "num_input_tokens_seen": 3007216,
      "step": 5290
    },
    {
      "epoch": 92.90265486725664,
      "grad_norm": 7.333170742640505e-06,
      "learning_rate": 0.28721964010054907,
      "loss": 0.0,
      "num_input_tokens_seen": 3010432,
      "step": 5295
    },
    {
      "epoch": 92.99115044247787,
      "grad_norm": 7.978036592248827e-06,
      "learning_rate": 0.28719583710741503,
      "loss": 0.0,
      "num_input_tokens_seen": 3013056,
      "step": 5300
    },
    {
      "epoch": 93.070796460177,
      "grad_norm": 7.157663731049979e-06,
      "learning_rate": 0.28717201295695877,
      "loss": 0.0,
      "num_input_tokens_seen": 3015736,
      "step": 5305
    },
    {
      "epoch": 93.15929203539822,
      "grad_norm": 1.272394547413569e-05,
      "learning_rate": 0.28714816765285434,
      "loss": 0.0,
      "num_input_tokens_seen": 3018552,
      "step": 5310
    },
    {
      "epoch": 93.24778761061947,
      "grad_norm": 2.4609657884866465e-06,
      "learning_rate": 0.28712430119877896,
      "loss": 0.0,
      "num_input_tokens_seen": 3021416,
      "step": 5315
    },
    {
      "epoch": 93.33628318584071,
      "grad_norm": 1.5214490304060746e-05,
      "learning_rate": 0.28710041359841304,
      "loss": 0.0,
      "num_input_tokens_seen": 3024168,
      "step": 5320
    },
    {
      "epoch": 93.42477876106194,
      "grad_norm": 5.6995741033460945e-06,
      "learning_rate": 0.28707650485544056,
      "loss": 0.0,
      "num_input_tokens_seen": 3026856,
      "step": 5325
    },
    {
      "epoch": 93.51327433628319,
      "grad_norm": 1.3058319382253103e-05,
      "learning_rate": 0.28705257497354836,
      "loss": 0.0,
      "num_input_tokens_seen": 3029864,
      "step": 5330
    },
    {
      "epoch": 93.60176991150442,
      "grad_norm": 2.4836615466483636e-06,
      "learning_rate": 0.28702862395642675,
      "loss": 0.0,
      "num_input_tokens_seen": 3032888,
      "step": 5335
    },
    {
      "epoch": 93.69026548672566,
      "grad_norm": 9.12006180442404e-06,
      "learning_rate": 0.28700465180776935,
      "loss": 0.0,
      "num_input_tokens_seen": 3035608,
      "step": 5340
    },
    {
      "epoch": 93.77876106194691,
      "grad_norm": 6.13910697211395e-06,
      "learning_rate": 0.2869806585312729,
      "loss": 0.0,
      "num_input_tokens_seen": 3038536,
      "step": 5345
    },
    {
      "epoch": 93.86725663716814,
      "grad_norm": 9.552423762215767e-06,
      "learning_rate": 0.28695664413063754,
      "loss": 0.0,
      "num_input_tokens_seen": 3040920,
      "step": 5350
    },
    {
      "epoch": 93.95575221238938,
      "grad_norm": 4.05720811613719e-06,
      "learning_rate": 0.28693260860956654,
      "loss": 0.0,
      "num_input_tokens_seen": 3043688,
      "step": 5355
    },
    {
      "epoch": 94.03539823008849,
      "grad_norm": 2.4857777134457137e-06,
      "learning_rate": 0.2869085519717665,
      "loss": 0.0,
      "num_input_tokens_seen": 3046240,
      "step": 5360
    },
    {
      "epoch": 94.12389380530973,
      "grad_norm": 8.040336183512409e-07,
      "learning_rate": 0.28688447422094726,
      "loss": 0.0,
      "num_input_tokens_seen": 3049072,
      "step": 5365
    },
    {
      "epoch": 94.21238938053098,
      "grad_norm": 1.3127288184477948e-05,
      "learning_rate": 0.2868603753608219,
      "loss": 0.0,
      "num_input_tokens_seen": 3052000,
      "step": 5370
    },
    {
      "epoch": 94.30088495575221,
      "grad_norm": 8.754888767725788e-06,
      "learning_rate": 0.28683625539510665,
      "loss": 0.0,
      "num_input_tokens_seen": 3054752,
      "step": 5375
    },
    {
      "epoch": 94.38938053097345,
      "grad_norm": 8.120430720737204e-06,
      "learning_rate": 0.28681211432752135,
      "loss": 0.0,
      "num_input_tokens_seen": 3057792,
      "step": 5380
    },
    {
      "epoch": 94.47787610619469,
      "grad_norm": 1.510795937065268e-05,
      "learning_rate": 0.2867879521617887,
      "loss": 0.0,
      "num_input_tokens_seen": 3060912,
      "step": 5385
    },
    {
      "epoch": 94.56637168141593,
      "grad_norm": 6.285953986662207e-06,
      "learning_rate": 0.28676376890163485,
      "loss": 0.0,
      "num_input_tokens_seen": 3064096,
      "step": 5390
    },
    {
      "epoch": 94.65486725663717,
      "grad_norm": 6.538160050695296e-06,
      "learning_rate": 0.2867395645507891,
      "loss": 0.0,
      "num_input_tokens_seen": 3066896,
      "step": 5395
    },
    {
      "epoch": 94.7433628318584,
      "grad_norm": 1.5279762010322884e-05,
      "learning_rate": 0.2867153391129842,
      "loss": 0.0,
      "num_input_tokens_seen": 3069840,
      "step": 5400
    },
    {
      "epoch": 94.7433628318584,
      "eval_loss": 0.3785821497440338,
      "eval_runtime": 0.9457,
      "eval_samples_per_second": 26.436,
      "eval_steps_per_second": 13.747,
      "num_input_tokens_seen": 3069840,
      "step": 5400
    },
    {
      "epoch": 94.83185840707965,
      "grad_norm": 6.3126603890850674e-06,
      "learning_rate": 0.28669109259195585,
      "loss": 0.0,
      "num_input_tokens_seen": 3072432,
      "step": 5405
    },
    {
      "epoch": 94.92035398230088,
      "grad_norm": 1.2755456737068016e-05,
      "learning_rate": 0.2866668249914433,
      "loss": 0.0,
      "num_input_tokens_seen": 3075152,
      "step": 5410
    },
    {
      "epoch": 95.0,
      "grad_norm": 2.0567915726132924e-06,
      "learning_rate": 0.2866425363151889,
      "loss": 0.0,
      "num_input_tokens_seen": 3077616,
      "step": 5415
    },
    {
      "epoch": 95.08849557522124,
      "grad_norm": 3.7781182982143946e-06,
      "learning_rate": 0.2866182265669382,
      "loss": 0.0,
      "num_input_tokens_seen": 3080304,
      "step": 5420
    },
    {
      "epoch": 95.17699115044248,
      "grad_norm": 7.118732355593238e-06,
      "learning_rate": 0.28659389575044014,
      "loss": 0.0,
      "num_input_tokens_seen": 3083168,
      "step": 5425
    },
    {
      "epoch": 95.26548672566372,
      "grad_norm": 3.226055468985578e-06,
      "learning_rate": 0.28656954386944683,
      "loss": 0.0,
      "num_input_tokens_seen": 3086608,
      "step": 5430
    },
    {
      "epoch": 95.35398230088495,
      "grad_norm": 8.298664397443645e-06,
      "learning_rate": 0.28654517092771353,
      "loss": 0.0,
      "num_input_tokens_seen": 3089840,
      "step": 5435
    },
    {
      "epoch": 95.4424778761062,
      "grad_norm": 8.834174877847545e-06,
      "learning_rate": 0.286520776928999,
      "loss": 0.0,
      "num_input_tokens_seen": 3092464,
      "step": 5440
    },
    {
      "epoch": 95.53097345132744,
      "grad_norm": 9.578356184647419e-06,
      "learning_rate": 0.286496361877065,
      "loss": 0.0,
      "num_input_tokens_seen": 3095376,
      "step": 5445
    },
    {
      "epoch": 95.61946902654867,
      "grad_norm": 8.355544196092524e-06,
      "learning_rate": 0.28647192577567676,
      "loss": 0.0,
      "num_input_tokens_seen": 3098352,
      "step": 5450
    },
    {
      "epoch": 95.70796460176992,
      "grad_norm": 3.535879386618035e-06,
      "learning_rate": 0.28644746862860254,
      "loss": 0.0,
      "num_input_tokens_seen": 3100992,
      "step": 5455
    },
    {
      "epoch": 95.79646017699115,
      "grad_norm": 8.77792956543999e-07,
      "learning_rate": 0.2864229904396139,
      "loss": 0.0,
      "num_input_tokens_seen": 3104016,
      "step": 5460
    },
    {
      "epoch": 95.88495575221239,
      "grad_norm": 6.0454935919551644e-06,
      "learning_rate": 0.28639849121248573,
      "loss": 0.0,
      "num_input_tokens_seen": 3107088,
      "step": 5465
    },
    {
      "epoch": 95.97345132743362,
      "grad_norm": 4.778089987667045e-06,
      "learning_rate": 0.28637397095099615,
      "loss": 0.0,
      "num_input_tokens_seen": 3109696,
      "step": 5470
    },
    {
      "epoch": 96.05309734513274,
      "grad_norm": 8.965533197624609e-06,
      "learning_rate": 0.28634942965892646,
      "loss": 0.0,
      "num_input_tokens_seen": 3111880,
      "step": 5475
    },
    {
      "epoch": 96.14159292035399,
      "grad_norm": 3.2200096029555425e-06,
      "learning_rate": 0.28632486734006124,
      "loss": 0.0,
      "num_input_tokens_seen": 3114696,
      "step": 5480
    },
    {
      "epoch": 96.23008849557522,
      "grad_norm": 8.183047611964867e-06,
      "learning_rate": 0.28630028399818835,
      "loss": 0.0,
      "num_input_tokens_seen": 3118040,
      "step": 5485
    },
    {
      "epoch": 96.31858407079646,
      "grad_norm": 5.1480137699400075e-06,
      "learning_rate": 0.2862756796370987,
      "loss": 0.0,
      "num_input_tokens_seen": 3120856,
      "step": 5490
    },
    {
      "epoch": 96.40707964601769,
      "grad_norm": 1.558330222906079e-05,
      "learning_rate": 0.2862510542605868,
      "loss": 0.0,
      "num_input_tokens_seen": 3123560,
      "step": 5495
    },
    {
      "epoch": 96.49557522123894,
      "grad_norm": 4.589311402014573e-07,
      "learning_rate": 0.2862264078724501,
      "loss": 0.0,
      "num_input_tokens_seen": 3126200,
      "step": 5500
    },
    {
      "epoch": 96.58407079646018,
      "grad_norm": 3.7958411667204928e-06,
      "learning_rate": 0.28620174047648933,
      "loss": 0.0,
      "num_input_tokens_seen": 3129336,
      "step": 5505
    },
    {
      "epoch": 96.67256637168141,
      "grad_norm": 3.7388774671853753e-06,
      "learning_rate": 0.2861770520765086,
      "loss": 0.0,
      "num_input_tokens_seen": 3132296,
      "step": 5510
    },
    {
      "epoch": 96.76106194690266,
      "grad_norm": 2.616256779219839e-06,
      "learning_rate": 0.2861523426763151,
      "loss": 0.0,
      "num_input_tokens_seen": 3135160,
      "step": 5515
    },
    {
      "epoch": 96.84955752212389,
      "grad_norm": 8.177626114047598e-06,
      "learning_rate": 0.2861276122797194,
      "loss": 0.0,
      "num_input_tokens_seen": 3137912,
      "step": 5520
    },
    {
      "epoch": 96.93805309734513,
      "grad_norm": 2.7976798264717218e-06,
      "learning_rate": 0.28610286089053516,
      "loss": 0.0,
      "num_input_tokens_seen": 3140872,
      "step": 5525
    },
    {
      "epoch": 97.01769911504425,
      "grad_norm": 8.705530376573734e-07,
      "learning_rate": 0.28607808851257943,
      "loss": 0.0,
      "num_input_tokens_seen": 3143568,
      "step": 5530
    },
    {
      "epoch": 97.10619469026548,
      "grad_norm": 7.264301530085504e-06,
      "learning_rate": 0.28605329514967237,
      "loss": 0.0,
      "num_input_tokens_seen": 3146368,
      "step": 5535
    },
    {
      "epoch": 97.19469026548673,
      "grad_norm": 7.986803211679216e-06,
      "learning_rate": 0.2860284808056374,
      "loss": 0.0,
      "num_input_tokens_seen": 3149040,
      "step": 5540
    },
    {
      "epoch": 97.28318584070796,
      "grad_norm": 3.0409169085032772e-06,
      "learning_rate": 0.28600364548430135,
      "loss": 0.0,
      "num_input_tokens_seen": 3152224,
      "step": 5545
    },
    {
      "epoch": 97.3716814159292,
      "grad_norm": 4.981218808097765e-06,
      "learning_rate": 0.28597878918949393,
      "loss": 0.0,
      "num_input_tokens_seen": 3155216,
      "step": 5550
    },
    {
      "epoch": 97.46017699115045,
      "grad_norm": 5.273623173707165e-06,
      "learning_rate": 0.2859539119250485,
      "loss": 0.0,
      "num_input_tokens_seen": 3157984,
      "step": 5555
    },
    {
      "epoch": 97.54867256637168,
      "grad_norm": 2.834217184499721e-06,
      "learning_rate": 0.2859290136948013,
      "loss": 0.0,
      "num_input_tokens_seen": 3160928,
      "step": 5560
    },
    {
      "epoch": 97.63716814159292,
      "grad_norm": 2.9708717192988843e-06,
      "learning_rate": 0.28590409450259197,
      "loss": 0.0,
      "num_input_tokens_seen": 3163920,
      "step": 5565
    },
    {
      "epoch": 97.72566371681415,
      "grad_norm": 3.281513272668235e-06,
      "learning_rate": 0.28587915435226346,
      "loss": 0.0,
      "num_input_tokens_seen": 3166416,
      "step": 5570
    },
    {
      "epoch": 97.8141592920354,
      "grad_norm": 1.58724797074683e-05,
      "learning_rate": 0.2858541932476617,
      "loss": 0.0,
      "num_input_tokens_seen": 3169456,
      "step": 5575
    },
    {
      "epoch": 97.90265486725664,
      "grad_norm": 5.964952833892312e-06,
      "learning_rate": 0.2858292111926361,
      "loss": 0.0,
      "num_input_tokens_seen": 3172208,
      "step": 5580
    },
    {
      "epoch": 97.99115044247787,
      "grad_norm": 1.2183214494143613e-05,
      "learning_rate": 0.28580420819103924,
      "loss": 0.0,
      "num_input_tokens_seen": 3175584,
      "step": 5585
    },
    {
      "epoch": 98.070796460177,
      "grad_norm": 4.998957138013793e-06,
      "learning_rate": 0.2857791842467269,
      "loss": 0.0,
      "num_input_tokens_seen": 3178096,
      "step": 5590
    },
    {
      "epoch": 98.15929203539822,
      "grad_norm": 3.0503570087603293e-06,
      "learning_rate": 0.2857541393635579,
      "loss": 0.0,
      "num_input_tokens_seen": 3181152,
      "step": 5595
    },
    {
      "epoch": 98.24778761061947,
      "grad_norm": 7.21739161235746e-06,
      "learning_rate": 0.2857290735453948,
      "loss": 0.0,
      "num_input_tokens_seen": 3183600,
      "step": 5600
    },
    {
      "epoch": 98.24778761061947,
      "eval_loss": 0.39556920528411865,
      "eval_runtime": 0.9408,
      "eval_samples_per_second": 26.572,
      "eval_steps_per_second": 13.817,
      "num_input_tokens_seen": 3183600,
      "step": 5600
    },
    {
      "epoch": 98.33628318584071,
      "grad_norm": 4.905630248686066e-06,
      "learning_rate": 0.28570398679610276,
      "loss": 0.0,
      "num_input_tokens_seen": 3186528,
      "step": 5605
    },
    {
      "epoch": 98.42477876106194,
      "grad_norm": 3.255623141740216e-06,
      "learning_rate": 0.2856788791195506,
      "loss": 0.0,
      "num_input_tokens_seen": 3189840,
      "step": 5610
    },
    {
      "epoch": 98.51327433628319,
      "grad_norm": 3.229627964174142e-06,
      "learning_rate": 0.28565375051961023,
      "loss": 0.0,
      "num_input_tokens_seen": 3192512,
      "step": 5615
    },
    {
      "epoch": 98.60176991150442,
      "grad_norm": 1.2250823147041956e-06,
      "learning_rate": 0.28562860100015686,
      "loss": 0.0,
      "num_input_tokens_seen": 3195648,
      "step": 5620
    },
    {
      "epoch": 98.69026548672566,
      "grad_norm": 1.0644415851857048e-05,
      "learning_rate": 0.2856034305650687,
      "loss": 0.0,
      "num_input_tokens_seen": 3198160,
      "step": 5625
    },
    {
      "epoch": 98.77876106194691,
      "grad_norm": 7.3345972850802355e-06,
      "learning_rate": 0.28557823921822756,
      "loss": 0.0,
      "num_input_tokens_seen": 3201184,
      "step": 5630
    },
    {
      "epoch": 98.86725663716814,
      "grad_norm": 1.0097201084136032e-05,
      "learning_rate": 0.2855530269635181,
      "loss": 0.0,
      "num_input_tokens_seen": 3204016,
      "step": 5635
    },
    {
      "epoch": 98.95575221238938,
      "grad_norm": 1.746657108014915e-05,
      "learning_rate": 0.2855277938048284,
      "loss": 0.0,
      "num_input_tokens_seen": 3206624,
      "step": 5640
    },
    {
      "epoch": 99.03539823008849,
      "grad_norm": 2.4694882085896097e-06,
      "learning_rate": 0.2855025397460498,
      "loss": 0.0,
      "num_input_tokens_seen": 3208856,
      "step": 5645
    },
    {
      "epoch": 99.12389380530973,
      "grad_norm": 1.423750745743746e-05,
      "learning_rate": 0.28547726479107666,
      "loss": 0.0,
      "num_input_tokens_seen": 3211992,
      "step": 5650
    },
    {
      "epoch": 99.21238938053098,
      "grad_norm": 2.8929041491210228e-06,
      "learning_rate": 0.2854519689438068,
      "loss": 0.0,
      "num_input_tokens_seen": 3215064,
      "step": 5655
    },
    {
      "epoch": 99.30088495575221,
      "grad_norm": 2.9804689347656677e-06,
      "learning_rate": 0.2854266522081412,
      "loss": 0.0,
      "num_input_tokens_seen": 3217736,
      "step": 5660
    },
    {
      "epoch": 99.38938053097345,
      "grad_norm": 5.857941232534358e-06,
      "learning_rate": 0.28540131458798385,
      "loss": 0.0,
      "num_input_tokens_seen": 3220600,
      "step": 5665
    },
    {
      "epoch": 99.47787610619469,
      "grad_norm": 5.448228876048233e-06,
      "learning_rate": 0.28537595608724226,
      "loss": 0.0,
      "num_input_tokens_seen": 3223592,
      "step": 5670
    },
    {
      "epoch": 99.56637168141593,
      "grad_norm": 4.407000233186409e-06,
      "learning_rate": 0.28535057670982705,
      "loss": 0.0,
      "num_input_tokens_seen": 3226248,
      "step": 5675
    },
    {
      "epoch": 99.65486725663717,
      "grad_norm": 3.2127420581673505e-06,
      "learning_rate": 0.285325176459652,
      "loss": 0.0,
      "num_input_tokens_seen": 3229432,
      "step": 5680
    },
    {
      "epoch": 99.7433628318584,
      "grad_norm": 9.831165698415134e-06,
      "learning_rate": 0.28529975534063406,
      "loss": 0.0,
      "num_input_tokens_seen": 3231896,
      "step": 5685
    },
    {
      "epoch": 99.83185840707965,
      "grad_norm": 1.8643887642610935e-06,
      "learning_rate": 0.2852743133566936,
      "loss": 0.0,
      "num_input_tokens_seen": 3235032,
      "step": 5690
    },
    {
      "epoch": 99.92035398230088,
      "grad_norm": 4.8655547288944945e-06,
      "learning_rate": 0.2852488505117541,
      "loss": 0.0,
      "num_input_tokens_seen": 3237752,
      "step": 5695
    },
    {
      "epoch": 100.0,
      "grad_norm": 5.355445864552166e-07,
      "learning_rate": 0.28522336680974214,
      "loss": 0.0,
      "num_input_tokens_seen": 3239888,
      "step": 5700
    },
    {
      "epoch": 100.08849557522124,
      "grad_norm": 2.8886913696624106e-06,
      "learning_rate": 0.2851978622545877,
      "loss": 0.0,
      "num_input_tokens_seen": 3242672,
      "step": 5705
    },
    {
      "epoch": 100.17699115044248,
      "grad_norm": 4.4802509364672005e-06,
      "learning_rate": 0.285172336850224,
      "loss": 0.0,
      "num_input_tokens_seen": 3245408,
      "step": 5710
    },
    {
      "epoch": 100.26548672566372,
      "grad_norm": 3.3934311431949027e-06,
      "learning_rate": 0.2851467906005871,
      "loss": 0.0,
      "num_input_tokens_seen": 3248592,
      "step": 5715
    },
    {
      "epoch": 100.35398230088495,
      "grad_norm": 8.108844667731319e-06,
      "learning_rate": 0.28512122350961683,
      "loss": 0.0,
      "num_input_tokens_seen": 3251072,
      "step": 5720
    },
    {
      "epoch": 100.4424778761062,
      "grad_norm": 1.1402268000892946e-06,
      "learning_rate": 0.2850956355812559,
      "loss": 0.0,
      "num_input_tokens_seen": 3253872,
      "step": 5725
    },
    {
      "epoch": 100.53097345132744,
      "grad_norm": 2.7269518341199728e-06,
      "learning_rate": 0.28507002681945015,
      "loss": 0.0,
      "num_input_tokens_seen": 3256752,
      "step": 5730
    },
    {
      "epoch": 100.61946902654867,
      "grad_norm": 1.6866455325725838e-06,
      "learning_rate": 0.28504439722814895,
      "loss": 0.0,
      "num_input_tokens_seen": 3259584,
      "step": 5735
    },
    {
      "epoch": 100.70796460176992,
      "grad_norm": 4.127869033254683e-06,
      "learning_rate": 0.28501874681130457,
      "loss": 0.0,
      "num_input_tokens_seen": 3263120,
      "step": 5740
    },
    {
      "epoch": 100.79646017699115,
      "grad_norm": 1.3445212061924394e-05,
      "learning_rate": 0.2849930755728727,
      "loss": 0.0,
      "num_input_tokens_seen": 3266032,
      "step": 5745
    },
    {
      "epoch": 100.88495575221239,
      "grad_norm": 1.0505062846277724e-06,
      "learning_rate": 0.28496738351681217,
      "loss": 0.0,
      "num_input_tokens_seen": 3268736,
      "step": 5750
    },
    {
      "epoch": 100.97345132743362,
      "grad_norm": 1.0929511518043e-05,
      "learning_rate": 0.284941670647085,
      "loss": 0.0,
      "num_input_tokens_seen": 3272016,
      "step": 5755
    },
    {
      "epoch": 101.05309734513274,
      "grad_norm": 1.240762344423274e-06,
      "learning_rate": 0.2849159369676563,
      "loss": 0.0,
      "num_input_tokens_seen": 3274392,
      "step": 5760
    },
    {
      "epoch": 101.14159292035399,
      "grad_norm": 3.9250580812222324e-06,
      "learning_rate": 0.2848901824824948,
      "loss": 0.0,
      "num_input_tokens_seen": 3276952,
      "step": 5765
    },
    {
      "epoch": 101.23008849557522,
      "grad_norm": 3.7109812183189206e-06,
      "learning_rate": 0.284864407195572,
      "loss": 0.0,
      "num_input_tokens_seen": 3280104,
      "step": 5770
    },
    {
      "epoch": 101.31858407079646,
      "grad_norm": 2.959833636850817e-06,
      "learning_rate": 0.28483861111086284,
      "loss": 0.0,
      "num_input_tokens_seen": 3282840,
      "step": 5775
    },
    {
      "epoch": 101.40707964601769,
      "grad_norm": 2.8342963105387753e-06,
      "learning_rate": 0.2848127942323453,
      "loss": 0.0,
      "num_input_tokens_seen": 3285672,
      "step": 5780
    },
    {
      "epoch": 101.49557522123894,
      "grad_norm": 3.219944801458041e-06,
      "learning_rate": 0.2847869565640007,
      "loss": 0.0,
      "num_input_tokens_seen": 3288872,
      "step": 5785
    },
    {
      "epoch": 101.58407079646018,
      "grad_norm": 2.8587567157956073e-06,
      "learning_rate": 0.2847610981098136,
      "loss": 0.0,
      "num_input_tokens_seen": 3292264,
      "step": 5790
    },
    {
      "epoch": 101.67256637168141,
      "grad_norm": 4.762504886457464e-06,
      "learning_rate": 0.2847352188737716,
      "loss": 0.0,
      "num_input_tokens_seen": 3294808,
      "step": 5795
    },
    {
      "epoch": 101.76106194690266,
      "grad_norm": 1.465115747123491e-05,
      "learning_rate": 0.2847093188598658,
      "loss": 0.0,
      "num_input_tokens_seen": 3297896,
      "step": 5800
    },
    {
      "epoch": 101.76106194690266,
      "eval_loss": 0.38708847761154175,
      "eval_runtime": 0.9179,
      "eval_samples_per_second": 27.235,
      "eval_steps_per_second": 14.162,
      "num_input_tokens_seen": 3297896,
      "step": 5800
    },
    {
      "epoch": 101.84955752212389,
      "grad_norm": 2.9225434445834253e-06,
      "learning_rate": 0.28468339807209003,
      "loss": 0.0,
      "num_input_tokens_seen": 3300648,
      "step": 5805
    },
    {
      "epoch": 101.93805309734513,
      "grad_norm": 7.71751638239948e-06,
      "learning_rate": 0.2846574565144418,
      "loss": 0.0,
      "num_input_tokens_seen": 3303400,
      "step": 5810
    },
    {
      "epoch": 102.01769911504425,
      "grad_norm": 1.3598828445537947e-06,
      "learning_rate": 0.28463149419092154,
      "loss": 0.0,
      "num_input_tokens_seen": 3305792,
      "step": 5815
    },
    {
      "epoch": 102.10619469026548,
      "grad_norm": 1.2941313798364718e-06,
      "learning_rate": 0.284605511105533,
      "loss": 0.0,
      "num_input_tokens_seen": 3308880,
      "step": 5820
    },
    {
      "epoch": 102.19469026548673,
      "grad_norm": 1.2958608976987307e-06,
      "learning_rate": 0.28457950726228315,
      "loss": 0.0,
      "num_input_tokens_seen": 3312368,
      "step": 5825
    },
    {
      "epoch": 102.28318584070796,
      "grad_norm": 7.3799733399937395e-06,
      "learning_rate": 0.28455348266518193,
      "loss": 0.0,
      "num_input_tokens_seen": 3315072,
      "step": 5830
    },
    {
      "epoch": 102.3716814159292,
      "grad_norm": 6.492371539934538e-06,
      "learning_rate": 0.28452743731824287,
      "loss": 0.0,
      "num_input_tokens_seen": 3317728,
      "step": 5835
    },
    {
      "epoch": 102.46017699115045,
      "grad_norm": 1.2733941730402876e-05,
      "learning_rate": 0.28450137122548236,
      "loss": 0.0,
      "num_input_tokens_seen": 3320592,
      "step": 5840
    },
    {
      "epoch": 102.54867256637168,
      "grad_norm": 1.0832934549398487e-06,
      "learning_rate": 0.2844752843909201,
      "loss": 0.0,
      "num_input_tokens_seen": 3322976,
      "step": 5845
    },
    {
      "epoch": 102.63716814159292,
      "grad_norm": 9.709224286780227e-06,
      "learning_rate": 0.28444917681857923,
      "loss": 0.0,
      "num_input_tokens_seen": 3325872,
      "step": 5850
    },
    {
      "epoch": 102.72566371681415,
      "grad_norm": 7.281668331415858e-06,
      "learning_rate": 0.28442304851248557,
      "loss": 0.0,
      "num_input_tokens_seen": 3328432,
      "step": 5855
    },
    {
      "epoch": 102.8141592920354,
      "grad_norm": 2.1977873529976932e-06,
      "learning_rate": 0.2843968994766686,
      "loss": 0.0,
      "num_input_tokens_seen": 3331600,
      "step": 5860
    },
    {
      "epoch": 102.90265486725664,
      "grad_norm": 8.34025013318751e-06,
      "learning_rate": 0.28437072971516075,
      "loss": 0.0,
      "num_input_tokens_seen": 3335040,
      "step": 5865
    },
    {
      "epoch": 102.99115044247787,
      "grad_norm": 1.7243605725525413e-06,
      "learning_rate": 0.2843445392319979,
      "loss": 0.0,
      "num_input_tokens_seen": 3337920,
      "step": 5870
    },
    {
      "epoch": 103.070796460177,
      "grad_norm": 4.696688336025545e-07,
      "learning_rate": 0.28431832803121865,
      "loss": 0.0,
      "num_input_tokens_seen": 3340552,
      "step": 5875
    },
    {
      "epoch": 103.15929203539822,
      "grad_norm": 7.644524885108694e-06,
      "learning_rate": 0.28429209611686534,
      "loss": 0.0,
      "num_input_tokens_seen": 3342904,
      "step": 5880
    },
    {
      "epoch": 103.24778761061947,
      "grad_norm": 2.1358916910685366e-06,
      "learning_rate": 0.28426584349298323,
      "loss": 0.0,
      "num_input_tokens_seen": 3345480,
      "step": 5885
    },
    {
      "epoch": 103.33628318584071,
      "grad_norm": 5.622121534543112e-06,
      "learning_rate": 0.2842395701636207,
      "loss": 0.0,
      "num_input_tokens_seen": 3348456,
      "step": 5890
    },
    {
      "epoch": 103.42477876106194,
      "grad_norm": 1.729793780214095e-06,
      "learning_rate": 0.28421327613282954,
      "loss": 0.0,
      "num_input_tokens_seen": 3351000,
      "step": 5895
    },
    {
      "epoch": 103.51327433628319,
      "grad_norm": 3.069641707043047e-06,
      "learning_rate": 0.28418696140466454,
      "loss": 0.0,
      "num_input_tokens_seen": 3353624,
      "step": 5900
    },
    {
      "epoch": 103.60176991150442,
      "grad_norm": 5.894824425922707e-06,
      "learning_rate": 0.2841606259831838,
      "loss": 0.0,
      "num_input_tokens_seen": 3356504,
      "step": 5905
    },
    {
      "epoch": 103.69026548672566,
      "grad_norm": 2.033227247011382e-06,
      "learning_rate": 0.2841342698724486,
      "loss": 0.0,
      "num_input_tokens_seen": 3359080,
      "step": 5910
    },
    {
      "epoch": 103.77876106194691,
      "grad_norm": 1.504096871940419e-05,
      "learning_rate": 0.28410789307652334,
      "loss": 0.0,
      "num_input_tokens_seen": 3362216,
      "step": 5915
    },
    {
      "epoch": 103.86725663716814,
      "grad_norm": 5.029681233281735e-06,
      "learning_rate": 0.2840814955994756,
      "loss": 0.0,
      "num_input_tokens_seen": 3365960,
      "step": 5920
    },
    {
      "epoch": 103.95575221238938,
      "grad_norm": 8.646763490105513e-06,
      "learning_rate": 0.2840550774453763,
      "loss": 0.0,
      "num_input_tokens_seen": 3369192,
      "step": 5925
    },
    {
      "epoch": 104.03539823008849,
      "grad_norm": 5.243864052317804e-06,
      "learning_rate": 0.28402863861829947,
      "loss": 0.0,
      "num_input_tokens_seen": 3371848,
      "step": 5930
    },
    {
      "epoch": 104.12389380530973,
      "grad_norm": 3.3659880500636064e-06,
      "learning_rate": 0.2840021791223222,
      "loss": 0.0,
      "num_input_tokens_seen": 3374792,
      "step": 5935
    },
    {
      "epoch": 104.21238938053098,
      "grad_norm": 1.0396927791589405e-05,
      "learning_rate": 0.2839756989615249,
      "loss": 0.0,
      "num_input_tokens_seen": 3377816,
      "step": 5940
    },
    {
      "epoch": 104.30088495575221,
      "grad_norm": 9.120252002503548e-07,
      "learning_rate": 0.28394919813999125,
      "loss": 0.0,
      "num_input_tokens_seen": 3380808,
      "step": 5945
    },
    {
      "epoch": 104.38938053097345,
      "grad_norm": 2.27427517529577e-06,
      "learning_rate": 0.28392267666180787,
      "loss": 0.0,
      "num_input_tokens_seen": 3384008,
      "step": 5950
    },
    {
      "epoch": 104.47787610619469,
      "grad_norm": 4.938204256177414e-06,
      "learning_rate": 0.2838961345310648,
      "loss": 0.0,
      "num_input_tokens_seen": 3386472,
      "step": 5955
    },
    {
      "epoch": 104.56637168141593,
      "grad_norm": 4.593832727550762e-06,
      "learning_rate": 0.2838695717518552,
      "loss": 0.0,
      "num_input_tokens_seen": 3389272,
      "step": 5960
    },
    {
      "epoch": 104.65486725663717,
      "grad_norm": 9.174766091746278e-06,
      "learning_rate": 0.28384298832827526,
      "loss": 0.0,
      "num_input_tokens_seen": 3392392,
      "step": 5965
    },
    {
      "epoch": 104.7433628318584,
      "grad_norm": 4.127657575736521e-06,
      "learning_rate": 0.28381638426442457,
      "loss": 0.0,
      "num_input_tokens_seen": 3395304,
      "step": 5970
    },
    {
      "epoch": 104.83185840707965,
      "grad_norm": 7.833178642613348e-06,
      "learning_rate": 0.2837897595644057,
      "loss": 0.0,
      "num_input_tokens_seen": 3397896,
      "step": 5975
    },
    {
      "epoch": 104.92035398230088,
      "grad_norm": 5.309138941811398e-06,
      "learning_rate": 0.28376311423232475,
      "loss": 0.0,
      "num_input_tokens_seen": 3400552,
      "step": 5980
    },
    {
      "epoch": 105.0,
      "grad_norm": 7.199303126981249e-06,
      "learning_rate": 0.2837364482722905,
      "loss": 0.0,
      "num_input_tokens_seen": 3402936,
      "step": 5985
    },
    {
      "epoch": 105.08849557522124,
      "grad_norm": 3.823163751803804e-06,
      "learning_rate": 0.28370976168841533,
      "loss": 0.0,
      "num_input_tokens_seen": 3405640,
      "step": 5990
    },
    {
      "epoch": 105.17699115044248,
      "grad_norm": 8.13579754321836e-06,
      "learning_rate": 0.2836830544848146,
      "loss": 0.0,
      "num_input_tokens_seen": 3408232,
      "step": 5995
    },
    {
      "epoch": 105.26548672566372,
      "grad_norm": 7.552147508249618e-06,
      "learning_rate": 0.2836563266656069,
      "loss": 0.0,
      "num_input_tokens_seen": 3411544,
      "step": 6000
    },
    {
      "epoch": 105.26548672566372,
      "eval_loss": 0.39194348454475403,
      "eval_runtime": 0.9414,
      "eval_samples_per_second": 26.557,
      "eval_steps_per_second": 13.81,
      "num_input_tokens_seen": 3411544,
      "step": 6000
    },
    {
      "epoch": 105.35398230088495,
      "grad_norm": 1.6060544112406205e-06,
      "learning_rate": 0.283629578234914,
      "loss": 0.0,
      "num_input_tokens_seen": 3414664,
      "step": 6005
    },
    {
      "epoch": 105.4424778761062,
      "grad_norm": 1.640885102460743e-06,
      "learning_rate": 0.2836028091968608,
      "loss": 0.0,
      "num_input_tokens_seen": 3417528,
      "step": 6010
    },
    {
      "epoch": 105.53097345132744,
      "grad_norm": 3.6557187286234694e-06,
      "learning_rate": 0.28357601955557554,
      "loss": 0.0,
      "num_input_tokens_seen": 3420408,
      "step": 6015
    },
    {
      "epoch": 105.61946902654867,
      "grad_norm": 7.786599780956749e-06,
      "learning_rate": 0.2835492093151894,
      "loss": 0.0,
      "num_input_tokens_seen": 3423272,
      "step": 6020
    },
    {
      "epoch": 105.70796460176992,
      "grad_norm": 1.7630585489314399e-06,
      "learning_rate": 0.2835223784798369,
      "loss": 0.0,
      "num_input_tokens_seen": 3426264,
      "step": 6025
    },
    {
      "epoch": 105.79646017699115,
      "grad_norm": 2.4380735794693464e-06,
      "learning_rate": 0.2834955270536557,
      "loss": 0.0,
      "num_input_tokens_seen": 3429304,
      "step": 6030
    },
    {
      "epoch": 105.88495575221239,
      "grad_norm": 6.758303356946271e-07,
      "learning_rate": 0.2834686550407866,
      "loss": 0.0,
      "num_input_tokens_seen": 3432120,
      "step": 6035
    },
    {
      "epoch": 105.97345132743362,
      "grad_norm": 7.810292117937934e-06,
      "learning_rate": 0.28344176244537367,
      "loss": 0.0,
      "num_input_tokens_seen": 3434552,
      "step": 6040
    },
    {
      "epoch": 106.05309734513274,
      "grad_norm": 5.205596607993357e-06,
      "learning_rate": 0.28341484927156396,
      "loss": 0.0,
      "num_input_tokens_seen": 3436568,
      "step": 6045
    },
    {
      "epoch": 106.14159292035399,
      "grad_norm": 2.5173019366775407e-06,
      "learning_rate": 0.28338791552350795,
      "loss": 0.0,
      "num_input_tokens_seen": 3440088,
      "step": 6050
    },
    {
      "epoch": 106.23008849557522,
      "grad_norm": 2.6564669042272726e-06,
      "learning_rate": 0.28336096120535914,
      "loss": 0.0,
      "num_input_tokens_seen": 3443400,
      "step": 6055
    },
    {
      "epoch": 106.31858407079646,
      "grad_norm": 4.999465090804733e-06,
      "learning_rate": 0.2833339863212741,
      "loss": 0.0,
      "num_input_tokens_seen": 3446120,
      "step": 6060
    },
    {
      "epoch": 106.40707964601769,
      "grad_norm": 5.723108188249171e-06,
      "learning_rate": 0.28330699087541283,
      "loss": 0.0,
      "num_input_tokens_seen": 3448520,
      "step": 6065
    },
    {
      "epoch": 106.49557522123894,
      "grad_norm": 4.3669965634762775e-06,
      "learning_rate": 0.2832799748719384,
      "loss": 0.0,
      "num_input_tokens_seen": 3451192,
      "step": 6070
    },
    {
      "epoch": 106.58407079646018,
      "grad_norm": 4.829083081858698e-06,
      "learning_rate": 0.28325293831501686,
      "loss": 0.0,
      "num_input_tokens_seen": 3454376,
      "step": 6075
    },
    {
      "epoch": 106.67256637168141,
      "grad_norm": 4.783206804859219e-06,
      "learning_rate": 0.2832258812088177,
      "loss": 0.0,
      "num_input_tokens_seen": 3456872,
      "step": 6080
    },
    {
      "epoch": 106.76106194690266,
      "grad_norm": 1.961733687494416e-06,
      "learning_rate": 0.2831988035575134,
      "loss": 0.0,
      "num_input_tokens_seen": 3459512,
      "step": 6085
    },
    {
      "epoch": 106.84955752212389,
      "grad_norm": 3.420240773266414e-06,
      "learning_rate": 0.28317170536527975,
      "loss": 0.0,
      "num_input_tokens_seen": 3462472,
      "step": 6090
    },
    {
      "epoch": 106.93805309734513,
      "grad_norm": 1.540603307148558e-06,
      "learning_rate": 0.2831445866362956,
      "loss": 0.0,
      "num_input_tokens_seen": 3465096,
      "step": 6095
    },
    {
      "epoch": 107.01769911504425,
      "grad_norm": 4.0676791286387015e-06,
      "learning_rate": 0.2831174473747429,
      "loss": 0.0,
      "num_input_tokens_seen": 3467568,
      "step": 6100
    },
    {
      "epoch": 107.10619469026548,
      "grad_norm": 4.596388862410095e-06,
      "learning_rate": 0.2830902875848071,
      "loss": 0.0,
      "num_input_tokens_seen": 3470832,
      "step": 6105
    },
    {
      "epoch": 107.19469026548673,
      "grad_norm": 3.92007041227771e-06,
      "learning_rate": 0.28306310727067635,
      "loss": 0.0,
      "num_input_tokens_seen": 3474032,
      "step": 6110
    },
    {
      "epoch": 107.28318584070796,
      "grad_norm": 2.386195092185517e-06,
      "learning_rate": 0.2830359064365423,
      "loss": 0.0,
      "num_input_tokens_seen": 3476944,
      "step": 6115
    },
    {
      "epoch": 107.3716814159292,
      "grad_norm": 5.933271950198105e-06,
      "learning_rate": 0.28300868508659965,
      "loss": 0.0,
      "num_input_tokens_seen": 3479840,
      "step": 6120
    },
    {
      "epoch": 107.46017699115045,
      "grad_norm": 1.1817024869742454e-06,
      "learning_rate": 0.28298144322504626,
      "loss": 0.0,
      "num_input_tokens_seen": 3482960,
      "step": 6125
    },
    {
      "epoch": 107.54867256637168,
      "grad_norm": 7.710846148256678e-06,
      "learning_rate": 0.2829541808560832,
      "loss": 0.0,
      "num_input_tokens_seen": 3485280,
      "step": 6130
    },
    {
      "epoch": 107.63716814159292,
      "grad_norm": 4.327190708863782e-06,
      "learning_rate": 0.2829268979839146,
      "loss": 0.0,
      "num_input_tokens_seen": 3488304,
      "step": 6135
    },
    {
      "epoch": 107.72566371681415,
      "grad_norm": 1.8247881143906852e-06,
      "learning_rate": 0.2828995946127479,
      "loss": 0.0,
      "num_input_tokens_seen": 3491232,
      "step": 6140
    },
    {
      "epoch": 107.8141592920354,
      "grad_norm": 7.242450010380708e-06,
      "learning_rate": 0.2828722707467936,
      "loss": 0.0,
      "num_input_tokens_seen": 3494368,
      "step": 6145
    },
    {
      "epoch": 107.90265486725664,
      "grad_norm": 4.23063465859741e-06,
      "learning_rate": 0.2828449263902653,
      "loss": 0.0,
      "num_input_tokens_seen": 3496912,
      "step": 6150
    },
    {
      "epoch": 107.99115044247787,
      "grad_norm": 1.4614139445257024e-06,
      "learning_rate": 0.28281756154738,
      "loss": 0.0,
      "num_input_tokens_seen": 3499632,
      "step": 6155
    },
    {
      "epoch": 108.070796460177,
      "grad_norm": 1.1395134151825914e-06,
      "learning_rate": 0.28279017622235764,
      "loss": 0.0,
      "num_input_tokens_seen": 3502320,
      "step": 6160
    },
    {
      "epoch": 108.15929203539822,
      "grad_norm": 3.575568371161353e-06,
      "learning_rate": 0.28276277041942127,
      "loss": 0.0,
      "num_input_tokens_seen": 3505104,
      "step": 6165
    },
    {
      "epoch": 108.24778761061947,
      "grad_norm": 2.577141458459664e-06,
      "learning_rate": 0.2827353441427974,
      "loss": 0.0,
      "num_input_tokens_seen": 3507744,
      "step": 6170
    },
    {
      "epoch": 108.33628318584071,
      "grad_norm": 4.559305125440005e-06,
      "learning_rate": 0.2827078973967153,
      "loss": 0.0,
      "num_input_tokens_seen": 3510848,
      "step": 6175
    },
    {
      "epoch": 108.42477876106194,
      "grad_norm": 1.0862028148039826e-06,
      "learning_rate": 0.2826804301854078,
      "loss": 0.0,
      "num_input_tokens_seen": 3513952,
      "step": 6180
    },
    {
      "epoch": 108.51327433628319,
      "grad_norm": 1.0676974852685817e-05,
      "learning_rate": 0.2826529425131105,
      "loss": 0.0,
      "num_input_tokens_seen": 3516992,
      "step": 6185
    },
    {
      "epoch": 108.60176991150442,
      "grad_norm": 1.7182989040520624e-06,
      "learning_rate": 0.2826254343840625,
      "loss": 0.0,
      "num_input_tokens_seen": 3519952,
      "step": 6190
    },
    {
      "epoch": 108.69026548672566,
      "grad_norm": 4.042374257551273e-06,
      "learning_rate": 0.2825979058025059,
      "loss": 0.0,
      "num_input_tokens_seen": 3522688,
      "step": 6195
    },
    {
      "epoch": 108.77876106194691,
      "grad_norm": 9.53942617343273e-06,
      "learning_rate": 0.2825703567726858,
      "loss": 0.0,
      "num_input_tokens_seen": 3525472,
      "step": 6200
    },
    {
      "epoch": 108.77876106194691,
      "eval_loss": 0.4034156799316406,
      "eval_runtime": 0.9299,
      "eval_samples_per_second": 26.884,
      "eval_steps_per_second": 13.98,
      "num_input_tokens_seen": 3525472,
      "step": 6200
    },
    {
      "epoch": 108.86725663716814,
      "grad_norm": 6.004881470289547e-06,
      "learning_rate": 0.2825427872988508,
      "loss": 0.0,
      "num_input_tokens_seen": 3528720,
      "step": 6205
    },
    {
      "epoch": 108.95575221238938,
      "grad_norm": 6.326859875116497e-06,
      "learning_rate": 0.28251519738525227,
      "loss": 0.0,
      "num_input_tokens_seen": 3531264,
      "step": 6210
    },
    {
      "epoch": 109.03539823008849,
      "grad_norm": 5.74574096390279e-06,
      "learning_rate": 0.28248758703614507,
      "loss": 0.0,
      "num_input_tokens_seen": 3533456,
      "step": 6215
    },
    {
      "epoch": 109.12389380530973,
      "grad_norm": 5.970582606096286e-06,
      "learning_rate": 0.28245995625578696,
      "loss": 0.0,
      "num_input_tokens_seen": 3536624,
      "step": 6220
    },
    {
      "epoch": 109.21238938053098,
      "grad_norm": 1.1766745046770666e-05,
      "learning_rate": 0.282432305048439,
      "loss": 0.0,
      "num_input_tokens_seen": 3539520,
      "step": 6225
    },
    {
      "epoch": 109.30088495575221,
      "grad_norm": 5.028723535360768e-07,
      "learning_rate": 0.28240463341836536,
      "loss": 0.0,
      "num_input_tokens_seen": 3542160,
      "step": 6230
    },
    {
      "epoch": 109.38938053097345,
      "grad_norm": 4.864241873292485e-06,
      "learning_rate": 0.2823769413698334,
      "loss": 0.0,
      "num_input_tokens_seen": 3544944,
      "step": 6235
    },
    {
      "epoch": 109.47787610619469,
      "grad_norm": 1.9876190435752505e-06,
      "learning_rate": 0.2823492289071135,
      "loss": 0.0,
      "num_input_tokens_seen": 3547424,
      "step": 6240
    },
    {
      "epoch": 109.56637168141593,
      "grad_norm": 2.344625045225257e-06,
      "learning_rate": 0.2823214960344793,
      "loss": 0.0,
      "num_input_tokens_seen": 3550416,
      "step": 6245
    },
    {
      "epoch": 109.65486725663717,
      "grad_norm": 3.9153246689238586e-06,
      "learning_rate": 0.28229374275620756,
      "loss": 0.0,
      "num_input_tokens_seen": 3553728,
      "step": 6250
    },
    {
      "epoch": 109.7433628318584,
      "grad_norm": 1.2720953463940532e-06,
      "learning_rate": 0.28226596907657814,
      "loss": 0.0,
      "num_input_tokens_seen": 3556992,
      "step": 6255
    },
    {
      "epoch": 109.83185840707965,
      "grad_norm": 1.0839229389603133e-06,
      "learning_rate": 0.28223817499987414,
      "loss": 0.0,
      "num_input_tokens_seen": 3559776,
      "step": 6260
    },
    {
      "epoch": 109.92035398230088,
      "grad_norm": 2.831737901942688e-06,
      "learning_rate": 0.2822103605303818,
      "loss": 0.0,
      "num_input_tokens_seen": 3562640,
      "step": 6265
    },
    {
      "epoch": 110.0,
      "grad_norm": 6.756088737347454e-07,
      "learning_rate": 0.2821825256723903,
      "loss": 0.0,
      "num_input_tokens_seen": 3565040,
      "step": 6270
    },
    {
      "epoch": 110.08849557522124,
      "grad_norm": 2.571578079368919e-06,
      "learning_rate": 0.2821546704301923,
      "loss": 0.0,
      "num_input_tokens_seen": 3567632,
      "step": 6275
    },
    {
      "epoch": 110.17699115044248,
      "grad_norm": 2.2433482627093326e-06,
      "learning_rate": 0.2821267948080834,
      "loss": 0.0,
      "num_input_tokens_seen": 3570816,
      "step": 6280
    },
    {
      "epoch": 110.26548672566372,
      "grad_norm": 1.6697358660167083e-06,
      "learning_rate": 0.28209889881036226,
      "loss": 0.0,
      "num_input_tokens_seen": 3573392,
      "step": 6285
    },
    {
      "epoch": 110.35398230088495,
      "grad_norm": 7.419398571073543e-06,
      "learning_rate": 0.28207098244133094,
      "loss": 0.0,
      "num_input_tokens_seen": 3576192,
      "step": 6290
    },
    {
      "epoch": 110.4424778761062,
      "grad_norm": 4.1724465518200304e-06,
      "learning_rate": 0.2820430457052943,
      "loss": 0.0,
      "num_input_tokens_seen": 3578928,
      "step": 6295
    },
    {
      "epoch": 110.53097345132744,
      "grad_norm": 4.223645646561636e-06,
      "learning_rate": 0.28201508860656077,
      "loss": 0.0,
      "num_input_tokens_seen": 3581680,
      "step": 6300
    },
    {
      "epoch": 110.61946902654867,
      "grad_norm": 3.505168706396944e-06,
      "learning_rate": 0.2819871111494415,
      "loss": 0.0,
      "num_input_tokens_seen": 3584880,
      "step": 6305
    },
    {
      "epoch": 110.70796460176992,
      "grad_norm": 6.081432275095722e-06,
      "learning_rate": 0.28195911333825113,
      "loss": 0.0,
      "num_input_tokens_seen": 3587760,
      "step": 6310
    },
    {
      "epoch": 110.79646017699115,
      "grad_norm": 2.263700253024581e-06,
      "learning_rate": 0.28193109517730713,
      "loss": 0.0,
      "num_input_tokens_seen": 3590752,
      "step": 6315
    },
    {
      "epoch": 110.88495575221239,
      "grad_norm": 1.3871551800548332e-06,
      "learning_rate": 0.2819030566709303,
      "loss": 0.0,
      "num_input_tokens_seen": 3593648,
      "step": 6320
    },
    {
      "epoch": 110.97345132743362,
      "grad_norm": 2.5103088319156086e-06,
      "learning_rate": 0.2818749978234445,
      "loss": 0.0,
      "num_input_tokens_seen": 3596384,
      "step": 6325
    },
    {
      "epoch": 111.05309734513274,
      "grad_norm": 6.441619916586205e-06,
      "learning_rate": 0.2818469186391768,
      "loss": 0.0,
      "num_input_tokens_seen": 3598640,
      "step": 6330
    },
    {
      "epoch": 111.14159292035399,
      "grad_norm": 3.4211325328215025e-06,
      "learning_rate": 0.28181881912245743,
      "loss": 0.0,
      "num_input_tokens_seen": 3601328,
      "step": 6335
    },
    {
      "epoch": 111.23008849557522,
      "grad_norm": 8.913207238947507e-07,
      "learning_rate": 0.2817906992776195,
      "loss": 0.0,
      "num_input_tokens_seen": 3604192,
      "step": 6340
    },
    {
      "epoch": 111.31858407079646,
      "grad_norm": 1.8401816532787052e-06,
      "learning_rate": 0.28176255910899967,
      "loss": 0.0,
      "num_input_tokens_seen": 3606800,
      "step": 6345
    },
    {
      "epoch": 111.40707964601769,
      "grad_norm": 2.5008516786328983e-06,
      "learning_rate": 0.2817343986209373,
      "loss": 0.0,
      "num_input_tokens_seen": 3609728,
      "step": 6350
    },
    {
      "epoch": 111.49557522123894,
      "grad_norm": 5.574369879468577e-06,
      "learning_rate": 0.2817062178177753,
      "loss": 0.0,
      "num_input_tokens_seen": 3612480,
      "step": 6355
    },
    {
      "epoch": 111.58407079646018,
      "grad_norm": 2.0235231659171404e-06,
      "learning_rate": 0.2816780167038593,
      "loss": 0.0,
      "num_input_tokens_seen": 3615760,
      "step": 6360
    },
    {
      "epoch": 111.67256637168141,
      "grad_norm": 2.8737561024172464e-06,
      "learning_rate": 0.28164979528353834,
      "loss": 0.0,
      "num_input_tokens_seen": 3618656,
      "step": 6365
    },
    {
      "epoch": 111.76106194690266,
      "grad_norm": 3.7827453525096644e-06,
      "learning_rate": 0.28162155356116453,
      "loss": 0.0,
      "num_input_tokens_seen": 3621264,
      "step": 6370
    },
    {
      "epoch": 111.84955752212389,
      "grad_norm": 1.1472453707028762e-06,
      "learning_rate": 0.28159329154109314,
      "loss": 0.0,
      "num_input_tokens_seen": 3624000,
      "step": 6375
    },
    {
      "epoch": 111.93805309734513,
      "grad_norm": 5.60845455765957e-06,
      "learning_rate": 0.28156500922768246,
      "loss": 0.0,
      "num_input_tokens_seen": 3627072,
      "step": 6380
    },
    {
      "epoch": 112.01769911504425,
      "grad_norm": 3.014007916135597e-06,
      "learning_rate": 0.28153670662529406,
      "loss": 0.0,
      "num_input_tokens_seen": 3629896,
      "step": 6385
    },
    {
      "epoch": 112.10619469026548,
      "grad_norm": 6.613041705350042e-07,
      "learning_rate": 0.28150838373829246,
      "loss": 0.0,
      "num_input_tokens_seen": 3632520,
      "step": 6390
    },
    {
      "epoch": 112.19469026548673,
      "grad_norm": 9.81797984422883e-06,
      "learning_rate": 0.2814800405710455,
      "loss": 0.0,
      "num_input_tokens_seen": 3635432,
      "step": 6395
    },
    {
      "epoch": 112.28318584070796,
      "grad_norm": 2.9945929327368503e-06,
      "learning_rate": 0.2814516771279239,
      "loss": 0.0,
      "num_input_tokens_seen": 3638584,
      "step": 6400
    },
    {
      "epoch": 112.28318584070796,
      "eval_loss": 0.41279280185699463,
      "eval_runtime": 0.9376,
      "eval_samples_per_second": 26.662,
      "eval_steps_per_second": 13.864,
      "num_input_tokens_seen": 3638584,
      "step": 6400
    },
    {
      "epoch": 112.3716814159292,
      "grad_norm": 4.5655096982955e-06,
      "learning_rate": 0.28142329341330186,
      "loss": 0.0,
      "num_input_tokens_seen": 3641288,
      "step": 6405
    },
    {
      "epoch": 112.46017699115045,
      "grad_norm": 2.089519739456591e-06,
      "learning_rate": 0.2813948894315564,
      "loss": 0.0,
      "num_input_tokens_seen": 3643800,
      "step": 6410
    },
    {
      "epoch": 112.54867256637168,
      "grad_norm": 4.490387709665811e-06,
      "learning_rate": 0.2813664651870677,
      "loss": 0.0,
      "num_input_tokens_seen": 3646792,
      "step": 6415
    },
    {
      "epoch": 112.63716814159292,
      "grad_norm": 8.071289130384685e-07,
      "learning_rate": 0.28133802068421926,
      "loss": 0.0,
      "num_input_tokens_seen": 3649960,
      "step": 6420
    },
    {
      "epoch": 112.72566371681415,
      "grad_norm": 1.0283761184837203e-06,
      "learning_rate": 0.28130955592739754,
      "loss": 0.0,
      "num_input_tokens_seen": 3653080,
      "step": 6425
    },
    {
      "epoch": 112.8141592920354,
      "grad_norm": 6.037210482645605e-07,
      "learning_rate": 0.2812810709209922,
      "loss": 0.0,
      "num_input_tokens_seen": 3655432,
      "step": 6430
    },
    {
      "epoch": 112.90265486725664,
      "grad_norm": 2.893356167987804e-06,
      "learning_rate": 0.2812525656693959,
      "loss": 0.0,
      "num_input_tokens_seen": 3658296,
      "step": 6435
    },
    {
      "epoch": 112.99115044247787,
      "grad_norm": 1.4307754554465646e-06,
      "learning_rate": 0.28122404017700453,
      "loss": 0.0,
      "num_input_tokens_seen": 3661048,
      "step": 6440
    },
    {
      "epoch": 113.070796460177,
      "grad_norm": 3.616989260990522e-06,
      "learning_rate": 0.2811954944482171,
      "loss": 0.0,
      "num_input_tokens_seen": 3663752,
      "step": 6445
    },
    {
      "epoch": 113.15929203539822,
      "grad_norm": 2.33480545830389e-06,
      "learning_rate": 0.2811669284874358,
      "loss": 0.0,
      "num_input_tokens_seen": 3666504,
      "step": 6450
    },
    {
      "epoch": 113.24778761061947,
      "grad_norm": 1.8145689182347269e-06,
      "learning_rate": 0.2811383422990657,
      "loss": 0.0,
      "num_input_tokens_seen": 3669464,
      "step": 6455
    },
    {
      "epoch": 113.33628318584071,
      "grad_norm": 5.135151695867535e-06,
      "learning_rate": 0.2811097358875152,
      "loss": 0.0,
      "num_input_tokens_seen": 3672584,
      "step": 6460
    },
    {
      "epoch": 113.42477876106194,
      "grad_norm": 2.878039822462597e-06,
      "learning_rate": 0.2810811092571959,
      "loss": 0.0,
      "num_input_tokens_seen": 3675400,
      "step": 6465
    },
    {
      "epoch": 113.51327433628319,
      "grad_norm": 7.295124078154913e-07,
      "learning_rate": 0.28105246241252224,
      "loss": 0.0,
      "num_input_tokens_seen": 3678488,
      "step": 6470
    },
    {
      "epoch": 113.60176991150442,
      "grad_norm": 8.516425964444352e-07,
      "learning_rate": 0.28102379535791194,
      "loss": 0.0,
      "num_input_tokens_seen": 3681240,
      "step": 6475
    },
    {
      "epoch": 113.69026548672566,
      "grad_norm": 1.1380113846826134e-06,
      "learning_rate": 0.2809951080977859,
      "loss": 0.0,
      "num_input_tokens_seen": 3684456,
      "step": 6480
    },
    {
      "epoch": 113.77876106194691,
      "grad_norm": 2.962567350550671e-06,
      "learning_rate": 0.28096640063656797,
      "loss": 0.0,
      "num_input_tokens_seen": 3687240,
      "step": 6485
    },
    {
      "epoch": 113.86725663716814,
      "grad_norm": 3.2509383345313836e-06,
      "learning_rate": 0.2809376729786852,
      "loss": 0.0,
      "num_input_tokens_seen": 3689736,
      "step": 6490
    },
    {
      "epoch": 113.95575221238938,
      "grad_norm": 3.876380560541293e-06,
      "learning_rate": 0.28090892512856785,
      "loss": 0.0,
      "num_input_tokens_seen": 3692584,
      "step": 6495
    },
    {
      "epoch": 114.03539823008849,
      "grad_norm": 5.931256964686327e-06,
      "learning_rate": 0.2808801570906491,
      "loss": 0.0,
      "num_input_tokens_seen": 3694592,
      "step": 6500
    },
    {
      "epoch": 114.12389380530973,
      "grad_norm": 1.6424693285443936e-06,
      "learning_rate": 0.2808513688693654,
      "loss": 0.0,
      "num_input_tokens_seen": 3697552,
      "step": 6505
    },
    {
      "epoch": 114.21238938053098,
      "grad_norm": 4.631248884834349e-06,
      "learning_rate": 0.28082256046915627,
      "loss": 0.0,
      "num_input_tokens_seen": 3700240,
      "step": 6510
    },
    {
      "epoch": 114.30088495575221,
      "grad_norm": 4.553704911813838e-06,
      "learning_rate": 0.28079373189446427,
      "loss": 0.0,
      "num_input_tokens_seen": 3703280,
      "step": 6515
    },
    {
      "epoch": 114.38938053097345,
      "grad_norm": 3.377109578650561e-06,
      "learning_rate": 0.28076488314973513,
      "loss": 0.0,
      "num_input_tokens_seen": 3706368,
      "step": 6520
    },
    {
      "epoch": 114.47787610619469,
      "grad_norm": 3.315898993605515e-06,
      "learning_rate": 0.28073601423941774,
      "loss": 0.0,
      "num_input_tokens_seen": 3708960,
      "step": 6525
    },
    {
      "epoch": 114.56637168141593,
      "grad_norm": 7.3036717367358506e-06,
      "learning_rate": 0.28070712516796403,
      "loss": 0.0,
      "num_input_tokens_seen": 3712064,
      "step": 6530
    },
    {
      "epoch": 114.65486725663717,
      "grad_norm": 4.499315195971576e-07,
      "learning_rate": 0.28067821593982906,
      "loss": 0.0,
      "num_input_tokens_seen": 3715072,
      "step": 6535
    },
    {
      "epoch": 114.7433628318584,
      "grad_norm": 5.083532869321061e-06,
      "learning_rate": 0.28064928655947097,
      "loss": 0.0,
      "num_input_tokens_seen": 3718304,
      "step": 6540
    },
    {
      "epoch": 114.83185840707965,
      "grad_norm": 3.659087951746187e-06,
      "learning_rate": 0.28062033703135103,
      "loss": 0.0,
      "num_input_tokens_seen": 3720928,
      "step": 6545
    },
    {
      "epoch": 114.92035398230088,
      "grad_norm": 3.8036662317608716e-06,
      "learning_rate": 0.2805913673599337,
      "loss": 0.0,
      "num_input_tokens_seen": 3723792,
      "step": 6550
    },
    {
      "epoch": 115.0,
      "grad_norm": 3.122393650301092e-07,
      "learning_rate": 0.2805623775496864,
      "loss": 0.0,
      "num_input_tokens_seen": 3726240,
      "step": 6555
    },
    {
      "epoch": 115.08849557522124,
      "grad_norm": 8.868903023540042e-06,
      "learning_rate": 0.2805333676050797,
      "loss": 0.0,
      "num_input_tokens_seen": 3729184,
      "step": 6560
    },
    {
      "epoch": 115.17699115044248,
      "grad_norm": 2.1956632281217026e-06,
      "learning_rate": 0.2805043375305873,
      "loss": 0.0,
      "num_input_tokens_seen": 3732400,
      "step": 6565
    },
    {
      "epoch": 115.26548672566372,
      "grad_norm": 3.1212086923915194e-06,
      "learning_rate": 0.2804752873306861,
      "loss": 0.0,
      "num_input_tokens_seen": 3735344,
      "step": 6570
    },
    {
      "epoch": 115.35398230088495,
      "grad_norm": 2.4761682198004564e-06,
      "learning_rate": 0.2804462170098559,
      "loss": 0.0,
      "num_input_tokens_seen": 3738208,
      "step": 6575
    },
    {
      "epoch": 115.4424778761062,
      "grad_norm": 3.0259809591370868e-06,
      "learning_rate": 0.2804171265725797,
      "loss": 0.0,
      "num_input_tokens_seen": 3741184,
      "step": 6580
    },
    {
      "epoch": 115.53097345132744,
      "grad_norm": 3.7790910027979407e-06,
      "learning_rate": 0.28038801602334373,
      "loss": 0.0,
      "num_input_tokens_seen": 3743648,
      "step": 6585
    },
    {
      "epoch": 115.61946902654867,
      "grad_norm": 1.2967542488695472e-06,
      "learning_rate": 0.28035888536663717,
      "loss": 0.0,
      "num_input_tokens_seen": 3746544,
      "step": 6590
    },
    {
      "epoch": 115.70796460176992,
      "grad_norm": 5.396545020630583e-06,
      "learning_rate": 0.2803297346069522,
      "loss": 0.0,
      "num_input_tokens_seen": 3749008,
      "step": 6595
    },
    {
      "epoch": 115.79646017699115,
      "grad_norm": 6.649771421507467e-06,
      "learning_rate": 0.28030056374878437,
      "loss": 0.0,
      "num_input_tokens_seen": 3752608,
      "step": 6600
    },
    {
      "epoch": 115.79646017699115,
      "eval_loss": 0.4014095067977905,
      "eval_runtime": 0.9189,
      "eval_samples_per_second": 27.205,
      "eval_steps_per_second": 14.147,
      "num_input_tokens_seen": 3752608,
      "step": 6600
    },
    {
      "epoch": 115.88495575221239,
      "grad_norm": 1.8778681578623946e-06,
      "learning_rate": 0.2802713727966321,
      "loss": 0.0,
      "num_input_tokens_seen": 3755152,
      "step": 6605
    },
    {
      "epoch": 115.97345132743362,
      "grad_norm": 1.8126271470464417e-06,
      "learning_rate": 0.28024216175499717,
      "loss": 0.0,
      "num_input_tokens_seen": 3758048,
      "step": 6610
    },
    {
      "epoch": 116.05309734513274,
      "grad_norm": 2.566056764408131e-06,
      "learning_rate": 0.2802129306283841,
      "loss": 0.0,
      "num_input_tokens_seen": 3760768,
      "step": 6615
    },
    {
      "epoch": 116.14159292035399,
      "grad_norm": 5.246491127763875e-06,
      "learning_rate": 0.28018367942130074,
      "loss": 0.0,
      "num_input_tokens_seen": 3763296,
      "step": 6620
    },
    {
      "epoch": 116.23008849557522,
      "grad_norm": 1.5780573221491068e-06,
      "learning_rate": 0.28015440813825804,
      "loss": 0.0,
      "num_input_tokens_seen": 3766656,
      "step": 6625
    },
    {
      "epoch": 116.31858407079646,
      "grad_norm": 3.5800467230728827e-06,
      "learning_rate": 0.28012511678377006,
      "loss": 0.0,
      "num_input_tokens_seen": 3769568,
      "step": 6630
    },
    {
      "epoch": 116.40707964601769,
      "grad_norm": 4.50969628218445e-06,
      "learning_rate": 0.28009580536235373,
      "loss": 0.0,
      "num_input_tokens_seen": 3772480,
      "step": 6635
    },
    {
      "epoch": 116.49557522123894,
      "grad_norm": 1.1545778306754073e-06,
      "learning_rate": 0.28006647387852934,
      "loss": 0.0,
      "num_input_tokens_seen": 3775360,
      "step": 6640
    },
    {
      "epoch": 116.58407079646018,
      "grad_norm": 7.052641421978478e-07,
      "learning_rate": 0.28003712233682015,
      "loss": 0.0,
      "num_input_tokens_seen": 3778352,
      "step": 6645
    },
    {
      "epoch": 116.67256637168141,
      "grad_norm": 3.890377684001578e-06,
      "learning_rate": 0.2800077507417526,
      "loss": 0.0,
      "num_input_tokens_seen": 3781216,
      "step": 6650
    },
    {
      "epoch": 116.76106194690266,
      "grad_norm": 3.996365194325335e-06,
      "learning_rate": 0.2799783590978561,
      "loss": 0.0,
      "num_input_tokens_seen": 3783664,
      "step": 6655
    },
    {
      "epoch": 116.84955752212389,
      "grad_norm": 6.993755505391164e-06,
      "learning_rate": 0.2799489474096632,
      "loss": 0.0,
      "num_input_tokens_seen": 3786576,
      "step": 6660
    },
    {
      "epoch": 116.93805309734513,
      "grad_norm": 5.268841505312594e-06,
      "learning_rate": 0.27991951568170953,
      "loss": 0.0,
      "num_input_tokens_seen": 3789504,
      "step": 6665
    },
    {
      "epoch": 117.01769911504425,
      "grad_norm": 6.807047157053603e-06,
      "learning_rate": 0.2798900639185339,
      "loss": 0.0,
      "num_input_tokens_seen": 3791752,
      "step": 6670
    },
    {
      "epoch": 117.10619469026548,
      "grad_norm": 7.140005891415058e-06,
      "learning_rate": 0.2798605921246781,
      "loss": 0.0,
      "num_input_tokens_seen": 3794520,
      "step": 6675
    },
    {
      "epoch": 117.19469026548673,
      "grad_norm": 3.160890400977223e-06,
      "learning_rate": 0.2798311003046871,
      "loss": 0.0,
      "num_input_tokens_seen": 3797096,
      "step": 6680
    },
    {
      "epoch": 117.28318584070796,
      "grad_norm": 1.093963533094211e-06,
      "learning_rate": 0.2798015884631089,
      "loss": 0.0,
      "num_input_tokens_seen": 3799576,
      "step": 6685
    },
    {
      "epoch": 117.3716814159292,
      "grad_norm": 1.8338051859245752e-06,
      "learning_rate": 0.27977205660449445,
      "loss": 0.0,
      "num_input_tokens_seen": 3802632,
      "step": 6690
    },
    {
      "epoch": 117.46017699115045,
      "grad_norm": 2.9383779747149674e-06,
      "learning_rate": 0.2797425047333981,
      "loss": 0.0,
      "num_input_tokens_seen": 3805864,
      "step": 6695
    },
    {
      "epoch": 117.54867256637168,
      "grad_norm": 4.917459136777325e-06,
      "learning_rate": 0.27971293285437715,
      "loss": 0.0,
      "num_input_tokens_seen": 3809096,
      "step": 6700
    },
    {
      "epoch": 117.63716814159292,
      "grad_norm": 3.287883600933128e-06,
      "learning_rate": 0.2796833409719918,
      "loss": 0.0,
      "num_input_tokens_seen": 3811912,
      "step": 6705
    },
    {
      "epoch": 117.72566371681415,
      "grad_norm": 2.6435270683577983e-06,
      "learning_rate": 0.27965372909080566,
      "loss": 0.0,
      "num_input_tokens_seen": 3814424,
      "step": 6710
    },
    {
      "epoch": 117.8141592920354,
      "grad_norm": 5.0904209274449386e-06,
      "learning_rate": 0.27962409721538506,
      "loss": 0.0,
      "num_input_tokens_seen": 3817496,
      "step": 6715
    },
    {
      "epoch": 117.90265486725664,
      "grad_norm": 1.1088524161095847e-06,
      "learning_rate": 0.27959444535029976,
      "loss": 0.0,
      "num_input_tokens_seen": 3820184,
      "step": 6720
    },
    {
      "epoch": 117.99115044247787,
      "grad_norm": 6.124179890321102e-06,
      "learning_rate": 0.27956477350012243,
      "loss": 0.0,
      "num_input_tokens_seen": 3823400,
      "step": 6725
    },
    {
      "epoch": 118.070796460177,
      "grad_norm": 2.953082685053232e-06,
      "learning_rate": 0.27953508166942875,
      "loss": 0.0,
      "num_input_tokens_seen": 3825952,
      "step": 6730
    },
    {
      "epoch": 118.15929203539822,
      "grad_norm": 3.6296173675509635e-06,
      "learning_rate": 0.27950536986279767,
      "loss": 0.0,
      "num_input_tokens_seen": 3828912,
      "step": 6735
    },
    {
      "epoch": 118.24778761061947,
      "grad_norm": 1.3488142940332182e-06,
      "learning_rate": 0.2794756380848111,
      "loss": 0.0,
      "num_input_tokens_seen": 3831824,
      "step": 6740
    },
    {
      "epoch": 118.33628318584071,
      "grad_norm": 1.1574784366530366e-06,
      "learning_rate": 0.279445886340054,
      "loss": 0.0,
      "num_input_tokens_seen": 3834688,
      "step": 6745
    },
    {
      "epoch": 118.42477876106194,
      "grad_norm": 3.3695471302053193e-06,
      "learning_rate": 0.27941611463311455,
      "loss": 0.0,
      "num_input_tokens_seen": 3837680,
      "step": 6750
    },
    {
      "epoch": 118.51327433628319,
      "grad_norm": 3.107046950390213e-06,
      "learning_rate": 0.2793863229685839,
      "loss": 0.0,
      "num_input_tokens_seen": 3840176,
      "step": 6755
    },
    {
      "epoch": 118.60176991150442,
      "grad_norm": 1.6883093394426396e-06,
      "learning_rate": 0.27935651135105627,
      "loss": 0.0,
      "num_input_tokens_seen": 3843536,
      "step": 6760
    },
    {
      "epoch": 118.69026548672566,
      "grad_norm": 3.5395726172282593e-06,
      "learning_rate": 0.279326679785129,
      "loss": 0.0,
      "num_input_tokens_seen": 3846512,
      "step": 6765
    },
    {
      "epoch": 118.77876106194691,
      "grad_norm": 3.279932798250229e-06,
      "learning_rate": 0.2792968282754024,
      "loss": 0.0,
      "num_input_tokens_seen": 3849184,
      "step": 6770
    },
    {
      "epoch": 118.86725663716814,
      "grad_norm": 1.5161696182985906e-06,
      "learning_rate": 0.2792669568264801,
      "loss": 0.0,
      "num_input_tokens_seen": 3852064,
      "step": 6775
    },
    {
      "epoch": 118.95575221238938,
      "grad_norm": 7.331223514484009e-06,
      "learning_rate": 0.27923706544296856,
      "loss": 0.0,
      "num_input_tokens_seen": 3854688,
      "step": 6780
    },
    {
      "epoch": 119.03539823008849,
      "grad_norm": 4.011179044027813e-06,
      "learning_rate": 0.2792071541294775,
      "loss": 0.0,
      "num_input_tokens_seen": 3856896,
      "step": 6785
    },
    {
      "epoch": 119.12389380530973,
      "grad_norm": 3.297665898571722e-06,
      "learning_rate": 0.27917722289061947,
      "loss": 0.0,
      "num_input_tokens_seen": 3859152,
      "step": 6790
    },
    {
      "epoch": 119.21238938053098,
      "grad_norm": 3.4152069474657765e-06,
      "learning_rate": 0.27914727173101034,
      "loss": 0.0,
      "num_input_tokens_seen": 3862256,
      "step": 6795
    },
    {
      "epoch": 119.30088495575221,
      "grad_norm": 2.4761127406236483e-06,
      "learning_rate": 0.279117300655269,
      "loss": 0.0,
      "num_input_tokens_seen": 3865376,
      "step": 6800
    },
    {
      "epoch": 119.30088495575221,
      "eval_loss": 0.413632333278656,
      "eval_runtime": 0.9441,
      "eval_samples_per_second": 26.481,
      "eval_steps_per_second": 13.77,
      "num_input_tokens_seen": 3865376,
      "step": 6800
    },
    {
      "epoch": 119.38938053097345,
      "grad_norm": 1.7857523744169157e-06,
      "learning_rate": 0.2790873096680173,
      "loss": 0.0,
      "num_input_tokens_seen": 3868048,
      "step": 6805
    },
    {
      "epoch": 119.47787610619469,
      "grad_norm": 1.6539314628971624e-06,
      "learning_rate": 0.2790572987738802,
      "loss": 0.0,
      "num_input_tokens_seen": 3870816,
      "step": 6810
    },
    {
      "epoch": 119.56637168141593,
      "grad_norm": 4.1911630432878155e-06,
      "learning_rate": 0.27902726797748584,
      "loss": 0.0,
      "num_input_tokens_seen": 3874048,
      "step": 6815
    },
    {
      "epoch": 119.65486725663717,
      "grad_norm": 2.0297484297771007e-06,
      "learning_rate": 0.2789972172834652,
      "loss": 0.0,
      "num_input_tokens_seen": 3876832,
      "step": 6820
    },
    {
      "epoch": 119.7433628318584,
      "grad_norm": 2.7727110136765987e-06,
      "learning_rate": 0.2789671466964527,
      "loss": 0.0,
      "num_input_tokens_seen": 3879296,
      "step": 6825
    },
    {
      "epoch": 119.83185840707965,
      "grad_norm": 4.3897871364606544e-06,
      "learning_rate": 0.2789370562210854,
      "loss": 0.0,
      "num_input_tokens_seen": 3882592,
      "step": 6830
    },
    {
      "epoch": 119.92035398230088,
      "grad_norm": 3.456360900599975e-06,
      "learning_rate": 0.27890694586200376,
      "loss": 0.0,
      "num_input_tokens_seen": 3885584,
      "step": 6835
    },
    {
      "epoch": 120.0,
      "grad_norm": 4.630872808775166e-06,
      "learning_rate": 0.2788768156238511,
      "loss": 0.0,
      "num_input_tokens_seen": 3888056,
      "step": 6840
    },
    {
      "epoch": 120.08849557522124,
      "grad_norm": 4.375242042442551e-06,
      "learning_rate": 0.27884666551127385,
      "loss": 0.0,
      "num_input_tokens_seen": 3890744,
      "step": 6845
    },
    {
      "epoch": 120.17699115044248,
      "grad_norm": 3.999995897174813e-06,
      "learning_rate": 0.2788164955289217,
      "loss": 0.0,
      "num_input_tokens_seen": 3893640,
      "step": 6850
    },
    {
      "epoch": 120.26548672566372,
      "grad_norm": 2.1822904727741843e-06,
      "learning_rate": 0.27878630568144697,
      "loss": 0.0,
      "num_input_tokens_seen": 3896264,
      "step": 6855
    },
    {
      "epoch": 120.35398230088495,
      "grad_norm": 3.2314619602402672e-06,
      "learning_rate": 0.2787560959735056,
      "loss": 0.0,
      "num_input_tokens_seen": 3899192,
      "step": 6860
    },
    {
      "epoch": 120.4424778761062,
      "grad_norm": 2.4225992092397064e-06,
      "learning_rate": 0.27872586640975616,
      "loss": 0.0,
      "num_input_tokens_seen": 3901880,
      "step": 6865
    },
    {
      "epoch": 120.53097345132744,
      "grad_norm": 1.96384416994988e-06,
      "learning_rate": 0.27869561699486045,
      "loss": 0.0,
      "num_input_tokens_seen": 3904824,
      "step": 6870
    },
    {
      "epoch": 120.61946902654867,
      "grad_norm": 2.0964187115168897e-06,
      "learning_rate": 0.2786653477334833,
      "loss": 0.0,
      "num_input_tokens_seen": 3907720,
      "step": 6875
    },
    {
      "epoch": 120.70796460176992,
      "grad_norm": 2.5468459625699325e-06,
      "learning_rate": 0.2786350586302926,
      "loss": 0.0,
      "num_input_tokens_seen": 3911064,
      "step": 6880
    },
    {
      "epoch": 120.79646017699115,
      "grad_norm": 3.301451670267852e-06,
      "learning_rate": 0.27860474968995935,
      "loss": 0.0,
      "num_input_tokens_seen": 3913784,
      "step": 6885
    },
    {
      "epoch": 120.88495575221239,
      "grad_norm": 5.2054256229894236e-06,
      "learning_rate": 0.27857442091715756,
      "loss": 0.0,
      "num_input_tokens_seen": 3916504,
      "step": 6890
    },
    {
      "epoch": 120.97345132743362,
      "grad_norm": 2.01965758606093e-06,
      "learning_rate": 0.27854407231656425,
      "loss": 0.0,
      "num_input_tokens_seen": 3919400,
      "step": 6895
    },
    {
      "epoch": 121.05309734513274,
      "grad_norm": 2.4063390355877345e-06,
      "learning_rate": 0.2785137038928596,
      "loss": 0.0,
      "num_input_tokens_seen": 3921912,
      "step": 6900
    },
    {
      "epoch": 121.14159292035399,
      "grad_norm": 3.7687573239963967e-06,
      "learning_rate": 0.27848331565072687,
      "loss": 0.0,
      "num_input_tokens_seen": 3925128,
      "step": 6905
    },
    {
      "epoch": 121.23008849557522,
      "grad_norm": 3.864171958412044e-06,
      "learning_rate": 0.27845290759485225,
      "loss": 0.0,
      "num_input_tokens_seen": 3928264,
      "step": 6910
    },
    {
      "epoch": 121.31858407079646,
      "grad_norm": 1.2314530977164395e-06,
      "learning_rate": 0.278422479729925,
      "loss": 0.0,
      "num_input_tokens_seen": 3930824,
      "step": 6915
    },
    {
      "epoch": 121.40707964601769,
      "grad_norm": 8.771188504397287e-07,
      "learning_rate": 0.2783920320606375,
      "loss": 0.0,
      "num_input_tokens_seen": 3933864,
      "step": 6920
    },
    {
      "epoch": 121.49557522123894,
      "grad_norm": 2.6947134301735787e-06,
      "learning_rate": 0.2783615645916852,
      "loss": 0.0,
      "num_input_tokens_seen": 3936504,
      "step": 6925
    },
    {
      "epoch": 121.58407079646018,
      "grad_norm": 6.234613465494476e-07,
      "learning_rate": 0.2783310773277666,
      "loss": 0.0,
      "num_input_tokens_seen": 3939752,
      "step": 6930
    },
    {
      "epoch": 121.67256637168141,
      "grad_norm": 2.1099717741890345e-06,
      "learning_rate": 0.2783005702735831,
      "loss": 0.0,
      "num_input_tokens_seen": 3942872,
      "step": 6935
    },
    {
      "epoch": 121.76106194690266,
      "grad_norm": 3.747256414499134e-06,
      "learning_rate": 0.2782700434338394,
      "loss": 0.0,
      "num_input_tokens_seen": 3945368,
      "step": 6940
    },
    {
      "epoch": 121.84955752212389,
      "grad_norm": 8.903879802346637e-07,
      "learning_rate": 0.278239496813243,
      "loss": 0.0,
      "num_input_tokens_seen": 3948120,
      "step": 6945
    },
    {
      "epoch": 121.93805309734513,
      "grad_norm": 2.3803333988325903e-06,
      "learning_rate": 0.27820893041650463,
      "loss": 0.0,
      "num_input_tokens_seen": 3951208,
      "step": 6950
    },
    {
      "epoch": 122.01769911504425,
      "grad_norm": 2.3865293314884184e-06,
      "learning_rate": 0.27817834424833804,
      "loss": 0.0,
      "num_input_tokens_seen": 3953560,
      "step": 6955
    },
    {
      "epoch": 122.10619469026548,
      "grad_norm": 3.477048721833853e-06,
      "learning_rate": 0.27814773831345996,
      "loss": 0.0,
      "num_input_tokens_seen": 3955960,
      "step": 6960
    },
    {
      "epoch": 122.19469026548673,
      "grad_norm": 5.5165551202662755e-06,
      "learning_rate": 0.2781171126165902,
      "loss": 0.0,
      "num_input_tokens_seen": 3958984,
      "step": 6965
    },
    {
      "epoch": 122.28318584070796,
      "grad_norm": 2.6355198770033894e-06,
      "learning_rate": 0.2780864671624517,
      "loss": 0.0,
      "num_input_tokens_seen": 3962200,
      "step": 6970
    },
    {
      "epoch": 122.3716814159292,
      "grad_norm": 3.882851615344407e-06,
      "learning_rate": 0.27805580195577034,
      "loss": 0.0,
      "num_input_tokens_seen": 3964696,
      "step": 6975
    },
    {
      "epoch": 122.46017699115045,
      "grad_norm": 3.325897750983131e-06,
      "learning_rate": 0.2780251170012751,
      "loss": 0.0,
      "num_input_tokens_seen": 3967752,
      "step": 6980
    },
    {
      "epoch": 122.54867256637168,
      "grad_norm": 1.4203116052158293e-06,
      "learning_rate": 0.27799441230369787,
      "loss": 0.0,
      "num_input_tokens_seen": 3970712,
      "step": 6985
    },
    {
      "epoch": 122.63716814159292,
      "grad_norm": 4.30909676651936e-06,
      "learning_rate": 0.27796368786777387,
      "loss": 0.0,
      "num_input_tokens_seen": 3974008,
      "step": 6990
    },
    {
      "epoch": 122.72566371681415,
      "grad_norm": 4.551810889097396e-06,
      "learning_rate": 0.277932943698241,
      "loss": 0.0,
      "num_input_tokens_seen": 3976472,
      "step": 6995
    },
    {
      "epoch": 122.8141592920354,
      "grad_norm": 3.155375225105672e-06,
      "learning_rate": 0.2779021797998406,
      "loss": 0.0,
      "num_input_tokens_seen": 3979464,
      "step": 7000
    },
    {
      "epoch": 122.8141592920354,
      "eval_loss": 0.407375693321228,
      "eval_runtime": 0.9338,
      "eval_samples_per_second": 26.773,
      "eval_steps_per_second": 13.922,
      "num_input_tokens_seen": 3979464,
      "step": 7000
    },
    {
      "epoch": 122.90265486725664,
      "grad_norm": 4.651253220799845e-06,
      "learning_rate": 0.2778713961773167,
      "loss": 0.0,
      "num_input_tokens_seen": 3982408,
      "step": 7005
    },
    {
      "epoch": 122.99115044247787,
      "grad_norm": 1.4820664091530489e-06,
      "learning_rate": 0.2778405928354166,
      "loss": 0.0,
      "num_input_tokens_seen": 3985096,
      "step": 7010
    },
    {
      "epoch": 123.070796460177,
      "grad_norm": 1.1768074728024658e-06,
      "learning_rate": 0.27780976977889055,
      "loss": 0.0,
      "num_input_tokens_seen": 3988384,
      "step": 7015
    },
    {
      "epoch": 123.15929203539822,
      "grad_norm": 1.434250634702039e-06,
      "learning_rate": 0.27777892701249185,
      "loss": 0.0,
      "num_input_tokens_seen": 3991360,
      "step": 7020
    },
    {
      "epoch": 123.24778761061947,
      "grad_norm": 1.737721731842612e-06,
      "learning_rate": 0.2777480645409768,
      "loss": 0.0,
      "num_input_tokens_seen": 3994176,
      "step": 7025
    },
    {
      "epoch": 123.33628318584071,
      "grad_norm": 3.4436475289112423e-06,
      "learning_rate": 0.27771718236910486,
      "loss": 0.0,
      "num_input_tokens_seen": 3996816,
      "step": 7030
    },
    {
      "epoch": 123.42477876106194,
      "grad_norm": 1.8211046608485049e-06,
      "learning_rate": 0.27768628050163835,
      "loss": 0.0,
      "num_input_tokens_seen": 3999552,
      "step": 7035
    },
    {
      "epoch": 123.51327433628319,
      "grad_norm": 2.9162717964936746e-06,
      "learning_rate": 0.2776553589433428,
      "loss": 0.0,
      "num_input_tokens_seen": 4002288,
      "step": 7040
    },
    {
      "epoch": 123.60176991150442,
      "grad_norm": 3.795616066781804e-06,
      "learning_rate": 0.27762441769898666,
      "loss": 0.0,
      "num_input_tokens_seen": 4005264,
      "step": 7045
    },
    {
      "epoch": 123.69026548672566,
      "grad_norm": 2.852388888641144e-06,
      "learning_rate": 0.2775934567733415,
      "loss": 0.0,
      "num_input_tokens_seen": 4008352,
      "step": 7050
    },
    {
      "epoch": 123.77876106194691,
      "grad_norm": 3.563500968084554e-06,
      "learning_rate": 0.2775624761711819,
      "loss": 0.0,
      "num_input_tokens_seen": 4010816,
      "step": 7055
    },
    {
      "epoch": 123.86725663716814,
      "grad_norm": 5.068246764494688e-07,
      "learning_rate": 0.2775314758972854,
      "loss": 0.0,
      "num_input_tokens_seen": 4013584,
      "step": 7060
    },
    {
      "epoch": 123.95575221238938,
      "grad_norm": 7.284073490154697e-06,
      "learning_rate": 0.2775004559564327,
      "loss": 0.0,
      "num_input_tokens_seen": 4016400,
      "step": 7065
    },
    {
      "epoch": 124.03539823008849,
      "grad_norm": 2.6385296223452315e-06,
      "learning_rate": 0.2774694163534073,
      "loss": 0.0,
      "num_input_tokens_seen": 4019080,
      "step": 7070
    },
    {
      "epoch": 124.12389380530973,
      "grad_norm": 3.490981498543988e-06,
      "learning_rate": 0.27743835709299614,
      "loss": 0.0,
      "num_input_tokens_seen": 4022488,
      "step": 7075
    },
    {
      "epoch": 124.21238938053098,
      "grad_norm": 1.6377737210859777e-06,
      "learning_rate": 0.2774072781799888,
      "loss": 0.0,
      "num_input_tokens_seen": 4025128,
      "step": 7080
    },
    {
      "epoch": 124.30088495575221,
      "grad_norm": 2.8329607175692217e-06,
      "learning_rate": 0.27737617961917804,
      "loss": 0.0,
      "num_input_tokens_seen": 4027976,
      "step": 7085
    },
    {
      "epoch": 124.38938053097345,
      "grad_norm": 1.1439022955528344e-06,
      "learning_rate": 0.27734506141535964,
      "loss": 0.0,
      "num_input_tokens_seen": 4031496,
      "step": 7090
    },
    {
      "epoch": 124.47787610619469,
      "grad_norm": 4.5753054109809455e-06,
      "learning_rate": 0.2773139235733325,
      "loss": 0.0,
      "num_input_tokens_seen": 4034536,
      "step": 7095
    },
    {
      "epoch": 124.56637168141593,
      "grad_norm": 3.6338060453999788e-06,
      "learning_rate": 0.2772827660978984,
      "loss": 0.0,
      "num_input_tokens_seen": 4037096,
      "step": 7100
    },
    {
      "epoch": 124.65486725663717,
      "grad_norm": 4.07169500249438e-06,
      "learning_rate": 0.27725158899386226,
      "loss": 0.0,
      "num_input_tokens_seen": 4039448,
      "step": 7105
    },
    {
      "epoch": 124.7433628318584,
      "grad_norm": 3.5508842302078847e-06,
      "learning_rate": 0.27722039226603196,
      "loss": 0.0,
      "num_input_tokens_seen": 4042760,
      "step": 7110
    },
    {
      "epoch": 124.83185840707965,
      "grad_norm": 2.80201106761524e-06,
      "learning_rate": 0.2771891759192184,
      "loss": 0.0,
      "num_input_tokens_seen": 4045176,
      "step": 7115
    },
    {
      "epoch": 124.92035398230088,
      "grad_norm": 2.346901055716444e-06,
      "learning_rate": 0.2771579399582355,
      "loss": 0.0,
      "num_input_tokens_seen": 4047864,
      "step": 7120
    },
    {
      "epoch": 125.0,
      "grad_norm": 8.2645084376054e-07,
      "learning_rate": 0.2771266843879004,
      "loss": 0.0,
      "num_input_tokens_seen": 4050248,
      "step": 7125
    },
    {
      "epoch": 125.08849557522124,
      "grad_norm": 3.392387270650943e-06,
      "learning_rate": 0.2770954092130329,
      "loss": 0.0,
      "num_input_tokens_seen": 4053352,
      "step": 7130
    },
    {
      "epoch": 125.17699115044248,
      "grad_norm": 2.370608399360208e-06,
      "learning_rate": 0.27706411443845613,
      "loss": 0.0,
      "num_input_tokens_seen": 4056088,
      "step": 7135
    },
    {
      "epoch": 125.26548672566372,
      "grad_norm": 2.3745396902086213e-06,
      "learning_rate": 0.27703280006899617,
      "loss": 0.0,
      "num_input_tokens_seen": 4058968,
      "step": 7140
    },
    {
      "epoch": 125.35398230088495,
      "grad_norm": 1.6972855974017875e-06,
      "learning_rate": 0.277001466109482,
      "loss": 0.0,
      "num_input_tokens_seen": 4061976,
      "step": 7145
    },
    {
      "epoch": 125.4424778761062,
      "grad_norm": 1.5718990198365645e-06,
      "learning_rate": 0.2769701125647458,
      "loss": 0.0,
      "num_input_tokens_seen": 4064696,
      "step": 7150
    },
    {
      "epoch": 125.53097345132744,
      "grad_norm": 5.397087988967542e-06,
      "learning_rate": 0.27693873943962266,
      "loss": 0.0,
      "num_input_tokens_seen": 4067592,
      "step": 7155
    },
    {
      "epoch": 125.61946902654867,
      "grad_norm": 2.3461698219762184e-06,
      "learning_rate": 0.2769073467389506,
      "loss": 0.0,
      "num_input_tokens_seen": 4070696,
      "step": 7160
    },
    {
      "epoch": 125.70796460176992,
      "grad_norm": 4.251372047292534e-06,
      "learning_rate": 0.2768759344675709,
      "loss": 0.0,
      "num_input_tokens_seen": 4073112,
      "step": 7165
    },
    {
      "epoch": 125.79646017699115,
      "grad_norm": 2.785760671031312e-06,
      "learning_rate": 0.27684450263032767,
      "loss": 0.0,
      "num_input_tokens_seen": 4076072,
      "step": 7170
    },
    {
      "epoch": 125.88495575221239,
      "grad_norm": 9.322918117504742e-07,
      "learning_rate": 0.2768130512320682,
      "loss": 0.0,
      "num_input_tokens_seen": 4078920,
      "step": 7175
    },
    {
      "epoch": 125.97345132743362,
      "grad_norm": 7.3579521995270625e-06,
      "learning_rate": 0.27678158027764244,
      "loss": 0.0,
      "num_input_tokens_seen": 4082120,
      "step": 7180
    },
    {
      "epoch": 126.05309734513274,
      "grad_norm": 1.2371692719170824e-06,
      "learning_rate": 0.27675008977190385,
      "loss": 0.0,
      "num_input_tokens_seen": 4084544,
      "step": 7185
    },
    {
      "epoch": 126.14159292035399,
      "grad_norm": 3.168425564581412e-06,
      "learning_rate": 0.2767185797197086,
      "loss": 0.0,
      "num_input_tokens_seen": 4087728,
      "step": 7190
    },
    {
      "epoch": 126.23008849557522,
      "grad_norm": 1.663713987909432e-06,
      "learning_rate": 0.2766870501259159,
      "loss": 0.0,
      "num_input_tokens_seen": 4090704,
      "step": 7195
    },
    {
      "epoch": 126.31858407079646,
      "grad_norm": 6.459242740675109e-07,
      "learning_rate": 0.276655500995388,
      "loss": 0.0,
      "num_input_tokens_seen": 4093296,
      "step": 7200
    },
    {
      "epoch": 126.31858407079646,
      "eval_loss": 0.40883001685142517,
      "eval_runtime": 0.9342,
      "eval_samples_per_second": 26.762,
      "eval_steps_per_second": 13.916,
      "num_input_tokens_seen": 4093296,
      "step": 7200
    },
    {
      "epoch": 126.40707964601769,
      "grad_norm": 1.406119281455176e-06,
      "learning_rate": 0.27662393233299015,
      "loss": 0.0,
      "num_input_tokens_seen": 4096144,
      "step": 7205
    },
    {
      "epoch": 126.49557522123894,
      "grad_norm": 2.4896346531022573e-06,
      "learning_rate": 0.27659234414359074,
      "loss": 0.0,
      "num_input_tokens_seen": 4098608,
      "step": 7210
    },
    {
      "epoch": 126.58407079646018,
      "grad_norm": 3.908672624675091e-06,
      "learning_rate": 0.27656073643206097,
      "loss": 0.0,
      "num_input_tokens_seen": 4101792,
      "step": 7215
    },
    {
      "epoch": 126.67256637168141,
      "grad_norm": 1.4107804418017622e-06,
      "learning_rate": 0.27652910920327517,
      "loss": 0.0,
      "num_input_tokens_seen": 4105184,
      "step": 7220
    },
    {
      "epoch": 126.76106194690266,
      "grad_norm": 3.983769147453131e-06,
      "learning_rate": 0.2764974624621107,
      "loss": 0.0,
      "num_input_tokens_seen": 4107952,
      "step": 7225
    },
    {
      "epoch": 126.84955752212389,
      "grad_norm": 3.553057467797771e-06,
      "learning_rate": 0.2764657962134479,
      "loss": 0.0,
      "num_input_tokens_seen": 4110480,
      "step": 7230
    },
    {
      "epoch": 126.93805309734513,
      "grad_norm": 5.219312129156606e-07,
      "learning_rate": 0.27643411046217,
      "loss": 0.0,
      "num_input_tokens_seen": 4113280,
      "step": 7235
    },
    {
      "epoch": 127.01769911504425,
      "grad_norm": 1.3866506378690246e-06,
      "learning_rate": 0.27640240521316334,
      "loss": 0.0,
      "num_input_tokens_seen": 4115912,
      "step": 7240
    },
    {
      "epoch": 127.10619469026548,
      "grad_norm": 2.599622121124412e-06,
      "learning_rate": 0.2763706804713174,
      "loss": 0.0,
      "num_input_tokens_seen": 4118840,
      "step": 7245
    },
    {
      "epoch": 127.19469026548673,
      "grad_norm": 4.0783652366371825e-06,
      "learning_rate": 0.2763389362415245,
      "loss": 0.0,
      "num_input_tokens_seen": 4121624,
      "step": 7250
    },
    {
      "epoch": 127.28318584070796,
      "grad_norm": 2.447305178066017e-06,
      "learning_rate": 0.27630717252867987,
      "loss": 0.0,
      "num_input_tokens_seen": 4124152,
      "step": 7255
    },
    {
      "epoch": 127.3716814159292,
      "grad_norm": 5.003168439543515e-07,
      "learning_rate": 0.276275389337682,
      "loss": 0.0,
      "num_input_tokens_seen": 4127016,
      "step": 7260
    },
    {
      "epoch": 127.46017699115045,
      "grad_norm": 2.0464924546104157e-06,
      "learning_rate": 0.2762435866734322,
      "loss": 0.0,
      "num_input_tokens_seen": 4129592,
      "step": 7265
    },
    {
      "epoch": 127.54867256637168,
      "grad_norm": 3.9745498270349344e-07,
      "learning_rate": 0.27621176454083485,
      "loss": 0.0,
      "num_input_tokens_seen": 4132920,
      "step": 7270
    },
    {
      "epoch": 127.63716814159292,
      "grad_norm": 2.276132136103115e-06,
      "learning_rate": 0.2761799229447973,
      "loss": 0.0,
      "num_input_tokens_seen": 4136024,
      "step": 7275
    },
    {
      "epoch": 127.72566371681415,
      "grad_norm": 7.875633855292108e-06,
      "learning_rate": 0.27614806189023006,
      "loss": 0.0,
      "num_input_tokens_seen": 4138840,
      "step": 7280
    },
    {
      "epoch": 127.8141592920354,
      "grad_norm": 2.725835201999871e-06,
      "learning_rate": 0.27611618138204636,
      "loss": 0.0,
      "num_input_tokens_seen": 4142152,
      "step": 7285
    },
    {
      "epoch": 127.90265486725664,
      "grad_norm": 3.948205176129704e-06,
      "learning_rate": 0.2760842814251626,
      "loss": 0.0,
      "num_input_tokens_seen": 4144856,
      "step": 7290
    },
    {
      "epoch": 127.99115044247787,
      "grad_norm": 2.159420091629727e-06,
      "learning_rate": 0.2760523620244982,
      "loss": 0.0,
      "num_input_tokens_seen": 4148088,
      "step": 7295
    },
    {
      "epoch": 128.07079646017698,
      "grad_norm": 4.992456979380222e-06,
      "learning_rate": 0.27602042318497544,
      "loss": 0.0,
      "num_input_tokens_seen": 4150680,
      "step": 7300
    },
    {
      "epoch": 128.15929203539824,
      "grad_norm": 4.14657097280724e-06,
      "learning_rate": 0.2759884649115198,
      "loss": 0.0,
      "num_input_tokens_seen": 4153544,
      "step": 7305
    },
    {
      "epoch": 128.24778761061947,
      "grad_norm": 1.643747850721411e-06,
      "learning_rate": 0.2759564872090596,
      "loss": 0.0,
      "num_input_tokens_seen": 4156136,
      "step": 7310
    },
    {
      "epoch": 128.3362831858407,
      "grad_norm": 2.422985289740609e-06,
      "learning_rate": 0.2759244900825262,
      "loss": 0.0,
      "num_input_tokens_seen": 4158792,
      "step": 7315
    },
    {
      "epoch": 128.42477876106196,
      "grad_norm": 4.1400830923521426e-06,
      "learning_rate": 0.2758924735368539,
      "loss": 0.0,
      "num_input_tokens_seen": 4161672,
      "step": 7320
    },
    {
      "epoch": 128.5132743362832,
      "grad_norm": 3.872611614497146e-06,
      "learning_rate": 0.27586043757698014,
      "loss": 0.0,
      "num_input_tokens_seen": 4164632,
      "step": 7325
    },
    {
      "epoch": 128.60176991150442,
      "grad_norm": 3.5173750347894384e-06,
      "learning_rate": 0.27582838220784534,
      "loss": 0.0,
      "num_input_tokens_seen": 4167768,
      "step": 7330
    },
    {
      "epoch": 128.69026548672565,
      "grad_norm": 1.8281762095284648e-06,
      "learning_rate": 0.27579630743439265,
      "loss": 0.0,
      "num_input_tokens_seen": 4170616,
      "step": 7335
    },
    {
      "epoch": 128.7787610619469,
      "grad_norm": 1.611850393601344e-06,
      "learning_rate": 0.2757642132615686,
      "loss": 0.0,
      "num_input_tokens_seen": 4173320,
      "step": 7340
    },
    {
      "epoch": 128.86725663716814,
      "grad_norm": 2.340627361263614e-06,
      "learning_rate": 0.2757320996943223,
      "loss": 0.0,
      "num_input_tokens_seen": 4176216,
      "step": 7345
    },
    {
      "epoch": 128.95575221238937,
      "grad_norm": 5.252745268080616e-06,
      "learning_rate": 0.2756999667376062,
      "loss": 0.0,
      "num_input_tokens_seen": 4178760,
      "step": 7350
    },
    {
      "epoch": 129.0353982300885,
      "grad_norm": 2.025994035648182e-06,
      "learning_rate": 0.2756678143963756,
      "loss": 0.0,
      "num_input_tokens_seen": 4181184,
      "step": 7355
    },
    {
      "epoch": 129.12389380530973,
      "grad_norm": 3.8053046864661155e-06,
      "learning_rate": 0.2756356426755888,
      "loss": 0.0,
      "num_input_tokens_seen": 4184208,
      "step": 7360
    },
    {
      "epoch": 129.21238938053096,
      "grad_norm": 2.5977421955758473e-06,
      "learning_rate": 0.27560345158020705,
      "loss": 0.0,
      "num_input_tokens_seen": 4187680,
      "step": 7365
    },
    {
      "epoch": 129.30088495575222,
      "grad_norm": 1.7599833199710702e-06,
      "learning_rate": 0.27557124111519465,
      "loss": 0.0,
      "num_input_tokens_seen": 4190400,
      "step": 7370
    },
    {
      "epoch": 129.38938053097345,
      "grad_norm": 1.4635571687904303e-06,
      "learning_rate": 0.27553901128551883,
      "loss": 0.0,
      "num_input_tokens_seen": 4192720,
      "step": 7375
    },
    {
      "epoch": 129.47787610619469,
      "grad_norm": 2.6848951506508456e-07,
      "learning_rate": 0.2755067620961498,
      "loss": 0.0,
      "num_input_tokens_seen": 4195584,
      "step": 7380
    },
    {
      "epoch": 129.56637168141592,
      "grad_norm": 3.4480333397368668e-06,
      "learning_rate": 0.27547449355206094,
      "loss": 0.0,
      "num_input_tokens_seen": 4198048,
      "step": 7385
    },
    {
      "epoch": 129.65486725663717,
      "grad_norm": 1.635221678952803e-06,
      "learning_rate": 0.2754422056582283,
      "loss": 0.0,
      "num_input_tokens_seen": 4201184,
      "step": 7390
    },
    {
      "epoch": 129.7433628318584,
      "grad_norm": 3.1470490284846164e-06,
      "learning_rate": 0.27540989841963115,
      "loss": 0.0,
      "num_input_tokens_seen": 4203936,
      "step": 7395
    },
    {
      "epoch": 129.83185840707964,
      "grad_norm": 1.1596499689403572e-06,
      "learning_rate": 0.27537757184125167,
      "loss": 0.0,
      "num_input_tokens_seen": 4207120,
      "step": 7400
    },
    {
      "epoch": 129.83185840707964,
      "eval_loss": 0.3879455327987671,
      "eval_runtime": 0.926,
      "eval_samples_per_second": 26.998,
      "eval_steps_per_second": 14.039,
      "num_input_tokens_seen": 4207120,
      "step": 7400
    },
    {
      "epoch": 129.9203539823009,
      "grad_norm": 2.054187461908441e-06,
      "learning_rate": 0.275345225928075,
      "loss": 0.0,
      "num_input_tokens_seen": 4210320,
      "step": 7405
    },
    {
      "epoch": 130.0,
      "grad_norm": 4.068767339049373e-06,
      "learning_rate": 0.2753128606850893,
      "loss": 0.0,
      "num_input_tokens_seen": 4212296,
      "step": 7410
    },
    {
      "epoch": 130.08849557522123,
      "grad_norm": 2.931662265837076e-06,
      "learning_rate": 0.2752804761172858,
      "loss": 0.0,
      "num_input_tokens_seen": 4215336,
      "step": 7415
    },
    {
      "epoch": 130.1769911504425,
      "grad_norm": 6.467627713391266e-07,
      "learning_rate": 0.27524807222965836,
      "loss": 0.0,
      "num_input_tokens_seen": 4218280,
      "step": 7420
    },
    {
      "epoch": 130.26548672566372,
      "grad_norm": 1.985734570553177e-06,
      "learning_rate": 0.27521564902720436,
      "loss": 0.0,
      "num_input_tokens_seen": 4221432,
      "step": 7425
    },
    {
      "epoch": 130.35398230088495,
      "grad_norm": 1.315662757406244e-06,
      "learning_rate": 0.2751832065149236,
      "loss": 0.0,
      "num_input_tokens_seen": 4224120,
      "step": 7430
    },
    {
      "epoch": 130.44247787610618,
      "grad_norm": 1.2041650734317955e-06,
      "learning_rate": 0.2751507446978193,
      "loss": 0.0,
      "num_input_tokens_seen": 4226808,
      "step": 7435
    },
    {
      "epoch": 130.53097345132744,
      "grad_norm": 2.2794738470111042e-06,
      "learning_rate": 0.2751182635808974,
      "loss": 0.0,
      "num_input_tokens_seen": 4229880,
      "step": 7440
    },
    {
      "epoch": 130.61946902654867,
      "grad_norm": 3.62906666850904e-06,
      "learning_rate": 0.27508576316916694,
      "loss": 0.0,
      "num_input_tokens_seen": 4232296,
      "step": 7445
    },
    {
      "epoch": 130.7079646017699,
      "grad_norm": 1.011663243843941e-06,
      "learning_rate": 0.2750532434676399,
      "loss": 0.0,
      "num_input_tokens_seen": 4235080,
      "step": 7450
    },
    {
      "epoch": 130.79646017699116,
      "grad_norm": 1.333742261522275e-06,
      "learning_rate": 0.27502070448133115,
      "loss": 0.0,
      "num_input_tokens_seen": 4238360,
      "step": 7455
    },
    {
      "epoch": 130.8849557522124,
      "grad_norm": 2.7180410597793525e-06,
      "learning_rate": 0.2749881462152587,
      "loss": 0.0,
      "num_input_tokens_seen": 4241000,
      "step": 7460
    },
    {
      "epoch": 130.97345132743362,
      "grad_norm": 1.5921714293654077e-06,
      "learning_rate": 0.2749555686744434,
      "loss": 0.0,
      "num_input_tokens_seen": 4244472,
      "step": 7465
    },
    {
      "epoch": 131.05309734513276,
      "grad_norm": 2.906670943048084e-06,
      "learning_rate": 0.2749229718639091,
      "loss": 0.0,
      "num_input_tokens_seen": 4246536,
      "step": 7470
    },
    {
      "epoch": 131.141592920354,
      "grad_norm": 1.947333657881245e-06,
      "learning_rate": 0.27489035578868265,
      "loss": 0.0,
      "num_input_tokens_seen": 4249544,
      "step": 7475
    },
    {
      "epoch": 131.23008849557522,
      "grad_norm": 2.3165996481111506e-06,
      "learning_rate": 0.2748577204537939,
      "loss": 0.0,
      "num_input_tokens_seen": 4252808,
      "step": 7480
    },
    {
      "epoch": 131.31858407079645,
      "grad_norm": 2.2897620510775596e-06,
      "learning_rate": 0.2748250658642756,
      "loss": 0.0,
      "num_input_tokens_seen": 4255528,
      "step": 7485
    },
    {
      "epoch": 131.4070796460177,
      "grad_norm": 2.6887184958468424e-06,
      "learning_rate": 0.2747923920251634,
      "loss": 0.0,
      "num_input_tokens_seen": 4258472,
      "step": 7490
    },
    {
      "epoch": 131.49557522123894,
      "grad_norm": 3.372822448000079e-06,
      "learning_rate": 0.27475969894149627,
      "loss": 0.0,
      "num_input_tokens_seen": 4261464,
      "step": 7495
    },
    {
      "epoch": 131.58407079646017,
      "grad_norm": 3.821893187705427e-06,
      "learning_rate": 0.2747269866183156,
      "loss": 0.0,
      "num_input_tokens_seen": 4263784,
      "step": 7500
    },
    {
      "epoch": 131.67256637168143,
      "grad_norm": 1.6657512560414034e-06,
      "learning_rate": 0.27469425506066625,
      "loss": 0.0,
      "num_input_tokens_seen": 4267032,
      "step": 7505
    },
    {
      "epoch": 131.76106194690266,
      "grad_norm": 2.578732164693065e-06,
      "learning_rate": 0.27466150427359576,
      "loss": 0.0,
      "num_input_tokens_seen": 4269560,
      "step": 7510
    },
    {
      "epoch": 131.8495575221239,
      "grad_norm": 7.081524131535843e-07,
      "learning_rate": 0.2746287342621547,
      "loss": 0.0,
      "num_input_tokens_seen": 4272664,
      "step": 7515
    },
    {
      "epoch": 131.93805309734512,
      "grad_norm": 3.395412477402715e-06,
      "learning_rate": 0.2745959450313966,
      "loss": 0.0,
      "num_input_tokens_seen": 4275608,
      "step": 7520
    },
    {
      "epoch": 132.01769911504425,
      "grad_norm": 1.1865499800478574e-06,
      "learning_rate": 0.27456313658637804,
      "loss": 0.0,
      "num_input_tokens_seen": 4277864,
      "step": 7525
    },
    {
      "epoch": 132.10619469026548,
      "grad_norm": 1.5715024801465916e-06,
      "learning_rate": 0.27453030893215846,
      "loss": 0.0,
      "num_input_tokens_seen": 4280696,
      "step": 7530
    },
    {
      "epoch": 132.1946902654867,
      "grad_norm": 1.074575266102329e-06,
      "learning_rate": 0.2744974620738003,
      "loss": 0.0,
      "num_input_tokens_seen": 4283400,
      "step": 7535
    },
    {
      "epoch": 132.28318584070797,
      "grad_norm": 2.633507619975717e-06,
      "learning_rate": 0.27446459601636897,
      "loss": 0.0,
      "num_input_tokens_seen": 4286648,
      "step": 7540
    },
    {
      "epoch": 132.3716814159292,
      "grad_norm": 1.1460482483016676e-06,
      "learning_rate": 0.2744317107649328,
      "loss": 0.0,
      "num_input_tokens_seen": 4289512,
      "step": 7545
    },
    {
      "epoch": 132.46017699115043,
      "grad_norm": 2.3290574517886853e-06,
      "learning_rate": 0.2743988063245631,
      "loss": 0.0,
      "num_input_tokens_seen": 4292472,
      "step": 7550
    },
    {
      "epoch": 132.5486725663717,
      "grad_norm": 1.6709360579625354e-06,
      "learning_rate": 0.2743658827003342,
      "loss": 0.0,
      "num_input_tokens_seen": 4295480,
      "step": 7555
    },
    {
      "epoch": 132.63716814159292,
      "grad_norm": 1.612057872080186e-06,
      "learning_rate": 0.27433293989732327,
      "loss": 0.0,
      "num_input_tokens_seen": 4298536,
      "step": 7560
    },
    {
      "epoch": 132.72566371681415,
      "grad_norm": 9.594857601769036e-07,
      "learning_rate": 0.27429997792061056,
      "loss": 0.0,
      "num_input_tokens_seen": 4301528,
      "step": 7565
    },
    {
      "epoch": 132.81415929203538,
      "grad_norm": 3.479040344700479e-07,
      "learning_rate": 0.27426699677527927,
      "loss": 0.0,
      "num_input_tokens_seen": 4304312,
      "step": 7570
    },
    {
      "epoch": 132.90265486725664,
      "grad_norm": 2.661088728928007e-06,
      "learning_rate": 0.2742339964664154,
      "loss": 0.0,
      "num_input_tokens_seen": 4306552,
      "step": 7575
    },
    {
      "epoch": 132.99115044247787,
      "grad_norm": 3.380514215223229e-07,
      "learning_rate": 0.274200976999108,
      "loss": 0.0,
      "num_input_tokens_seen": 4309288,
      "step": 7580
    },
    {
      "epoch": 133.07079646017698,
      "grad_norm": 1.1710498029060545e-06,
      "learning_rate": 0.27416793837844916,
      "loss": 0.0,
      "num_input_tokens_seen": 4312024,
      "step": 7585
    },
    {
      "epoch": 133.15929203539824,
      "grad_norm": 1.810665139601042e-06,
      "learning_rate": 0.27413488060953384,
      "loss": 0.0,
      "num_input_tokens_seen": 4314760,
      "step": 7590
    },
    {
      "epoch": 133.24778761061947,
      "grad_norm": 1.6951661336861434e-06,
      "learning_rate": 0.27410180369745996,
      "loss": 0.0,
      "num_input_tokens_seen": 4317400,
      "step": 7595
    },
    {
      "epoch": 133.3362831858407,
      "grad_norm": 1.9393255570321344e-06,
      "learning_rate": 0.27406870764732844,
      "loss": 0.0,
      "num_input_tokens_seen": 4320568,
      "step": 7600
    },
    {
      "epoch": 133.3362831858407,
      "eval_loss": 0.4211263358592987,
      "eval_runtime": 0.9103,
      "eval_samples_per_second": 27.464,
      "eval_steps_per_second": 14.281,
      "num_input_tokens_seen": 4320568,
      "step": 7600
    },
    {
      "epoch": 133.42477876106196,
      "grad_norm": 2.211630317106028e-06,
      "learning_rate": 0.27403559246424297,
      "loss": 0.0,
      "num_input_tokens_seen": 4322968,
      "step": 7605
    },
    {
      "epoch": 133.5132743362832,
      "grad_norm": 3.3390167573088547e-06,
      "learning_rate": 0.2740024581533105,
      "loss": 0.0,
      "num_input_tokens_seen": 4325864,
      "step": 7610
    },
    {
      "epoch": 133.60176991150442,
      "grad_norm": 2.4082764866761863e-06,
      "learning_rate": 0.2739693047196406,
      "loss": 0.0,
      "num_input_tokens_seen": 4328504,
      "step": 7615
    },
    {
      "epoch": 133.69026548672565,
      "grad_norm": 5.271755867397587e-07,
      "learning_rate": 0.27393613216834606,
      "loss": 0.0,
      "num_input_tokens_seen": 4331016,
      "step": 7620
    },
    {
      "epoch": 133.7787610619469,
      "grad_norm": 2.8120541628595674e-06,
      "learning_rate": 0.2739029405045424,
      "loss": 0.0,
      "num_input_tokens_seen": 4333976,
      "step": 7625
    },
    {
      "epoch": 133.86725663716814,
      "grad_norm": 1.2799810065189376e-06,
      "learning_rate": 0.2738697297333483,
      "loss": 0.0,
      "num_input_tokens_seen": 4336936,
      "step": 7630
    },
    {
      "epoch": 133.95575221238937,
      "grad_norm": 6.666624585704994e-07,
      "learning_rate": 0.2738364998598852,
      "loss": 0.0,
      "num_input_tokens_seen": 4340040,
      "step": 7635
    },
    {
      "epoch": 134.0353982300885,
      "grad_norm": 2.05279934561986e-06,
      "learning_rate": 0.27380325088927765,
      "loss": 0.0,
      "num_input_tokens_seen": 4342576,
      "step": 7640
    },
    {
      "epoch": 134.12389380530973,
      "grad_norm": 2.1348430436773924e-06,
      "learning_rate": 0.27376998282665294,
      "loss": 0.0,
      "num_input_tokens_seen": 4345248,
      "step": 7645
    },
    {
      "epoch": 134.21238938053096,
      "grad_norm": 2.7338310246705078e-06,
      "learning_rate": 0.27373669567714154,
      "loss": 0.0,
      "num_input_tokens_seen": 4349056,
      "step": 7650
    },
    {
      "epoch": 134.30088495575222,
      "grad_norm": 4.942755822412437e-07,
      "learning_rate": 0.27370338944587663,
      "loss": 0.0,
      "num_input_tokens_seen": 4351904,
      "step": 7655
    },
    {
      "epoch": 134.38938053097345,
      "grad_norm": 4.829840918318951e-07,
      "learning_rate": 0.27367006413799455,
      "loss": 0.0,
      "num_input_tokens_seen": 4354880,
      "step": 7660
    },
    {
      "epoch": 134.47787610619469,
      "grad_norm": 1.1831547226393013e-06,
      "learning_rate": 0.2736367197586345,
      "loss": 0.0,
      "num_input_tokens_seen": 4357456,
      "step": 7665
    },
    {
      "epoch": 134.56637168141592,
      "grad_norm": 2.951824171759654e-06,
      "learning_rate": 0.2736033563129385,
      "loss": 0.0,
      "num_input_tokens_seen": 4360032,
      "step": 7670
    },
    {
      "epoch": 134.65486725663717,
      "grad_norm": 1.8372145405010087e-06,
      "learning_rate": 0.27356997380605164,
      "loss": 0.0,
      "num_input_tokens_seen": 4362768,
      "step": 7675
    },
    {
      "epoch": 134.7433628318584,
      "grad_norm": 1.0507243359825225e-06,
      "learning_rate": 0.27353657224312194,
      "loss": 0.0,
      "num_input_tokens_seen": 4365536,
      "step": 7680
    },
    {
      "epoch": 134.83185840707964,
      "grad_norm": 1.4001676618136116e-06,
      "learning_rate": 0.2735031516293004,
      "loss": 0.0,
      "num_input_tokens_seen": 4368656,
      "step": 7685
    },
    {
      "epoch": 134.9203539823009,
      "grad_norm": 8.213565365622344e-07,
      "learning_rate": 0.2734697119697408,
      "loss": 0.0,
      "num_input_tokens_seen": 4371408,
      "step": 7690
    },
    {
      "epoch": 135.0,
      "grad_norm": 1.2125043213018216e-05,
      "learning_rate": 0.27343625326959997,
      "loss": 0.0,
      "num_input_tokens_seen": 4373936,
      "step": 7695
    },
    {
      "epoch": 135.08849557522123,
      "grad_norm": 1.4485783594864188e-07,
      "learning_rate": 0.27340277553403775,
      "loss": 0.0,
      "num_input_tokens_seen": 4376576,
      "step": 7700
    },
    {
      "epoch": 135.1769911504425,
      "grad_norm": 5.033311026636511e-06,
      "learning_rate": 0.2733692787682167,
      "loss": 0.0,
      "num_input_tokens_seen": 4379200,
      "step": 7705
    },
    {
      "epoch": 135.26548672566372,
      "grad_norm": 5.990499175823061e-06,
      "learning_rate": 0.27333576297730255,
      "loss": 0.0,
      "num_input_tokens_seen": 4381760,
      "step": 7710
    },
    {
      "epoch": 135.35398230088495,
      "grad_norm": 2.0432455585250864e-06,
      "learning_rate": 0.2733022281664638,
      "loss": 0.0,
      "num_input_tokens_seen": 4384368,
      "step": 7715
    },
    {
      "epoch": 135.44247787610618,
      "grad_norm": 1.1822924079751829e-06,
      "learning_rate": 0.273268674340872,
      "loss": 0.0,
      "num_input_tokens_seen": 4387568,
      "step": 7720
    },
    {
      "epoch": 135.53097345132744,
      "grad_norm": 1.3702095884582377e-06,
      "learning_rate": 0.27323510150570146,
      "loss": 0.0,
      "num_input_tokens_seen": 4390400,
      "step": 7725
    },
    {
      "epoch": 135.61946902654867,
      "grad_norm": 1.8252983409183798e-06,
      "learning_rate": 0.27320150966612966,
      "loss": 0.0,
      "num_input_tokens_seen": 4393760,
      "step": 7730
    },
    {
      "epoch": 135.7079646017699,
      "grad_norm": 4.600169631885365e-06,
      "learning_rate": 0.2731678988273368,
      "loss": 0.0,
      "num_input_tokens_seen": 4397072,
      "step": 7735
    },
    {
      "epoch": 135.79646017699116,
      "grad_norm": 9.859957117441809e-07,
      "learning_rate": 0.27313426899450605,
      "loss": 0.0,
      "num_input_tokens_seen": 4400416,
      "step": 7740
    },
    {
      "epoch": 135.8849557522124,
      "grad_norm": 3.031257165275747e-06,
      "learning_rate": 0.27310062017282366,
      "loss": 0.0,
      "num_input_tokens_seen": 4403344,
      "step": 7745
    },
    {
      "epoch": 135.97345132743362,
      "grad_norm": 2.0704792405012995e-06,
      "learning_rate": 0.2730669523674787,
      "loss": 0.0,
      "num_input_tokens_seen": 4406080,
      "step": 7750
    },
    {
      "epoch": 136.05309734513276,
      "grad_norm": 1.4861944919175585e-06,
      "learning_rate": 0.2730332655836631,
      "loss": 0.0,
      "num_input_tokens_seen": 4408264,
      "step": 7755
    },
    {
      "epoch": 136.141592920354,
      "grad_norm": 2.050362581940135e-06,
      "learning_rate": 0.2729995598265718,
      "loss": 0.0,
      "num_input_tokens_seen": 4410808,
      "step": 7760
    },
    {
      "epoch": 136.23008849557522,
      "grad_norm": 3.0440328373515513e-06,
      "learning_rate": 0.2729658351014027,
      "loss": 0.0,
      "num_input_tokens_seen": 4413320,
      "step": 7765
    },
    {
      "epoch": 136.31858407079645,
      "grad_norm": 2.187155587307643e-06,
      "learning_rate": 0.27293209141335656,
      "loss": 0.0,
      "num_input_tokens_seen": 4415896,
      "step": 7770
    },
    {
      "epoch": 136.4070796460177,
      "grad_norm": 2.1134374037501402e-06,
      "learning_rate": 0.27289832876763703,
      "loss": 0.0,
      "num_input_tokens_seen": 4418248,
      "step": 7775
    },
    {
      "epoch": 136.49557522123894,
      "grad_norm": 2.068020194201381e-06,
      "learning_rate": 0.27286454716945074,
      "loss": 0.0,
      "num_input_tokens_seen": 4421496,
      "step": 7780
    },
    {
      "epoch": 136.58407079646017,
      "grad_norm": 2.855913635357865e-06,
      "learning_rate": 0.27283074662400725,
      "loss": 0.0,
      "num_input_tokens_seen": 4424920,
      "step": 7785
    },
    {
      "epoch": 136.67256637168143,
      "grad_norm": 1.506061266809411e-07,
      "learning_rate": 0.2727969271365191,
      "loss": 0.0,
      "num_input_tokens_seen": 4428104,
      "step": 7790
    },
    {
      "epoch": 136.76106194690266,
      "grad_norm": 5.784546033282822e-07,
      "learning_rate": 0.2727630887122016,
      "loss": 0.0,
      "num_input_tokens_seen": 4431048,
      "step": 7795
    },
    {
      "epoch": 136.8495575221239,
      "grad_norm": 1.1050015018554404e-06,
      "learning_rate": 0.27272923135627314,
      "loss": 0.0,
      "num_input_tokens_seen": 4434056,
      "step": 7800
    },
    {
      "epoch": 136.8495575221239,
      "eval_loss": 0.4221038818359375,
      "eval_runtime": 0.9251,
      "eval_samples_per_second": 27.025,
      "eval_steps_per_second": 14.053,
      "num_input_tokens_seen": 4434056,
      "step": 7800
    },
    {
      "epoch": 136.93805309734512,
      "grad_norm": 3.824180566880386e-06,
      "learning_rate": 0.2726953550739548,
      "loss": 0.0,
      "num_input_tokens_seen": 4437096,
      "step": 7805
    },
    {
      "epoch": 137.01769911504425,
      "grad_norm": 3.354754426254658e-06,
      "learning_rate": 0.27266145987047086,
      "loss": 0.0,
      "num_input_tokens_seen": 4439648,
      "step": 7810
    },
    {
      "epoch": 137.10619469026548,
      "grad_norm": 2.237787157355342e-06,
      "learning_rate": 0.27262754575104836,
      "loss": 0.0,
      "num_input_tokens_seen": 4442304,
      "step": 7815
    },
    {
      "epoch": 137.1946902654867,
      "grad_norm": 3.6302756143413717e-06,
      "learning_rate": 0.27259361272091726,
      "loss": 0.0,
      "num_input_tokens_seen": 4445168,
      "step": 7820
    },
    {
      "epoch": 137.28318584070797,
      "grad_norm": 2.558928144935635e-06,
      "learning_rate": 0.27255966078531046,
      "loss": 0.0,
      "num_input_tokens_seen": 4447856,
      "step": 7825
    },
    {
      "epoch": 137.3716814159292,
      "grad_norm": 1.3781807410850888e-06,
      "learning_rate": 0.2725256899494638,
      "loss": 0.0,
      "num_input_tokens_seen": 4450352,
      "step": 7830
    },
    {
      "epoch": 137.46017699115043,
      "grad_norm": 1.2830137166019995e-06,
      "learning_rate": 0.272491700218616,
      "loss": 0.0,
      "num_input_tokens_seen": 4452832,
      "step": 7835
    },
    {
      "epoch": 137.5486725663717,
      "grad_norm": 8.643453384138411e-07,
      "learning_rate": 0.27245769159800876,
      "loss": 0.0,
      "num_input_tokens_seen": 4456032,
      "step": 7840
    },
    {
      "epoch": 137.63716814159292,
      "grad_norm": 2.3787374630046543e-06,
      "learning_rate": 0.2724236640928865,
      "loss": 0.0,
      "num_input_tokens_seen": 4459216,
      "step": 7845
    },
    {
      "epoch": 137.72566371681415,
      "grad_norm": 8.536243853995984e-07,
      "learning_rate": 0.27238961770849673,
      "loss": 0.0,
      "num_input_tokens_seen": 4462432,
      "step": 7850
    },
    {
      "epoch": 137.81415929203538,
      "grad_norm": 1.7645246543906978e-06,
      "learning_rate": 0.27235555245008997,
      "loss": 0.0,
      "num_input_tokens_seen": 4465264,
      "step": 7855
    },
    {
      "epoch": 137.90265486725664,
      "grad_norm": 3.5011140653296025e-07,
      "learning_rate": 0.2723214683229193,
      "loss": 0.0,
      "num_input_tokens_seen": 4467952,
      "step": 7860
    },
    {
      "epoch": 137.99115044247787,
      "grad_norm": 2.963653741971939e-06,
      "learning_rate": 0.27228736533224107,
      "loss": 0.0,
      "num_input_tokens_seen": 4470720,
      "step": 7865
    },
    {
      "epoch": 138.07079646017698,
      "grad_norm": 1.947083774211933e-06,
      "learning_rate": 0.27225324348331437,
      "loss": 0.0,
      "num_input_tokens_seen": 4473336,
      "step": 7870
    },
    {
      "epoch": 138.15929203539824,
      "grad_norm": 2.6873017304751556e-06,
      "learning_rate": 0.27221910278140116,
      "loss": 0.0,
      "num_input_tokens_seen": 4475688,
      "step": 7875
    },
    {
      "epoch": 138.24778761061947,
      "grad_norm": 8.336010068887845e-07,
      "learning_rate": 0.2721849432317664,
      "loss": 0.0,
      "num_input_tokens_seen": 4478488,
      "step": 7880
    },
    {
      "epoch": 138.3362831858407,
      "grad_norm": 1.235025820278679e-06,
      "learning_rate": 0.2721507648396779,
      "loss": 0.0,
      "num_input_tokens_seen": 4481448,
      "step": 7885
    },
    {
      "epoch": 138.42477876106196,
      "grad_norm": 8.724937856641191e-07,
      "learning_rate": 0.27211656761040653,
      "loss": 0.0,
      "num_input_tokens_seen": 4484328,
      "step": 7890
    },
    {
      "epoch": 138.5132743362832,
      "grad_norm": 7.095731575645914e-07,
      "learning_rate": 0.2720823515492257,
      "loss": 0.0,
      "num_input_tokens_seen": 4487240,
      "step": 7895
    },
    {
      "epoch": 138.60176991150442,
      "grad_norm": 3.375738288013963e-06,
      "learning_rate": 0.27204811666141215,
      "loss": 0.0,
      "num_input_tokens_seen": 4490040,
      "step": 7900
    },
    {
      "epoch": 138.69026548672565,
      "grad_norm": 1.004642740554118e-06,
      "learning_rate": 0.2720138629522452,
      "loss": 0.0,
      "num_input_tokens_seen": 4492888,
      "step": 7905
    },
    {
      "epoch": 138.7787610619469,
      "grad_norm": 2.5842957711574854e-06,
      "learning_rate": 0.2719795904270073,
      "loss": 0.0,
      "num_input_tokens_seen": 4496184,
      "step": 7910
    },
    {
      "epoch": 138.86725663716814,
      "grad_norm": 2.1217927042016527e-06,
      "learning_rate": 0.2719452990909837,
      "loss": 0.0,
      "num_input_tokens_seen": 4499032,
      "step": 7915
    },
    {
      "epoch": 138.95575221238937,
      "grad_norm": 1.5800801520526875e-06,
      "learning_rate": 0.2719109889494625,
      "loss": 0.0,
      "num_input_tokens_seen": 4501736,
      "step": 7920
    },
    {
      "epoch": 139.0353982300885,
      "grad_norm": 1.1334875580359949e-06,
      "learning_rate": 0.27187666000773475,
      "loss": 0.0,
      "num_input_tokens_seen": 4504168,
      "step": 7925
    },
    {
      "epoch": 139.12389380530973,
      "grad_norm": 3.297619059594581e-06,
      "learning_rate": 0.2718423122710944,
      "loss": 0.0,
      "num_input_tokens_seen": 4507304,
      "step": 7930
    },
    {
      "epoch": 139.21238938053096,
      "grad_norm": 4.119043524042354e-07,
      "learning_rate": 0.2718079457448384,
      "loss": 0.0,
      "num_input_tokens_seen": 4509976,
      "step": 7935
    },
    {
      "epoch": 139.30088495575222,
      "grad_norm": 4.211226951156277e-06,
      "learning_rate": 0.27177356043426637,
      "loss": 0.0,
      "num_input_tokens_seen": 4512648,
      "step": 7940
    },
    {
      "epoch": 139.38938053097345,
      "grad_norm": 2.8310605557635427e-06,
      "learning_rate": 0.27173915634468104,
      "loss": 0.0,
      "num_input_tokens_seen": 4515400,
      "step": 7945
    },
    {
      "epoch": 139.47787610619469,
      "grad_norm": 1.5799756738488213e-06,
      "learning_rate": 0.27170473348138796,
      "loss": 0.0,
      "num_input_tokens_seen": 4518744,
      "step": 7950
    },
    {
      "epoch": 139.56637168141592,
      "grad_norm": 2.1102755454194266e-06,
      "learning_rate": 0.27167029184969554,
      "loss": 0.0,
      "num_input_tokens_seen": 4521736,
      "step": 7955
    },
    {
      "epoch": 139.65486725663717,
      "grad_norm": 3.2922637274168665e-06,
      "learning_rate": 0.27163583145491504,
      "loss": 0.0,
      "num_input_tokens_seen": 4524632,
      "step": 7960
    },
    {
      "epoch": 139.7433628318584,
      "grad_norm": 1.495198262091435e-06,
      "learning_rate": 0.2716013523023608,
      "loss": 0.0,
      "num_input_tokens_seen": 4527400,
      "step": 7965
    },
    {
      "epoch": 139.83185840707964,
      "grad_norm": 2.209550530096749e-06,
      "learning_rate": 0.27156685439734995,
      "loss": 0.0,
      "num_input_tokens_seen": 4530216,
      "step": 7970
    },
    {
      "epoch": 139.9203539823009,
      "grad_norm": 3.5334937820152845e-06,
      "learning_rate": 0.2715323377452024,
      "loss": 0.0,
      "num_input_tokens_seen": 4533128,
      "step": 7975
    },
    {
      "epoch": 140.0,
      "grad_norm": 1.4905256762176577e-07,
      "learning_rate": 0.2714978023512411,
      "loss": 0.0,
      "num_input_tokens_seen": 4535568,
      "step": 7980
    },
    {
      "epoch": 140.08849557522123,
      "grad_norm": 1.4943009318812983e-06,
      "learning_rate": 0.2714632482207918,
      "loss": 0.0,
      "num_input_tokens_seen": 4538912,
      "step": 7985
    },
    {
      "epoch": 140.1769911504425,
      "grad_norm": 1.144013936027477e-06,
      "learning_rate": 0.2714286753591833,
      "loss": 0.0,
      "num_input_tokens_seen": 4541728,
      "step": 7990
    },
    {
      "epoch": 140.26548672566372,
      "grad_norm": 7.170142453105655e-07,
      "learning_rate": 0.27139408377174706,
      "loss": 0.0,
      "num_input_tokens_seen": 4545056,
      "step": 7995
    },
    {
      "epoch": 140.35398230088495,
      "grad_norm": 3.2837024264154024e-06,
      "learning_rate": 0.27135947346381756,
      "loss": 0.0,
      "num_input_tokens_seen": 4547840,
      "step": 8000
    },
    {
      "epoch": 140.35398230088495,
      "eval_loss": 0.425972580909729,
      "eval_runtime": 0.9271,
      "eval_samples_per_second": 26.966,
      "eval_steps_per_second": 14.023,
      "num_input_tokens_seen": 4547840,
      "step": 8000
    },
    {
      "epoch": 140.44247787610618,
      "grad_norm": 1.8611339100971236e-06,
      "learning_rate": 0.2713248444407322,
      "loss": 0.0,
      "num_input_tokens_seen": 4550544,
      "step": 8005
    },
    {
      "epoch": 140.53097345132744,
      "grad_norm": 2.380748810537625e-06,
      "learning_rate": 0.27129019670783106,
      "loss": 0.0,
      "num_input_tokens_seen": 4553536,
      "step": 8010
    },
    {
      "epoch": 140.61946902654867,
      "grad_norm": 8.501598927068699e-07,
      "learning_rate": 0.27125553027045746,
      "loss": 0.0,
      "num_input_tokens_seen": 4556096,
      "step": 8015
    },
    {
      "epoch": 140.7079646017699,
      "grad_norm": 9.299871521761816e-07,
      "learning_rate": 0.2712208451339572,
      "loss": 0.0,
      "num_input_tokens_seen": 4558784,
      "step": 8020
    },
    {
      "epoch": 140.79646017699116,
      "grad_norm": 1.6729885601307615e-06,
      "learning_rate": 0.27118614130367935,
      "loss": 0.0,
      "num_input_tokens_seen": 4561232,
      "step": 8025
    },
    {
      "epoch": 140.8849557522124,
      "grad_norm": 9.672380656411406e-07,
      "learning_rate": 0.2711514187849756,
      "loss": 0.0,
      "num_input_tokens_seen": 4564544,
      "step": 8030
    },
    {
      "epoch": 140.97345132743362,
      "grad_norm": 2.4918872441048734e-06,
      "learning_rate": 0.27111667758320057,
      "loss": 0.0,
      "num_input_tokens_seen": 4567280,
      "step": 8035
    },
    {
      "epoch": 141.05309734513276,
      "grad_norm": 2.13628072742722e-06,
      "learning_rate": 0.27108191770371176,
      "loss": 0.0,
      "num_input_tokens_seen": 4569480,
      "step": 8040
    },
    {
      "epoch": 141.141592920354,
      "grad_norm": 2.6584909846860683e-06,
      "learning_rate": 0.2710471391518697,
      "loss": 0.0,
      "num_input_tokens_seen": 4572360,
      "step": 8045
    },
    {
      "epoch": 141.23008849557522,
      "grad_norm": 3.3419446481275372e-06,
      "learning_rate": 0.2710123419330375,
      "loss": 0.0,
      "num_input_tokens_seen": 4575320,
      "step": 8050
    },
    {
      "epoch": 141.31858407079645,
      "grad_norm": 3.1182862585410476e-06,
      "learning_rate": 0.2709775260525816,
      "loss": 0.0,
      "num_input_tokens_seen": 4578104,
      "step": 8055
    },
    {
      "epoch": 141.4070796460177,
      "grad_norm": 8.383607337236754e-07,
      "learning_rate": 0.27094269151587075,
      "loss": 0.0,
      "num_input_tokens_seen": 4581096,
      "step": 8060
    },
    {
      "epoch": 141.49557522123894,
      "grad_norm": 7.536648354289355e-07,
      "learning_rate": 0.27090783832827703,
      "loss": 0.0,
      "num_input_tokens_seen": 4584184,
      "step": 8065
    },
    {
      "epoch": 141.58407079646017,
      "grad_norm": 2.3588422664033715e-06,
      "learning_rate": 0.2708729664951753,
      "loss": 0.0,
      "num_input_tokens_seen": 4587320,
      "step": 8070
    },
    {
      "epoch": 141.67256637168143,
      "grad_norm": 5.175935484658112e-07,
      "learning_rate": 0.27083807602194304,
      "loss": 0.0,
      "num_input_tokens_seen": 4589944,
      "step": 8075
    },
    {
      "epoch": 141.76106194690266,
      "grad_norm": 1.3177993878343841e-06,
      "learning_rate": 0.270803166913961,
      "loss": 0.0,
      "num_input_tokens_seen": 4592568,
      "step": 8080
    },
    {
      "epoch": 141.8495575221239,
      "grad_norm": 5.873963573321817e-07,
      "learning_rate": 0.27076823917661247,
      "loss": 0.0,
      "num_input_tokens_seen": 4595848,
      "step": 8085
    },
    {
      "epoch": 141.93805309734512,
      "grad_norm": 2.1653681869793218e-06,
      "learning_rate": 0.2707332928152838,
      "loss": 0.0,
      "num_input_tokens_seen": 4598648,
      "step": 8090
    },
    {
      "epoch": 142.01769911504425,
      "grad_norm": 5.08426694523223e-07,
      "learning_rate": 0.2706983278353641,
      "loss": 0.0,
      "num_input_tokens_seen": 4601320,
      "step": 8095
    },
    {
      "epoch": 142.10619469026548,
      "grad_norm": 3.683581553559634e-06,
      "learning_rate": 0.27066334424224553,
      "loss": 0.0,
      "num_input_tokens_seen": 4604488,
      "step": 8100
    },
    {
      "epoch": 142.1946902654867,
      "grad_norm": 5.387262831391126e-07,
      "learning_rate": 0.27062834204132297,
      "loss": 0.0,
      "num_input_tokens_seen": 4607032,
      "step": 8105
    },
    {
      "epoch": 142.28318584070797,
      "grad_norm": 1.8587423937788117e-06,
      "learning_rate": 0.27059332123799407,
      "loss": 0.0,
      "num_input_tokens_seen": 4609544,
      "step": 8110
    },
    {
      "epoch": 142.3716814159292,
      "grad_norm": 8.367191526303941e-07,
      "learning_rate": 0.27055828183765956,
      "loss": 0.0,
      "num_input_tokens_seen": 4612680,
      "step": 8115
    },
    {
      "epoch": 142.46017699115043,
      "grad_norm": 1.8290587604496977e-06,
      "learning_rate": 0.270523223845723,
      "loss": 0.0,
      "num_input_tokens_seen": 4615560,
      "step": 8120
    },
    {
      "epoch": 142.5486725663717,
      "grad_norm": 1.0620278771966696e-06,
      "learning_rate": 0.2704881472675907,
      "loss": 0.0,
      "num_input_tokens_seen": 4618184,
      "step": 8125
    },
    {
      "epoch": 142.63716814159292,
      "grad_norm": 3.884665147779742e-06,
      "learning_rate": 0.270453052108672,
      "loss": 0.0,
      "num_input_tokens_seen": 4620632,
      "step": 8130
    },
    {
      "epoch": 142.72566371681415,
      "grad_norm": 1.165385356216575e-06,
      "learning_rate": 0.2704179383743789,
      "loss": 0.0,
      "num_input_tokens_seen": 4623784,
      "step": 8135
    },
    {
      "epoch": 142.81415929203538,
      "grad_norm": 7.183800221355341e-07,
      "learning_rate": 0.27038280607012644,
      "loss": 0.0,
      "num_input_tokens_seen": 4627128,
      "step": 8140
    },
    {
      "epoch": 142.90265486725664,
      "grad_norm": 1.1918377822439652e-06,
      "learning_rate": 0.27034765520133247,
      "loss": 0.0,
      "num_input_tokens_seen": 4630616,
      "step": 8145
    },
    {
      "epoch": 142.99115044247787,
      "grad_norm": 1.988286612686352e-06,
      "learning_rate": 0.2703124857734177,
      "loss": 0.0,
      "num_input_tokens_seen": 4633144,
      "step": 8150
    },
    {
      "epoch": 143.07079646017698,
      "grad_norm": 1.168030053122493e-06,
      "learning_rate": 0.27027729779180565,
      "loss": 0.0,
      "num_input_tokens_seen": 4635648,
      "step": 8155
    },
    {
      "epoch": 143.15929203539824,
      "grad_norm": 1.2617624634003732e-06,
      "learning_rate": 0.27024209126192283,
      "loss": 0.0,
      "num_input_tokens_seen": 4638848,
      "step": 8160
    },
    {
      "epoch": 143.24778761061947,
      "grad_norm": 3.2698553695809096e-06,
      "learning_rate": 0.2702068661891984,
      "loss": 0.0,
      "num_input_tokens_seen": 4641744,
      "step": 8165
    },
    {
      "epoch": 143.3362831858407,
      "grad_norm": 2.126940472635397e-07,
      "learning_rate": 0.2701716225790647,
      "loss": 0.0,
      "num_input_tokens_seen": 4644528,
      "step": 8170
    },
    {
      "epoch": 143.42477876106196,
      "grad_norm": 1.668901632001507e-06,
      "learning_rate": 0.27013636043695655,
      "loss": 0.0,
      "num_input_tokens_seen": 4647616,
      "step": 8175
    },
    {
      "epoch": 143.5132743362832,
      "grad_norm": 1.188125679618679e-06,
      "learning_rate": 0.27010107976831194,
      "loss": 0.0,
      "num_input_tokens_seen": 4650352,
      "step": 8180
    },
    {
      "epoch": 143.60176991150442,
      "grad_norm": 3.431245431784191e-06,
      "learning_rate": 0.2700657805785715,
      "loss": 0.0,
      "num_input_tokens_seen": 4652992,
      "step": 8185
    },
    {
      "epoch": 143.69026548672565,
      "grad_norm": 1.2613695616892073e-06,
      "learning_rate": 0.2700304628731789,
      "loss": 0.0,
      "num_input_tokens_seen": 4655984,
      "step": 8190
    },
    {
      "epoch": 143.7787610619469,
      "grad_norm": 2.2148542484501377e-06,
      "learning_rate": 0.26999512665758046,
      "loss": 0.0,
      "num_input_tokens_seen": 4659088,
      "step": 8195
    },
    {
      "epoch": 143.86725663716814,
      "grad_norm": 1.7765278244041838e-06,
      "learning_rate": 0.2699597719372256,
      "loss": 0.0,
      "num_input_tokens_seen": 4662192,
      "step": 8200
    },
    {
      "epoch": 143.86725663716814,
      "eval_loss": 0.43572017550468445,
      "eval_runtime": 0.94,
      "eval_samples_per_second": 26.595,
      "eval_steps_per_second": 13.829,
      "num_input_tokens_seen": 4662192,
      "step": 8200
    },
    {
      "epoch": 143.95575221238937,
      "grad_norm": 2.297618266311474e-06,
      "learning_rate": 0.26992439871756635,
      "loss": 0.0,
      "num_input_tokens_seen": 4664672,
      "step": 8205
    },
    {
      "epoch": 144.0353982300885,
      "grad_norm": 9.57463726081187e-07,
      "learning_rate": 0.2698890070040578,
      "loss": 0.0,
      "num_input_tokens_seen": 4666768,
      "step": 8210
    },
    {
      "epoch": 144.12389380530973,
      "grad_norm": 1.433054080735019e-06,
      "learning_rate": 0.2698535968021577,
      "loss": 0.0,
      "num_input_tokens_seen": 4669056,
      "step": 8215
    },
    {
      "epoch": 144.21238938053096,
      "grad_norm": 4.112759597774129e-06,
      "learning_rate": 0.26981816811732684,
      "loss": 0.0,
      "num_input_tokens_seen": 4672064,
      "step": 8220
    },
    {
      "epoch": 144.30088495575222,
      "grad_norm": 1.440706341782061e-06,
      "learning_rate": 0.26978272095502875,
      "loss": 0.0,
      "num_input_tokens_seen": 4675344,
      "step": 8225
    },
    {
      "epoch": 144.38938053097345,
      "grad_norm": 1.855520963545132e-06,
      "learning_rate": 0.26974725532072974,
      "loss": 0.0,
      "num_input_tokens_seen": 4678432,
      "step": 8230
    },
    {
      "epoch": 144.47787610619469,
      "grad_norm": 1.926011236719205e-06,
      "learning_rate": 0.26971177121989914,
      "loss": 0.0,
      "num_input_tokens_seen": 4681344,
      "step": 8235
    },
    {
      "epoch": 144.56637168141592,
      "grad_norm": 1.1347461850164109e-06,
      "learning_rate": 0.2696762686580091,
      "loss": 0.0,
      "num_input_tokens_seen": 4684192,
      "step": 8240
    },
    {
      "epoch": 144.65486725663717,
      "grad_norm": 7.20717650892766e-07,
      "learning_rate": 0.26964074764053436,
      "loss": 0.0,
      "num_input_tokens_seen": 4686688,
      "step": 8245
    },
    {
      "epoch": 144.7433628318584,
      "grad_norm": 2.19362800635281e-06,
      "learning_rate": 0.2696052081729529,
      "loss": 0.0,
      "num_input_tokens_seen": 4689776,
      "step": 8250
    },
    {
      "epoch": 144.83185840707964,
      "grad_norm": 1.6048933275669697e-06,
      "learning_rate": 0.2695696502607453,
      "loss": 0.0,
      "num_input_tokens_seen": 4692384,
      "step": 8255
    },
    {
      "epoch": 144.9203539823009,
      "grad_norm": 1.9305489331600256e-06,
      "learning_rate": 0.26953407390939504,
      "loss": 0.0,
      "num_input_tokens_seen": 4695744,
      "step": 8260
    },
    {
      "epoch": 145.0,
      "grad_norm": 5.580675974670157e-07,
      "learning_rate": 0.26949847912438835,
      "loss": 0.0,
      "num_input_tokens_seen": 4698160,
      "step": 8265
    },
    {
      "epoch": 145.08849557522123,
      "grad_norm": 2.5248259589716326e-06,
      "learning_rate": 0.26946286591121454,
      "loss": 0.0,
      "num_input_tokens_seen": 4701280,
      "step": 8270
    },
    {
      "epoch": 145.1769911504425,
      "grad_norm": 6.652600745837844e-07,
      "learning_rate": 0.2694272342753655,
      "loss": 0.0,
      "num_input_tokens_seen": 4703744,
      "step": 8275
    },
    {
      "epoch": 145.26548672566372,
      "grad_norm": 2.722687781897548e-07,
      "learning_rate": 0.26939158422233617,
      "loss": 0.0,
      "num_input_tokens_seen": 4706736,
      "step": 8280
    },
    {
      "epoch": 145.35398230088495,
      "grad_norm": 1.662066210883495e-06,
      "learning_rate": 0.26935591575762413,
      "loss": 0.0,
      "num_input_tokens_seen": 4709312,
      "step": 8285
    },
    {
      "epoch": 145.44247787610618,
      "grad_norm": 7.525932232965715e-07,
      "learning_rate": 0.26932022888672996,
      "loss": 0.0,
      "num_input_tokens_seen": 4712288,
      "step": 8290
    },
    {
      "epoch": 145.53097345132744,
      "grad_norm": 9.225707913174119e-07,
      "learning_rate": 0.26928452361515703,
      "loss": 0.0,
      "num_input_tokens_seen": 4715216,
      "step": 8295
    },
    {
      "epoch": 145.61946902654867,
      "grad_norm": 1.267751031264197e-06,
      "learning_rate": 0.26924879994841155,
      "loss": 0.0,
      "num_input_tokens_seen": 4717840,
      "step": 8300
    },
    {
      "epoch": 145.7079646017699,
      "grad_norm": 7.996579824975925e-07,
      "learning_rate": 0.2692130578920025,
      "loss": 0.0,
      "num_input_tokens_seen": 4720560,
      "step": 8305
    },
    {
      "epoch": 145.79646017699116,
      "grad_norm": 2.0890181531285634e-06,
      "learning_rate": 0.26917729745144187,
      "loss": 0.0,
      "num_input_tokens_seen": 4723824,
      "step": 8310
    },
    {
      "epoch": 145.8849557522124,
      "grad_norm": 3.611046395235462e-06,
      "learning_rate": 0.2691415186322443,
      "loss": 0.0,
      "num_input_tokens_seen": 4726704,
      "step": 8315
    },
    {
      "epoch": 145.97345132743362,
      "grad_norm": 9.63198203862703e-07,
      "learning_rate": 0.2691057214399273,
      "loss": 0.0,
      "num_input_tokens_seen": 4729616,
      "step": 8320
    },
    {
      "epoch": 146.05309734513276,
      "grad_norm": 1.2794440635843785e-06,
      "learning_rate": 0.2690699058800113,
      "loss": 0.0,
      "num_input_tokens_seen": 4731848,
      "step": 8325
    },
    {
      "epoch": 146.141592920354,
      "grad_norm": 2.506681937575195e-07,
      "learning_rate": 0.2690340719580194,
      "loss": 0.0,
      "num_input_tokens_seen": 4734840,
      "step": 8330
    },
    {
      "epoch": 146.23008849557522,
      "grad_norm": 1.680533159742481e-06,
      "learning_rate": 0.2689982196794778,
      "loss": 0.0,
      "num_input_tokens_seen": 4737624,
      "step": 8335
    },
    {
      "epoch": 146.31858407079645,
      "grad_norm": 4.36949733284564e-07,
      "learning_rate": 0.2689623490499153,
      "loss": 0.0,
      "num_input_tokens_seen": 4740712,
      "step": 8340
    },
    {
      "epoch": 146.4070796460177,
      "grad_norm": 1.5520868146268185e-06,
      "learning_rate": 0.2689264600748636,
      "loss": 0.0,
      "num_input_tokens_seen": 4743656,
      "step": 8345
    },
    {
      "epoch": 146.49557522123894,
      "grad_norm": 1.3675189620698802e-06,
      "learning_rate": 0.26889055275985724,
      "loss": 0.0,
      "num_input_tokens_seen": 4746168,
      "step": 8350
    },
    {
      "epoch": 146.58407079646017,
      "grad_norm": 1.6949893506534863e-06,
      "learning_rate": 0.2688546271104335,
      "loss": 0.0,
      "num_input_tokens_seen": 4748712,
      "step": 8355
    },
    {
      "epoch": 146.67256637168143,
      "grad_norm": 1.5522243757004617e-06,
      "learning_rate": 0.26881868313213275,
      "loss": 0.0,
      "num_input_tokens_seen": 4751848,
      "step": 8360
    },
    {
      "epoch": 146.76106194690266,
      "grad_norm": 9.136535368270415e-07,
      "learning_rate": 0.2687827208304978,
      "loss": 0.0,
      "num_input_tokens_seen": 4754888,
      "step": 8365
    },
    {
      "epoch": 146.8495575221239,
      "grad_norm": 7.618065751557879e-07,
      "learning_rate": 0.26874674021107464,
      "loss": 0.0,
      "num_input_tokens_seen": 4757928,
      "step": 8370
    },
    {
      "epoch": 146.93805309734512,
      "grad_norm": 8.28911765893281e-07,
      "learning_rate": 0.2687107412794118,
      "loss": 0.0,
      "num_input_tokens_seen": 4760760,
      "step": 8375
    },
    {
      "epoch": 147.01769911504425,
      "grad_norm": 2.386605501669692e-06,
      "learning_rate": 0.26867472404106096,
      "loss": 0.0,
      "num_input_tokens_seen": 4763120,
      "step": 8380
    },
    {
      "epoch": 147.10619469026548,
      "grad_norm": 1.2971486285096034e-06,
      "learning_rate": 0.26863868850157624,
      "loss": 0.0,
      "num_input_tokens_seen": 4765792,
      "step": 8385
    },
    {
      "epoch": 147.1946902654867,
      "grad_norm": 1.7380691588186892e-06,
      "learning_rate": 0.26860263466651485,
      "loss": 0.0,
      "num_input_tokens_seen": 4768768,
      "step": 8390
    },
    {
      "epoch": 147.28318584070797,
      "grad_norm": 3.438392923271749e-06,
      "learning_rate": 0.26856656254143674,
      "loss": 0.0,
      "num_input_tokens_seen": 4771664,
      "step": 8395
    },
    {
      "epoch": 147.3716814159292,
      "grad_norm": 1.3516093986254418e-06,
      "learning_rate": 0.2685304721319047,
      "loss": 0.0,
      "num_input_tokens_seen": 4774160,
      "step": 8400
    },
    {
      "epoch": 147.3716814159292,
      "eval_loss": 0.4341481328010559,
      "eval_runtime": 0.9103,
      "eval_samples_per_second": 27.464,
      "eval_steps_per_second": 14.281,
      "num_input_tokens_seen": 4774160,
      "step": 8400
    },
    {
      "epoch": 147.46017699115043,
      "grad_norm": 8.994955464913801e-07,
      "learning_rate": 0.2684943634434843,
      "loss": 0.0,
      "num_input_tokens_seen": 4776896,
      "step": 8405
    },
    {
      "epoch": 147.5486725663717,
      "grad_norm": 1.2780436691173236e-06,
      "learning_rate": 0.268458236481744,
      "loss": 0.0,
      "num_input_tokens_seen": 4779664,
      "step": 8410
    },
    {
      "epoch": 147.63716814159292,
      "grad_norm": 1.9733815861400217e-06,
      "learning_rate": 0.2684220912522549,
      "loss": 0.0,
      "num_input_tokens_seen": 4782400,
      "step": 8415
    },
    {
      "epoch": 147.72566371681415,
      "grad_norm": 2.12465647564386e-06,
      "learning_rate": 0.2683859277605913,
      "loss": 0.0,
      "num_input_tokens_seen": 4785120,
      "step": 8420
    },
    {
      "epoch": 147.81415929203538,
      "grad_norm": 7.23618370557233e-07,
      "learning_rate": 0.2683497460123298,
      "loss": 0.0,
      "num_input_tokens_seen": 4787952,
      "step": 8425
    },
    {
      "epoch": 147.90265486725664,
      "grad_norm": 2.522502882129629e-06,
      "learning_rate": 0.26831354601305013,
      "loss": 0.0,
      "num_input_tokens_seen": 4790736,
      "step": 8430
    },
    {
      "epoch": 147.99115044247787,
      "grad_norm": 6.726561423420208e-07,
      "learning_rate": 0.26827732776833496,
      "loss": 0.0,
      "num_input_tokens_seen": 4794000,
      "step": 8435
    },
    {
      "epoch": 148.07079646017698,
      "grad_norm": 2.220075202785665e-06,
      "learning_rate": 0.26824109128376944,
      "loss": 0.0,
      "num_input_tokens_seen": 4796384,
      "step": 8440
    },
    {
      "epoch": 148.15929203539824,
      "grad_norm": 1.4330845488075283e-06,
      "learning_rate": 0.2682048365649417,
      "loss": 0.0,
      "num_input_tokens_seen": 4799536,
      "step": 8445
    },
    {
      "epoch": 148.24778761061947,
      "grad_norm": 1.7297571730523487e-06,
      "learning_rate": 0.2681685636174428,
      "loss": 0.0,
      "num_input_tokens_seen": 4801936,
      "step": 8450
    },
    {
      "epoch": 148.3362831858407,
      "grad_norm": 9.141668328993546e-07,
      "learning_rate": 0.2681322724468663,
      "loss": 0.0,
      "num_input_tokens_seen": 4804416,
      "step": 8455
    },
    {
      "epoch": 148.42477876106196,
      "grad_norm": 5.987529334561259e-07,
      "learning_rate": 0.2680959630588089,
      "loss": 0.0,
      "num_input_tokens_seen": 4807184,
      "step": 8460
    },
    {
      "epoch": 148.5132743362832,
      "grad_norm": 2.6630442562236567e-07,
      "learning_rate": 0.26805963545886985,
      "loss": 0.0,
      "num_input_tokens_seen": 4809936,
      "step": 8465
    },
    {
      "epoch": 148.60176991150442,
      "grad_norm": 1.9067272205575136e-06,
      "learning_rate": 0.26802328965265143,
      "loss": 0.0,
      "num_input_tokens_seen": 4812720,
      "step": 8470
    },
    {
      "epoch": 148.69026548672565,
      "grad_norm": 2.3890640932222595e-06,
      "learning_rate": 0.26798692564575854,
      "loss": 0.0,
      "num_input_tokens_seen": 4815968,
      "step": 8475
    },
    {
      "epoch": 148.7787610619469,
      "grad_norm": 1.1290983366052387e-06,
      "learning_rate": 0.26795054344379904,
      "loss": 0.0,
      "num_input_tokens_seen": 4818896,
      "step": 8480
    },
    {
      "epoch": 148.86725663716814,
      "grad_norm": 1.9242877442593453e-06,
      "learning_rate": 0.2679141430523835,
      "loss": 0.0,
      "num_input_tokens_seen": 4821952,
      "step": 8485
    },
    {
      "epoch": 148.95575221238937,
      "grad_norm": 1.814750589801406e-06,
      "learning_rate": 0.2678777244771252,
      "loss": 0.0,
      "num_input_tokens_seen": 4824848,
      "step": 8490
    },
    {
      "epoch": 149.0353982300885,
      "grad_norm": 9.21267599096609e-07,
      "learning_rate": 0.2678412877236405,
      "loss": 0.0,
      "num_input_tokens_seen": 4827704,
      "step": 8495
    },
    {
      "epoch": 149.12389380530973,
      "grad_norm": 1.1250490388192702e-06,
      "learning_rate": 0.2678048327975484,
      "loss": 0.0,
      "num_input_tokens_seen": 4830552,
      "step": 8500
    },
    {
      "epoch": 149.21238938053096,
      "grad_norm": 2.1502512481674785e-07,
      "learning_rate": 0.2677683597044706,
      "loss": 0.0,
      "num_input_tokens_seen": 4833368,
      "step": 8505
    },
    {
      "epoch": 149.30088495575222,
      "grad_norm": 6.522372473227733e-07,
      "learning_rate": 0.2677318684500318,
      "loss": 0.0,
      "num_input_tokens_seen": 4836584,
      "step": 8510
    },
    {
      "epoch": 149.38938053097345,
      "grad_norm": 1.949638999576564e-06,
      "learning_rate": 0.2676953590398593,
      "loss": 0.0,
      "num_input_tokens_seen": 4839592,
      "step": 8515
    },
    {
      "epoch": 149.47787610619469,
      "grad_norm": 7.812561193532019e-07,
      "learning_rate": 0.2676588314795834,
      "loss": 0.0,
      "num_input_tokens_seen": 4842328,
      "step": 8520
    },
    {
      "epoch": 149.56637168141592,
      "grad_norm": 1.173032273982244e-06,
      "learning_rate": 0.26762228577483715,
      "loss": 0.0,
      "num_input_tokens_seen": 4844728,
      "step": 8525
    },
    {
      "epoch": 149.65486725663717,
      "grad_norm": 7.457414312739274e-07,
      "learning_rate": 0.2675857219312563,
      "loss": 0.0,
      "num_input_tokens_seen": 4847288,
      "step": 8530
    },
    {
      "epoch": 149.7433628318584,
      "grad_norm": 9.66074026109709e-07,
      "learning_rate": 0.2675491399544794,
      "loss": 0.0,
      "num_input_tokens_seen": 4850424,
      "step": 8535
    },
    {
      "epoch": 149.83185840707964,
      "grad_norm": 1.872904931587982e-06,
      "learning_rate": 0.2675125398501479,
      "loss": 0.0,
      "num_input_tokens_seen": 4853496,
      "step": 8540
    },
    {
      "epoch": 149.9203539823009,
      "grad_norm": 3.0786708293817355e-07,
      "learning_rate": 0.26747592162390604,
      "loss": 0.0,
      "num_input_tokens_seen": 4856296,
      "step": 8545
    },
    {
      "epoch": 150.0,
      "grad_norm": 1.2924523389301612e-06,
      "learning_rate": 0.26743928528140076,
      "loss": 0.0,
      "num_input_tokens_seen": 4859016,
      "step": 8550
    },
    {
      "epoch": 150.08849557522123,
      "grad_norm": 9.03906936855492e-07,
      "learning_rate": 0.26740263082828186,
      "loss": 0.0,
      "num_input_tokens_seen": 4861608,
      "step": 8555
    },
    {
      "epoch": 150.1769911504425,
      "grad_norm": 2.6549898848315934e-06,
      "learning_rate": 0.2673659582702019,
      "loss": 0.0,
      "num_input_tokens_seen": 4864600,
      "step": 8560
    },
    {
      "epoch": 150.26548672566372,
      "grad_norm": 1.963453314601793e-06,
      "learning_rate": 0.2673292676128163,
      "loss": 0.0,
      "num_input_tokens_seen": 4867384,
      "step": 8565
    },
    {
      "epoch": 150.35398230088495,
      "grad_norm": 1.5582936612190679e-06,
      "learning_rate": 0.2672925588617831,
      "loss": 0.0,
      "num_input_tokens_seen": 4870200,
      "step": 8570
    },
    {
      "epoch": 150.44247787610618,
      "grad_norm": 7.766580552015512e-07,
      "learning_rate": 0.2672558320227634,
      "loss": 0.0,
      "num_input_tokens_seen": 4873048,
      "step": 8575
    },
    {
      "epoch": 150.53097345132744,
      "grad_norm": 1.7041674027495901e-06,
      "learning_rate": 0.2672190871014209,
      "loss": 0.0,
      "num_input_tokens_seen": 4875720,
      "step": 8580
    },
    {
      "epoch": 150.61946902654867,
      "grad_norm": 1.4939794255042216e-06,
      "learning_rate": 0.267182324103422,
      "loss": 0.0,
      "num_input_tokens_seen": 4878888,
      "step": 8585
    },
    {
      "epoch": 150.7079646017699,
      "grad_norm": 1.1828825563497958e-06,
      "learning_rate": 0.2671455430344362,
      "loss": 0.0,
      "num_input_tokens_seen": 4881928,
      "step": 8590
    },
    {
      "epoch": 150.79646017699116,
      "grad_norm": 1.229747113029589e-06,
      "learning_rate": 0.2671087439001355,
      "loss": 0.0,
      "num_input_tokens_seen": 4884856,
      "step": 8595
    },
    {
      "epoch": 150.8849557522124,
      "grad_norm": 9.956369240171625e-07,
      "learning_rate": 0.2670719267061948,
      "loss": 0.0,
      "num_input_tokens_seen": 4887640,
      "step": 8600
    },
    {
      "epoch": 150.8849557522124,
      "eval_loss": 0.4432040750980377,
      "eval_runtime": 0.9406,
      "eval_samples_per_second": 26.578,
      "eval_steps_per_second": 13.821,
      "num_input_tokens_seen": 4887640,
      "step": 8600
    },
    {
      "epoch": 150.97345132743362,
      "grad_norm": 1.0886783456953708e-06,
      "learning_rate": 0.2670350914582918,
      "loss": 0.0,
      "num_input_tokens_seen": 4890808,
      "step": 8605
    },
    {
      "epoch": 151.05309734513276,
      "grad_norm": 1.8027479882221087e-06,
      "learning_rate": 0.26699823816210694,
      "loss": 0.0,
      "num_input_tokens_seen": 4893112,
      "step": 8610
    },
    {
      "epoch": 151.141592920354,
      "grad_norm": 7.300220090655785e-07,
      "learning_rate": 0.26696136682332344,
      "loss": 0.0,
      "num_input_tokens_seen": 4895576,
      "step": 8615
    },
    {
      "epoch": 151.23008849557522,
      "grad_norm": 1.020078798319446e-06,
      "learning_rate": 0.2669244774476274,
      "loss": 0.0,
      "num_input_tokens_seen": 4898424,
      "step": 8620
    },
    {
      "epoch": 151.31858407079645,
      "grad_norm": 4.604262642260437e-07,
      "learning_rate": 0.2668875700407075,
      "loss": 0.0,
      "num_input_tokens_seen": 4901528,
      "step": 8625
    },
    {
      "epoch": 151.4070796460177,
      "grad_norm": 1.647059434617404e-06,
      "learning_rate": 0.26685064460825547,
      "loss": 0.0,
      "num_input_tokens_seen": 4904328,
      "step": 8630
    },
    {
      "epoch": 151.49557522123894,
      "grad_norm": 1.9853325738949934e-06,
      "learning_rate": 0.26681370115596553,
      "loss": 0.0,
      "num_input_tokens_seen": 4907400,
      "step": 8635
    },
    {
      "epoch": 151.58407079646017,
      "grad_norm": 1.21973152999999e-06,
      "learning_rate": 0.26677673968953497,
      "loss": 0.0,
      "num_input_tokens_seen": 4910600,
      "step": 8640
    },
    {
      "epoch": 151.67256637168143,
      "grad_norm": 2.035937768596341e-06,
      "learning_rate": 0.2667397602146636,
      "loss": 0.0,
      "num_input_tokens_seen": 4914008,
      "step": 8645
    },
    {
      "epoch": 151.76106194690266,
      "grad_norm": 9.016679314299836e-07,
      "learning_rate": 0.2667027627370542,
      "loss": 0.0,
      "num_input_tokens_seen": 4916760,
      "step": 8650
    },
    {
      "epoch": 151.8495575221239,
      "grad_norm": 9.068806434697763e-07,
      "learning_rate": 0.26666574726241216,
      "loss": 0.0,
      "num_input_tokens_seen": 4919976,
      "step": 8655
    },
    {
      "epoch": 151.93805309734512,
      "grad_norm": 1.8864385538108763e-06,
      "learning_rate": 0.2666287137964458,
      "loss": 0.0,
      "num_input_tokens_seen": 4922728,
      "step": 8660
    },
    {
      "epoch": 152.01769911504425,
      "grad_norm": 1.6715118817955954e-06,
      "learning_rate": 0.26659166234486614,
      "loss": 0.0,
      "num_input_tokens_seen": 4925128,
      "step": 8665
    },
    {
      "epoch": 152.10619469026548,
      "grad_norm": 2.8944025416421937e-06,
      "learning_rate": 0.2665545929133869,
      "loss": 0.0,
      "num_input_tokens_seen": 4927704,
      "step": 8670
    },
    {
      "epoch": 152.1946902654867,
      "grad_norm": 8.428706905760919e-07,
      "learning_rate": 0.2665175055077248,
      "loss": 0.0,
      "num_input_tokens_seen": 4930392,
      "step": 8675
    },
    {
      "epoch": 152.28318584070797,
      "grad_norm": 9.350391110274359e-07,
      "learning_rate": 0.2664804001335991,
      "loss": 0.0,
      "num_input_tokens_seen": 4933624,
      "step": 8680
    },
    {
      "epoch": 152.3716814159292,
      "grad_norm": 2.3547474370388954e-07,
      "learning_rate": 0.26644327679673185,
      "loss": 0.0,
      "num_input_tokens_seen": 4936360,
      "step": 8685
    },
    {
      "epoch": 152.46017699115043,
      "grad_norm": 7.122666261238919e-07,
      "learning_rate": 0.26640613550284803,
      "loss": 0.0,
      "num_input_tokens_seen": 4939368,
      "step": 8690
    },
    {
      "epoch": 152.5486725663717,
      "grad_norm": 3.163095811942185e-07,
      "learning_rate": 0.26636897625767525,
      "loss": 0.0,
      "num_input_tokens_seen": 4942776,
      "step": 8695
    },
    {
      "epoch": 152.63716814159292,
      "grad_norm": 7.810217539372388e-07,
      "learning_rate": 0.266331799066944,
      "loss": 0.0,
      "num_input_tokens_seen": 4945656,
      "step": 8700
    },
    {
      "epoch": 152.72566371681415,
      "grad_norm": 1.4472894918071688e-06,
      "learning_rate": 0.2662946039363874,
      "loss": 0.0,
      "num_input_tokens_seen": 4948312,
      "step": 8705
    },
    {
      "epoch": 152.81415929203538,
      "grad_norm": 1.1650843134702882e-06,
      "learning_rate": 0.2662573908717414,
      "loss": 0.0,
      "num_input_tokens_seen": 4951720,
      "step": 8710
    },
    {
      "epoch": 152.90265486725664,
      "grad_norm": 6.411382855731063e-07,
      "learning_rate": 0.2662201598787447,
      "loss": 0.0,
      "num_input_tokens_seen": 4954392,
      "step": 8715
    },
    {
      "epoch": 152.99115044247787,
      "grad_norm": 2.009155195992207e-06,
      "learning_rate": 0.2661829109631389,
      "loss": 0.0,
      "num_input_tokens_seen": 4957192,
      "step": 8720
    },
    {
      "epoch": 153.07079646017698,
      "grad_norm": 1.2237003375048516e-06,
      "learning_rate": 0.26614564413066816,
      "loss": 0.0,
      "num_input_tokens_seen": 4959344,
      "step": 8725
    },
    {
      "epoch": 153.15929203539824,
      "grad_norm": 1.6196739807128324e-06,
      "learning_rate": 0.2661083593870795,
      "loss": 0.0,
      "num_input_tokens_seen": 4963120,
      "step": 8730
    },
    {
      "epoch": 153.24778761061947,
      "grad_norm": 8.343562285517692e-07,
      "learning_rate": 0.26607105673812276,
      "loss": 0.0,
      "num_input_tokens_seen": 4965952,
      "step": 8735
    },
    {
      "epoch": 153.3362831858407,
      "grad_norm": 2.3750326363369823e-06,
      "learning_rate": 0.2660337361895504,
      "loss": 0.0,
      "num_input_tokens_seen": 4968464,
      "step": 8740
    },
    {
      "epoch": 153.42477876106196,
      "grad_norm": 1.583334324095631e-06,
      "learning_rate": 0.26599639774711775,
      "loss": 0.0,
      "num_input_tokens_seen": 4971920,
      "step": 8745
    },
    {
      "epoch": 153.5132743362832,
      "grad_norm": 6.503027520921023e-07,
      "learning_rate": 0.2659590414165829,
      "loss": 0.0,
      "num_input_tokens_seen": 4974480,
      "step": 8750
    },
    {
      "epoch": 153.60176991150442,
      "grad_norm": 1.3154509588275687e-06,
      "learning_rate": 0.2659216672037066,
      "loss": 0.0,
      "num_input_tokens_seen": 4977408,
      "step": 8755
    },
    {
      "epoch": 153.69026548672565,
      "grad_norm": 2.032919837802183e-06,
      "learning_rate": 0.26588427511425244,
      "loss": 0.0,
      "num_input_tokens_seen": 4980384,
      "step": 8760
    },
    {
      "epoch": 153.7787610619469,
      "grad_norm": 3.977224878326524e-07,
      "learning_rate": 0.26584686515398676,
      "loss": 0.0,
      "num_input_tokens_seen": 4983376,
      "step": 8765
    },
    {
      "epoch": 153.86725663716814,
      "grad_norm": 1.9783938114414923e-06,
      "learning_rate": 0.2658094373286787,
      "loss": 0.0,
      "num_input_tokens_seen": 4986032,
      "step": 8770
    },
    {
      "epoch": 153.95575221238937,
      "grad_norm": 1.1970090554314083e-06,
      "learning_rate": 0.2657719916441,
      "loss": 0.0,
      "num_input_tokens_seen": 4988576,
      "step": 8775
    },
    {
      "epoch": 154.0353982300885,
      "grad_norm": 8.171315357685671e-07,
      "learning_rate": 0.2657345281060253,
      "loss": 0.0,
      "num_input_tokens_seen": 4990864,
      "step": 8780
    },
    {
      "epoch": 154.12389380530973,
      "grad_norm": 1.4886775261402363e-06,
      "learning_rate": 0.26569704672023203,
      "loss": 0.0,
      "num_input_tokens_seen": 4993824,
      "step": 8785
    },
    {
      "epoch": 154.21238938053096,
      "grad_norm": 1.4394377103599254e-06,
      "learning_rate": 0.26565954749250015,
      "loss": 0.0,
      "num_input_tokens_seen": 4997072,
      "step": 8790
    },
    {
      "epoch": 154.30088495575222,
      "grad_norm": 2.2829183308203937e-06,
      "learning_rate": 0.2656220304286126,
      "loss": 0.0,
      "num_input_tokens_seen": 4999696,
      "step": 8795
    },
    {
      "epoch": 154.38938053097345,
      "grad_norm": 1.7466638837504433e-06,
      "learning_rate": 0.265584495534355,
      "loss": 0.0,
      "num_input_tokens_seen": 5002864,
      "step": 8800
    },
    {
      "epoch": 154.38938053097345,
      "eval_loss": 0.44304025173187256,
      "eval_runtime": 0.9395,
      "eval_samples_per_second": 26.61,
      "eval_steps_per_second": 13.837,
      "num_input_tokens_seen": 5002864,
      "step": 8800
    },
    {
      "epoch": 154.47787610619469,
      "grad_norm": 6.001200745231472e-07,
      "learning_rate": 0.2655469428155156,
      "loss": 0.0,
      "num_input_tokens_seen": 5005568,
      "step": 8805
    },
    {
      "epoch": 154.56637168141592,
      "grad_norm": 8.604246204413357e-07,
      "learning_rate": 0.2655093722778856,
      "loss": 0.0,
      "num_input_tokens_seen": 5008672,
      "step": 8810
    },
    {
      "epoch": 154.65486725663717,
      "grad_norm": 1.41474146175824e-07,
      "learning_rate": 0.2654717839272588,
      "loss": 0.0,
      "num_input_tokens_seen": 5011440,
      "step": 8815
    },
    {
      "epoch": 154.7433628318584,
      "grad_norm": 1.5609085721735028e-06,
      "learning_rate": 0.2654341777694318,
      "loss": 0.0,
      "num_input_tokens_seen": 5014576,
      "step": 8820
    },
    {
      "epoch": 154.83185840707964,
      "grad_norm": 1.1463945384093677e-06,
      "learning_rate": 0.265396553810204,
      "loss": 0.0,
      "num_input_tokens_seen": 5017440,
      "step": 8825
    },
    {
      "epoch": 154.9203539823009,
      "grad_norm": 1.7370635987390415e-06,
      "learning_rate": 0.26535891205537737,
      "loss": 0.0,
      "num_input_tokens_seen": 5019984,
      "step": 8830
    },
    {
      "epoch": 155.0,
      "grad_norm": 5.155172402737662e-06,
      "learning_rate": 0.26532125251075683,
      "loss": 0.0,
      "num_input_tokens_seen": 5022264,
      "step": 8835
    },
    {
      "epoch": 155.08849557522123,
      "grad_norm": 1.9065513470195583e-06,
      "learning_rate": 0.26528357518214996,
      "loss": 0.0,
      "num_input_tokens_seen": 5025208,
      "step": 8840
    },
    {
      "epoch": 155.1769911504425,
      "grad_norm": 3.7089532725076424e-06,
      "learning_rate": 0.26524588007536704,
      "loss": 0.0,
      "num_input_tokens_seen": 5027896,
      "step": 8845
    },
    {
      "epoch": 155.26548672566372,
      "grad_norm": 8.65292406615481e-07,
      "learning_rate": 0.26520816719622115,
      "loss": 0.0,
      "num_input_tokens_seen": 5031048,
      "step": 8850
    },
    {
      "epoch": 155.35398230088495,
      "grad_norm": 2.86147752603938e-07,
      "learning_rate": 0.2651704365505281,
      "loss": 0.0,
      "num_input_tokens_seen": 5033688,
      "step": 8855
    },
    {
      "epoch": 155.44247787610618,
      "grad_norm": 6.300498966993473e-07,
      "learning_rate": 0.26513268814410634,
      "loss": 0.0,
      "num_input_tokens_seen": 5036840,
      "step": 8860
    },
    {
      "epoch": 155.53097345132744,
      "grad_norm": 6.79839899930812e-07,
      "learning_rate": 0.2650949219827773,
      "loss": 0.0,
      "num_input_tokens_seen": 5039656,
      "step": 8865
    },
    {
      "epoch": 155.61946902654867,
      "grad_norm": 1.2234685300427373e-06,
      "learning_rate": 0.26505713807236486,
      "loss": 0.0,
      "num_input_tokens_seen": 5042568,
      "step": 8870
    },
    {
      "epoch": 155.7079646017699,
      "grad_norm": 1.1138057516291155e-06,
      "learning_rate": 0.26501933641869585,
      "loss": 0.0,
      "num_input_tokens_seen": 5045416,
      "step": 8875
    },
    {
      "epoch": 155.79646017699116,
      "grad_norm": 1.7852321434475016e-06,
      "learning_rate": 0.26498151702759976,
      "loss": 0.0,
      "num_input_tokens_seen": 5048088,
      "step": 8880
    },
    {
      "epoch": 155.8849557522124,
      "grad_norm": 4.615691295839497e-07,
      "learning_rate": 0.2649436799049088,
      "loss": 0.0,
      "num_input_tokens_seen": 5050776,
      "step": 8885
    },
    {
      "epoch": 155.97345132743362,
      "grad_norm": 1.99706028070068e-06,
      "learning_rate": 0.2649058250564579,
      "loss": 0.0,
      "num_input_tokens_seen": 5053928,
      "step": 8890
    },
    {
      "epoch": 156.05309734513276,
      "grad_norm": 9.188113381242147e-07,
      "learning_rate": 0.26486795248808476,
      "loss": 0.0,
      "num_input_tokens_seen": 5056392,
      "step": 8895
    },
    {
      "epoch": 156.141592920354,
      "grad_norm": 9.84950247584493e-07,
      "learning_rate": 0.2648300622056298,
      "loss": 0.0,
      "num_input_tokens_seen": 5059352,
      "step": 8900
    },
    {
      "epoch": 156.23008849557522,
      "grad_norm": 1.0143780855287332e-06,
      "learning_rate": 0.2647921542149363,
      "loss": 0.0,
      "num_input_tokens_seen": 5062344,
      "step": 8905
    },
    {
      "epoch": 156.31858407079645,
      "grad_norm": 1.947491227838327e-06,
      "learning_rate": 0.26475422852185,
      "loss": 0.0,
      "num_input_tokens_seen": 5065304,
      "step": 8910
    },
    {
      "epoch": 156.4070796460177,
      "grad_norm": 1.5824990668988903e-06,
      "learning_rate": 0.2647162851322196,
      "loss": 0.0,
      "num_input_tokens_seen": 5067720,
      "step": 8915
    },
    {
      "epoch": 156.49557522123894,
      "grad_norm": 1.1737329259631224e-06,
      "learning_rate": 0.2646783240518964,
      "loss": 0.0,
      "num_input_tokens_seen": 5071192,
      "step": 8920
    },
    {
      "epoch": 156.58407079646017,
      "grad_norm": 6.932904739187506e-07,
      "learning_rate": 0.26464034528673447,
      "loss": 0.0,
      "num_input_tokens_seen": 5074184,
      "step": 8925
    },
    {
      "epoch": 156.67256637168143,
      "grad_norm": 1.7572021988598863e-06,
      "learning_rate": 0.26460234884259065,
      "loss": 0.0,
      "num_input_tokens_seen": 5076616,
      "step": 8930
    },
    {
      "epoch": 156.76106194690266,
      "grad_norm": 1.3810505379296956e-06,
      "learning_rate": 0.2645643347253245,
      "loss": 0.0,
      "num_input_tokens_seen": 5078904,
      "step": 8935
    },
    {
      "epoch": 156.8495575221239,
      "grad_norm": 2.4241919618361862e-06,
      "learning_rate": 0.2645263029407982,
      "loss": 0.0,
      "num_input_tokens_seen": 5081912,
      "step": 8940
    },
    {
      "epoch": 156.93805309734512,
      "grad_norm": 1.4474469480774133e-06,
      "learning_rate": 0.2644882534948767,
      "loss": 0.0,
      "num_input_tokens_seen": 5085192,
      "step": 8945
    },
    {
      "epoch": 157.01769911504425,
      "grad_norm": 3.1579475034959614e-07,
      "learning_rate": 0.2644501863934278,
      "loss": 0.0,
      "num_input_tokens_seen": 5087352,
      "step": 8950
    },
    {
      "epoch": 157.10619469026548,
      "grad_norm": 1.1474867278593592e-06,
      "learning_rate": 0.26441210164232193,
      "loss": 0.0,
      "num_input_tokens_seen": 5090616,
      "step": 8955
    },
    {
      "epoch": 157.1946902654867,
      "grad_norm": 1.6119929568958469e-06,
      "learning_rate": 0.26437399924743216,
      "loss": 0.0,
      "num_input_tokens_seen": 5093528,
      "step": 8960
    },
    {
      "epoch": 157.28318584070797,
      "grad_norm": 6.755896606591705e-07,
      "learning_rate": 0.26433587921463436,
      "loss": 0.0,
      "num_input_tokens_seen": 5096264,
      "step": 8965
    },
    {
      "epoch": 157.3716814159292,
      "grad_norm": 1.8368114069744479e-06,
      "learning_rate": 0.2642977415498072,
      "loss": 0.0,
      "num_input_tokens_seen": 5099288,
      "step": 8970
    },
    {
      "epoch": 157.46017699115043,
      "grad_norm": 9.063800234798691e-07,
      "learning_rate": 0.26425958625883195,
      "loss": 0.0,
      "num_input_tokens_seen": 5102136,
      "step": 8975
    },
    {
      "epoch": 157.5486725663717,
      "grad_norm": 7.810195370439033e-07,
      "learning_rate": 0.2642214133475926,
      "loss": 0.0,
      "num_input_tokens_seen": 5104952,
      "step": 8980
    },
    {
      "epoch": 157.63716814159292,
      "grad_norm": 1.3804611853629467e-06,
      "learning_rate": 0.26418322282197587,
      "loss": 0.0,
      "num_input_tokens_seen": 5107672,
      "step": 8985
    },
    {
      "epoch": 157.72566371681415,
      "grad_norm": 2.571811364759924e-06,
      "learning_rate": 0.2641450146878714,
      "loss": 0.0,
      "num_input_tokens_seen": 5110504,
      "step": 8990
    },
    {
      "epoch": 157.81415929203538,
      "grad_norm": 1.11361441668123e-06,
      "learning_rate": 0.26410678895117107,
      "loss": 0.0,
      "num_input_tokens_seen": 5113192,
      "step": 8995
    },
    {
      "epoch": 157.90265486725664,
      "grad_norm": 6.888587904541055e-07,
      "learning_rate": 0.26406854561777,
      "loss": 0.0,
      "num_input_tokens_seen": 5116216,
      "step": 9000
    },
    {
      "epoch": 157.90265486725664,
      "eval_loss": 0.4386080205440521,
      "eval_runtime": 0.9434,
      "eval_samples_per_second": 26.501,
      "eval_steps_per_second": 13.78,
      "num_input_tokens_seen": 5116216,
      "step": 9000
    },
    {
      "epoch": 157.99115044247787,
      "grad_norm": 1.0334141506973538e-06,
      "learning_rate": 0.26403028469356576,
      "loss": 0.0,
      "num_input_tokens_seen": 5119064,
      "step": 9005
    },
    {
      "epoch": 158.07079646017698,
      "grad_norm": 1.0628543805069057e-06,
      "learning_rate": 0.2639920061844585,
      "loss": 0.0,
      "num_input_tokens_seen": 5121440,
      "step": 9010
    },
    {
      "epoch": 158.15929203539824,
      "grad_norm": 5.750115406044642e-07,
      "learning_rate": 0.2639537100963515,
      "loss": 0.0,
      "num_input_tokens_seen": 5124080,
      "step": 9015
    },
    {
      "epoch": 158.24778761061947,
      "grad_norm": 1.7414230342183146e-06,
      "learning_rate": 0.26391539643515033,
      "loss": 0.0,
      "num_input_tokens_seen": 5126608,
      "step": 9020
    },
    {
      "epoch": 158.3362831858407,
      "grad_norm": 1.7845078446043772e-06,
      "learning_rate": 0.26387706520676346,
      "loss": 0.0,
      "num_input_tokens_seen": 5129872,
      "step": 9025
    },
    {
      "epoch": 158.42477876106196,
      "grad_norm": 1.4294797665570513e-06,
      "learning_rate": 0.26383871641710205,
      "loss": 0.0,
      "num_input_tokens_seen": 5132784,
      "step": 9030
    },
    {
      "epoch": 158.5132743362832,
      "grad_norm": 7.858392336856923e-07,
      "learning_rate": 0.26380035007208,
      "loss": 0.0,
      "num_input_tokens_seen": 5135456,
      "step": 9035
    },
    {
      "epoch": 158.60176991150442,
      "grad_norm": 1.175102397610317e-06,
      "learning_rate": 0.26376196617761394,
      "loss": 0.0,
      "num_input_tokens_seen": 5138720,
      "step": 9040
    },
    {
      "epoch": 158.69026548672565,
      "grad_norm": 9.693820857137325e-07,
      "learning_rate": 0.263723564739623,
      "loss": 0.0,
      "num_input_tokens_seen": 5141424,
      "step": 9045
    },
    {
      "epoch": 158.7787610619469,
      "grad_norm": 1.7227825992449652e-06,
      "learning_rate": 0.2636851457640293,
      "loss": 0.0,
      "num_input_tokens_seen": 5144544,
      "step": 9050
    },
    {
      "epoch": 158.86725663716814,
      "grad_norm": 9.72829866441316e-07,
      "learning_rate": 0.26364670925675737,
      "loss": 0.0,
      "num_input_tokens_seen": 5147264,
      "step": 9055
    },
    {
      "epoch": 158.95575221238937,
      "grad_norm": 1.9631072518677684e-06,
      "learning_rate": 0.2636082552237347,
      "loss": 0.0,
      "num_input_tokens_seen": 5150592,
      "step": 9060
    },
    {
      "epoch": 159.0353982300885,
      "grad_norm": 3.9169088950075093e-07,
      "learning_rate": 0.26356978367089146,
      "loss": 0.0,
      "num_input_tokens_seen": 5153320,
      "step": 9065
    },
    {
      "epoch": 159.12389380530973,
      "grad_norm": 1.3532903722079936e-06,
      "learning_rate": 0.26353129460416036,
      "loss": 0.0,
      "num_input_tokens_seen": 5156296,
      "step": 9070
    },
    {
      "epoch": 159.21238938053096,
      "grad_norm": 8.96110918802151e-07,
      "learning_rate": 0.2634927880294769,
      "loss": 0.0,
      "num_input_tokens_seen": 5159224,
      "step": 9075
    },
    {
      "epoch": 159.30088495575222,
      "grad_norm": 5.45693922049395e-07,
      "learning_rate": 0.26345426395277927,
      "loss": 0.0,
      "num_input_tokens_seen": 5162072,
      "step": 9080
    },
    {
      "epoch": 159.38938053097345,
      "grad_norm": 2.0904196844639955e-06,
      "learning_rate": 0.2634157223800084,
      "loss": 0.0,
      "num_input_tokens_seen": 5164808,
      "step": 9085
    },
    {
      "epoch": 159.47787610619469,
      "grad_norm": 7.680549174438056e-07,
      "learning_rate": 0.26337716331710787,
      "loss": 0.0,
      "num_input_tokens_seen": 5167720,
      "step": 9090
    },
    {
      "epoch": 159.56637168141592,
      "grad_norm": 7.592185511384741e-07,
      "learning_rate": 0.2633385867700239,
      "loss": 0.0,
      "num_input_tokens_seen": 5170504,
      "step": 9095
    },
    {
      "epoch": 159.65486725663717,
      "grad_norm": 2.313399363629287e-06,
      "learning_rate": 0.2632999927447056,
      "loss": 0.0,
      "num_input_tokens_seen": 5173800,
      "step": 9100
    },
    {
      "epoch": 159.7433628318584,
      "grad_norm": 1.8627988538355567e-06,
      "learning_rate": 0.2632613812471046,
      "loss": 0.0,
      "num_input_tokens_seen": 5176872,
      "step": 9105
    },
    {
      "epoch": 159.83185840707964,
      "grad_norm": 3.197208684468933e-07,
      "learning_rate": 0.2632227522831753,
      "loss": 0.0,
      "num_input_tokens_seen": 5179384,
      "step": 9110
    },
    {
      "epoch": 159.9203539823009,
      "grad_norm": 1.413410927852965e-06,
      "learning_rate": 0.26318410585887475,
      "loss": 0.0,
      "num_input_tokens_seen": 5181848,
      "step": 9115
    },
    {
      "epoch": 160.0,
      "grad_norm": 1.0365952221036423e-05,
      "learning_rate": 0.2631454419801627,
      "loss": 0.0,
      "num_input_tokens_seen": 5184128,
      "step": 9120
    },
    {
      "epoch": 160.08849557522123,
      "grad_norm": 1.0116607427335111e-06,
      "learning_rate": 0.2631067606530016,
      "loss": 0.0,
      "num_input_tokens_seen": 5187152,
      "step": 9125
    },
    {
      "epoch": 160.1769911504425,
      "grad_norm": 4.734545768769749e-07,
      "learning_rate": 0.2630680618833567,
      "loss": 0.0,
      "num_input_tokens_seen": 5189888,
      "step": 9130
    },
    {
      "epoch": 160.26548672566372,
      "grad_norm": 5.837300136590784e-07,
      "learning_rate": 0.26302934567719566,
      "loss": 0.0,
      "num_input_tokens_seen": 5192880,
      "step": 9135
    },
    {
      "epoch": 160.35398230088495,
      "grad_norm": 2.1910798864155367e-07,
      "learning_rate": 0.2629906120404892,
      "loss": 0.0,
      "num_input_tokens_seen": 5195648,
      "step": 9140
    },
    {
      "epoch": 160.44247787610618,
      "grad_norm": 1.3909684639656916e-06,
      "learning_rate": 0.26295186097921036,
      "loss": 0.0,
      "num_input_tokens_seen": 5198352,
      "step": 9145
    },
    {
      "epoch": 160.53097345132744,
      "grad_norm": 6.163019179439289e-07,
      "learning_rate": 0.2629130924993351,
      "loss": 0.0,
      "num_input_tokens_seen": 5200992,
      "step": 9150
    },
    {
      "epoch": 160.61946902654867,
      "grad_norm": 8.044802939366491e-07,
      "learning_rate": 0.2628743066068421,
      "loss": 0.0,
      "num_input_tokens_seen": 5204016,
      "step": 9155
    },
    {
      "epoch": 160.7079646017699,
      "grad_norm": 7.983333034644602e-07,
      "learning_rate": 0.26283550330771244,
      "loss": 0.0,
      "num_input_tokens_seen": 5207072,
      "step": 9160
    },
    {
      "epoch": 160.79646017699116,
      "grad_norm": 2.2133244215183367e-07,
      "learning_rate": 0.2627966826079303,
      "loss": 0.0,
      "num_input_tokens_seen": 5210256,
      "step": 9165
    },
    {
      "epoch": 160.8849557522124,
      "grad_norm": 7.287785024345794e-07,
      "learning_rate": 0.26275784451348216,
      "loss": 0.0,
      "num_input_tokens_seen": 5213088,
      "step": 9170
    },
    {
      "epoch": 160.97345132743362,
      "grad_norm": 3.061009863358777e-07,
      "learning_rate": 0.2627189890303574,
      "loss": 0.0,
      "num_input_tokens_seen": 5215904,
      "step": 9175
    },
    {
      "epoch": 161.05309734513276,
      "grad_norm": 7.594845783387427e-07,
      "learning_rate": 0.262680116164548,
      "loss": 0.0,
      "num_input_tokens_seen": 5218216,
      "step": 9180
    },
    {
      "epoch": 161.141592920354,
      "grad_norm": 1.5278326372936135e-06,
      "learning_rate": 0.2626412259220487,
      "loss": 0.0,
      "num_input_tokens_seen": 5220856,
      "step": 9185
    },
    {
      "epoch": 161.23008849557522,
      "grad_norm": 2.279214641021099e-06,
      "learning_rate": 0.2626023183088568,
      "loss": 0.0,
      "num_input_tokens_seen": 5223368,
      "step": 9190
    },
    {
      "epoch": 161.31858407079645,
      "grad_norm": 2.4150747321982635e-06,
      "learning_rate": 0.26256339333097234,
      "loss": 0.0,
      "num_input_tokens_seen": 5226040,
      "step": 9195
    },
    {
      "epoch": 161.4070796460177,
      "grad_norm": 5.783414280813304e-07,
      "learning_rate": 0.2625244509943981,
      "loss": 0.0,
      "num_input_tokens_seen": 5229496,
      "step": 9200
    },
    {
      "epoch": 161.4070796460177,
      "eval_loss": 0.44973084330558777,
      "eval_runtime": 0.9304,
      "eval_samples_per_second": 26.87,
      "eval_steps_per_second": 13.972,
      "num_input_tokens_seen": 5229496,
      "step": 9200
    },
    {
      "epoch": 161.49557522123894,
      "grad_norm": 1.740652692205913e-06,
      "learning_rate": 0.2624854913051395,
      "loss": 0.0,
      "num_input_tokens_seen": 5232456,
      "step": 9205
    },
    {
      "epoch": 161.58407079646017,
      "grad_norm": 1.299213977290492e-06,
      "learning_rate": 0.26244651426920446,
      "loss": 0.0,
      "num_input_tokens_seen": 5235352,
      "step": 9210
    },
    {
      "epoch": 161.67256637168143,
      "grad_norm": 1.174952899418713e-06,
      "learning_rate": 0.26240751989260386,
      "loss": 0.0,
      "num_input_tokens_seen": 5238584,
      "step": 9215
    },
    {
      "epoch": 161.76106194690266,
      "grad_norm": 1.5093143019839772e-06,
      "learning_rate": 0.2623685081813511,
      "loss": 0.0,
      "num_input_tokens_seen": 5241432,
      "step": 9220
    },
    {
      "epoch": 161.8495575221239,
      "grad_norm": 1.150908246927429e-06,
      "learning_rate": 0.2623294791414623,
      "loss": 0.0,
      "num_input_tokens_seen": 5243896,
      "step": 9225
    },
    {
      "epoch": 161.93805309734512,
      "grad_norm": 1.1311979051242815e-06,
      "learning_rate": 0.26229043277895614,
      "loss": 0.0,
      "num_input_tokens_seen": 5246424,
      "step": 9230
    },
    {
      "epoch": 162.01769911504425,
      "grad_norm": 1.8442494820192223e-06,
      "learning_rate": 0.2622513690998542,
      "loss": 0.0,
      "num_input_tokens_seen": 5248776,
      "step": 9235
    },
    {
      "epoch": 162.10619469026548,
      "grad_norm": 1.0949501074719592e-06,
      "learning_rate": 0.26221228811018044,
      "loss": 0.0,
      "num_input_tokens_seen": 5251608,
      "step": 9240
    },
    {
      "epoch": 162.1946902654867,
      "grad_norm": 3.0772244485888223e-07,
      "learning_rate": 0.2621731898159617,
      "loss": 0.0,
      "num_input_tokens_seen": 5254232,
      "step": 9245
    },
    {
      "epoch": 162.28318584070797,
      "grad_norm": 1.7691534139885334e-06,
      "learning_rate": 0.26213407422322743,
      "loss": 0.0,
      "num_input_tokens_seen": 5257112,
      "step": 9250
    },
    {
      "epoch": 162.3716814159292,
      "grad_norm": 8.651633152112481e-07,
      "learning_rate": 0.2620949413380098,
      "loss": 0.0,
      "num_input_tokens_seen": 5259864,
      "step": 9255
    },
    {
      "epoch": 162.46017699115043,
      "grad_norm": 5.826173605782969e-07,
      "learning_rate": 0.26205579116634353,
      "loss": 0.0,
      "num_input_tokens_seen": 5263208,
      "step": 9260
    },
    {
      "epoch": 162.5486725663717,
      "grad_norm": 1.442405960005999e-06,
      "learning_rate": 0.26201662371426604,
      "loss": 0.0,
      "num_input_tokens_seen": 5266040,
      "step": 9265
    },
    {
      "epoch": 162.63716814159292,
      "grad_norm": 1.3707209518543095e-06,
      "learning_rate": 0.2619774389878175,
      "loss": 0.0,
      "num_input_tokens_seen": 5268712,
      "step": 9270
    },
    {
      "epoch": 162.72566371681415,
      "grad_norm": 2.5841563910944387e-07,
      "learning_rate": 0.2619382369930407,
      "loss": 0.0,
      "num_input_tokens_seen": 5271544,
      "step": 9275
    },
    {
      "epoch": 162.81415929203538,
      "grad_norm": 9.527062161396316e-07,
      "learning_rate": 0.261899017735981,
      "loss": 0.0,
      "num_input_tokens_seen": 5274616,
      "step": 9280
    },
    {
      "epoch": 162.90265486725664,
      "grad_norm": 7.39605695798673e-07,
      "learning_rate": 0.2618597812226866,
      "loss": 0.0,
      "num_input_tokens_seen": 5277656,
      "step": 9285
    },
    {
      "epoch": 162.99115044247787,
      "grad_norm": 3.2651999504196283e-07,
      "learning_rate": 0.2618205274592082,
      "loss": 0.0,
      "num_input_tokens_seen": 5280744,
      "step": 9290
    },
    {
      "epoch": 163.07079646017698,
      "grad_norm": 1.5065702427818906e-06,
      "learning_rate": 0.2617812564515992,
      "loss": 0.0,
      "num_input_tokens_seen": 5283384,
      "step": 9295
    },
    {
      "epoch": 163.15929203539824,
      "grad_norm": 2.5946926598408027e-06,
      "learning_rate": 0.2617419682059158,
      "loss": 0.0,
      "num_input_tokens_seen": 5286088,
      "step": 9300
    },
    {
      "epoch": 163.24778761061947,
      "grad_norm": 1.169985807791818e-06,
      "learning_rate": 0.26170266272821663,
      "loss": 0.0,
      "num_input_tokens_seen": 5289016,
      "step": 9305
    },
    {
      "epoch": 163.3362831858407,
      "grad_norm": 5.555161806114484e-07,
      "learning_rate": 0.26166334002456315,
      "loss": 0.0,
      "num_input_tokens_seen": 5291928,
      "step": 9310
    },
    {
      "epoch": 163.42477876106196,
      "grad_norm": 1.4536717571900226e-06,
      "learning_rate": 0.2616240001010194,
      "loss": 0.0,
      "num_input_tokens_seen": 5294776,
      "step": 9315
    },
    {
      "epoch": 163.5132743362832,
      "grad_norm": 1.8220508763988619e-06,
      "learning_rate": 0.26158464296365197,
      "loss": 0.0,
      "num_input_tokens_seen": 5297224,
      "step": 9320
    },
    {
      "epoch": 163.60176991150442,
      "grad_norm": 1.6166519571925164e-06,
      "learning_rate": 0.2615452686185304,
      "loss": 0.0,
      "num_input_tokens_seen": 5300056,
      "step": 9325
    },
    {
      "epoch": 163.69026548672565,
      "grad_norm": 4.776167088493821e-07,
      "learning_rate": 0.26150587707172673,
      "loss": 0.0,
      "num_input_tokens_seen": 5303032,
      "step": 9330
    },
    {
      "epoch": 163.7787610619469,
      "grad_norm": 2.1290040876920102e-06,
      "learning_rate": 0.2614664683293154,
      "loss": 0.0,
      "num_input_tokens_seen": 5306136,
      "step": 9335
    },
    {
      "epoch": 163.86725663716814,
      "grad_norm": 1.2265254554222338e-06,
      "learning_rate": 0.26142704239737397,
      "loss": 0.0,
      "num_input_tokens_seen": 5309416,
      "step": 9340
    },
    {
      "epoch": 163.95575221238937,
      "grad_norm": 1.3173472552807652e-06,
      "learning_rate": 0.26138759928198235,
      "loss": 0.0,
      "num_input_tokens_seen": 5312040,
      "step": 9345
    },
    {
      "epoch": 164.0353982300885,
      "grad_norm": 1.1018003078788752e-06,
      "learning_rate": 0.26134813898922304,
      "loss": 0.0,
      "num_input_tokens_seen": 5314136,
      "step": 9350
    },
    {
      "epoch": 164.12389380530973,
      "grad_norm": 3.201470519798022e-07,
      "learning_rate": 0.26130866152518145,
      "loss": 0.0,
      "num_input_tokens_seen": 5317080,
      "step": 9355
    },
    {
      "epoch": 164.21238938053096,
      "grad_norm": 1.6069276398411603e-06,
      "learning_rate": 0.2612691668959455,
      "loss": 0.0,
      "num_input_tokens_seen": 5320120,
      "step": 9360
    },
    {
      "epoch": 164.30088495575222,
      "grad_norm": 1.5328344034060137e-06,
      "learning_rate": 0.2612296551076057,
      "loss": 0.0,
      "num_input_tokens_seen": 5323048,
      "step": 9365
    },
    {
      "epoch": 164.38938053097345,
      "grad_norm": 8.522584380443732e-07,
      "learning_rate": 0.26119012616625525,
      "loss": 0.0,
      "num_input_tokens_seen": 5325864,
      "step": 9370
    },
    {
      "epoch": 164.47787610619469,
      "grad_norm": 5.240049176791217e-07,
      "learning_rate": 0.26115058007799,
      "loss": 0.0,
      "num_input_tokens_seen": 5328824,
      "step": 9375
    },
    {
      "epoch": 164.56637168141592,
      "grad_norm": 1.8962005015055183e-06,
      "learning_rate": 0.26111101684890864,
      "loss": 0.0,
      "num_input_tokens_seen": 5331480,
      "step": 9380
    },
    {
      "epoch": 164.65486725663717,
      "grad_norm": 2.594476427475456e-06,
      "learning_rate": 0.26107143648511205,
      "loss": 0.0,
      "num_input_tokens_seen": 5334504,
      "step": 9385
    },
    {
      "epoch": 164.7433628318584,
      "grad_norm": 7.52058042507997e-07,
      "learning_rate": 0.2610318389927042,
      "loss": 0.0,
      "num_input_tokens_seen": 5337752,
      "step": 9390
    },
    {
      "epoch": 164.83185840707964,
      "grad_norm": 9.514160410617478e-07,
      "learning_rate": 0.26099222437779146,
      "loss": 0.0,
      "num_input_tokens_seen": 5340600,
      "step": 9395
    },
    {
      "epoch": 164.9203539823009,
      "grad_norm": 1.0113332109540352e-06,
      "learning_rate": 0.26095259264648285,
      "loss": 0.0,
      "num_input_tokens_seen": 5343528,
      "step": 9400
    },
    {
      "epoch": 164.9203539823009,
      "eval_loss": 0.4416623115539551,
      "eval_runtime": 0.9501,
      "eval_samples_per_second": 26.314,
      "eval_steps_per_second": 13.683,
      "num_input_tokens_seen": 5343528,
      "step": 9400
    },
    {
      "epoch": 165.0,
      "grad_norm": 1.1688463246173342e-06,
      "learning_rate": 0.2609129438048902,
      "loss": 0.0,
      "num_input_tokens_seen": 5345792,
      "step": 9405
    },
    {
      "epoch": 165.08849557522123,
      "grad_norm": 2.050467173830839e-07,
      "learning_rate": 0.2608732778591278,
      "loss": 0.0,
      "num_input_tokens_seen": 5349024,
      "step": 9410
    },
    {
      "epoch": 165.1769911504425,
      "grad_norm": 6.003189128023223e-07,
      "learning_rate": 0.2608335948153126,
      "loss": 0.0,
      "num_input_tokens_seen": 5351648,
      "step": 9415
    },
    {
      "epoch": 165.26548672566372,
      "grad_norm": 6.018974545440869e-07,
      "learning_rate": 0.26079389467956426,
      "loss": 0.0,
      "num_input_tokens_seen": 5354752,
      "step": 9420
    },
    {
      "epoch": 165.35398230088495,
      "grad_norm": 2.4366556772292824e-06,
      "learning_rate": 0.26075417745800505,
      "loss": 0.0,
      "num_input_tokens_seen": 5357888,
      "step": 9425
    },
    {
      "epoch": 165.44247787610618,
      "grad_norm": 1.507936531197629e-06,
      "learning_rate": 0.26071444315675985,
      "loss": 0.0,
      "num_input_tokens_seen": 5360512,
      "step": 9430
    },
    {
      "epoch": 165.53097345132744,
      "grad_norm": 7.135993200790836e-07,
      "learning_rate": 0.2606746917819562,
      "loss": 0.0,
      "num_input_tokens_seen": 5363216,
      "step": 9435
    },
    {
      "epoch": 165.61946902654867,
      "grad_norm": 1.9940539459639695e-06,
      "learning_rate": 0.2606349233397242,
      "loss": 0.0,
      "num_input_tokens_seen": 5366000,
      "step": 9440
    },
    {
      "epoch": 165.7079646017699,
      "grad_norm": 5.69863175314822e-07,
      "learning_rate": 0.26059513783619676,
      "loss": 0.0,
      "num_input_tokens_seen": 5368528,
      "step": 9445
    },
    {
      "epoch": 165.79646017699116,
      "grad_norm": 1.7007643009492313e-06,
      "learning_rate": 0.26055533527750924,
      "loss": 0.0,
      "num_input_tokens_seen": 5371136,
      "step": 9450
    },
    {
      "epoch": 165.8849557522124,
      "grad_norm": 1.6074332052085083e-06,
      "learning_rate": 0.26051551566979964,
      "loss": 0.0,
      "num_input_tokens_seen": 5374288,
      "step": 9455
    },
    {
      "epoch": 165.97345132743362,
      "grad_norm": 5.490290959642152e-07,
      "learning_rate": 0.26047567901920876,
      "loss": 0.0,
      "num_input_tokens_seen": 5377472,
      "step": 9460
    },
    {
      "epoch": 166.05309734513276,
      "grad_norm": 5.187025635677855e-07,
      "learning_rate": 0.2604358253318798,
      "loss": 0.0,
      "num_input_tokens_seen": 5379560,
      "step": 9465
    },
    {
      "epoch": 166.141592920354,
      "grad_norm": 6.283880225055327e-07,
      "learning_rate": 0.26039595461395876,
      "loss": 0.0,
      "num_input_tokens_seen": 5382488,
      "step": 9470
    },
    {
      "epoch": 166.23008849557522,
      "grad_norm": 7.323864110730938e-07,
      "learning_rate": 0.26035606687159424,
      "loss": 0.0,
      "num_input_tokens_seen": 5386232,
      "step": 9475
    },
    {
      "epoch": 166.31858407079645,
      "grad_norm": 1.0003367378885741e-06,
      "learning_rate": 0.26031616211093733,
      "loss": 0.0,
      "num_input_tokens_seen": 5389112,
      "step": 9480
    },
    {
      "epoch": 166.4070796460177,
      "grad_norm": 1.2375026017252821e-06,
      "learning_rate": 0.26027624033814195,
      "loss": 0.0,
      "num_input_tokens_seen": 5391752,
      "step": 9485
    },
    {
      "epoch": 166.49557522123894,
      "grad_norm": 1.7135698726633564e-06,
      "learning_rate": 0.2602363015593645,
      "loss": 0.0,
      "num_input_tokens_seen": 5394056,
      "step": 9490
    },
    {
      "epoch": 166.58407079646017,
      "grad_norm": 3.3928154152818024e-07,
      "learning_rate": 0.26019634578076395,
      "loss": 0.0,
      "num_input_tokens_seen": 5396792,
      "step": 9495
    },
    {
      "epoch": 166.67256637168143,
      "grad_norm": 1.1863111240018043e-06,
      "learning_rate": 0.26015637300850214,
      "loss": 0.0,
      "num_input_tokens_seen": 5399960,
      "step": 9500
    },
    {
      "epoch": 166.76106194690266,
      "grad_norm": 4.0022237612902245e-07,
      "learning_rate": 0.26011638324874325,
      "loss": 0.0,
      "num_input_tokens_seen": 5403048,
      "step": 9505
    },
    {
      "epoch": 166.8495575221239,
      "grad_norm": 1.009422476272448e-06,
      "learning_rate": 0.2600763765076543,
      "loss": 0.0,
      "num_input_tokens_seen": 5406024,
      "step": 9510
    },
    {
      "epoch": 166.93805309734512,
      "grad_norm": 1.1389446399334702e-06,
      "learning_rate": 0.2600363527914048,
      "loss": 0.0,
      "num_input_tokens_seen": 5408744,
      "step": 9515
    },
    {
      "epoch": 167.01769911504425,
      "grad_norm": 1.4645391956946696e-06,
      "learning_rate": 0.25999631210616686,
      "loss": 0.0,
      "num_input_tokens_seen": 5410896,
      "step": 9520
    },
    {
      "epoch": 167.10619469026548,
      "grad_norm": 1.3896102473154315e-06,
      "learning_rate": 0.25995625445811527,
      "loss": 0.0,
      "num_input_tokens_seen": 5413488,
      "step": 9525
    },
    {
      "epoch": 167.1946902654867,
      "grad_norm": 1.2741423915940686e-06,
      "learning_rate": 0.2599161798534275,
      "loss": 0.0,
      "num_input_tokens_seen": 5416224,
      "step": 9530
    },
    {
      "epoch": 167.28318584070797,
      "grad_norm": 1.3902574664825806e-06,
      "learning_rate": 0.25987608829828346,
      "loss": 0.0,
      "num_input_tokens_seen": 5418960,
      "step": 9535
    },
    {
      "epoch": 167.3716814159292,
      "grad_norm": 8.274520268969354e-07,
      "learning_rate": 0.25983597979886586,
      "loss": 0.0,
      "num_input_tokens_seen": 5421488,
      "step": 9540
    },
    {
      "epoch": 167.46017699115043,
      "grad_norm": 1.2534086408777512e-06,
      "learning_rate": 0.2597958543613599,
      "loss": 0.0,
      "num_input_tokens_seen": 5424528,
      "step": 9545
    },
    {
      "epoch": 167.5486725663717,
      "grad_norm": 7.321469297494332e-07,
      "learning_rate": 0.25975571199195335,
      "loss": 0.0,
      "num_input_tokens_seen": 5427264,
      "step": 9550
    },
    {
      "epoch": 167.63716814159292,
      "grad_norm": 5.270725296213641e-07,
      "learning_rate": 0.25971555269683677,
      "loss": 0.0,
      "num_input_tokens_seen": 5430304,
      "step": 9555
    },
    {
      "epoch": 167.72566371681415,
      "grad_norm": 7.7315627322605e-07,
      "learning_rate": 0.25967537648220324,
      "loss": 0.0,
      "num_input_tokens_seen": 5433264,
      "step": 9560
    },
    {
      "epoch": 167.81415929203538,
      "grad_norm": 7.044670269351627e-07,
      "learning_rate": 0.2596351833542483,
      "loss": 0.0,
      "num_input_tokens_seen": 5436464,
      "step": 9565
    },
    {
      "epoch": 167.90265486725664,
      "grad_norm": 5.958944484518724e-07,
      "learning_rate": 0.25959497331917036,
      "loss": 0.0,
      "num_input_tokens_seen": 5439456,
      "step": 9570
    },
    {
      "epoch": 167.99115044247787,
      "grad_norm": 9.419439379598771e-07,
      "learning_rate": 0.2595547463831703,
      "loss": 0.0,
      "num_input_tokens_seen": 5442544,
      "step": 9575
    },
    {
      "epoch": 168.07079646017698,
      "grad_norm": 6.483644483523676e-07,
      "learning_rate": 0.25951450255245156,
      "loss": 0.0,
      "num_input_tokens_seen": 5444544,
      "step": 9580
    },
    {
      "epoch": 168.15929203539824,
      "grad_norm": 5.369107611841173e-07,
      "learning_rate": 0.2594742418332203,
      "loss": 0.0,
      "num_input_tokens_seen": 5447296,
      "step": 9585
    },
    {
      "epoch": 168.24778761061947,
      "grad_norm": 6.7873276066166e-07,
      "learning_rate": 0.2594339642316852,
      "loss": 0.0,
      "num_input_tokens_seen": 5449888,
      "step": 9590
    },
    {
      "epoch": 168.3362831858407,
      "grad_norm": 1.827297296586039e-06,
      "learning_rate": 0.2593936697540576,
      "loss": 0.0,
      "num_input_tokens_seen": 5452416,
      "step": 9595
    },
    {
      "epoch": 168.42477876106196,
      "grad_norm": 1.3565155541073182e-06,
      "learning_rate": 0.2593533584065514,
      "loss": 0.0,
      "num_input_tokens_seen": 5455520,
      "step": 9600
    },
    {
      "epoch": 168.42477876106196,
      "eval_loss": 0.44280382990837097,
      "eval_runtime": 0.9379,
      "eval_samples_per_second": 26.655,
      "eval_steps_per_second": 13.861,
      "num_input_tokens_seen": 5455520,
      "step": 9600
    },
    {
      "epoch": 168.5132743362832,
      "grad_norm": 8.965440656538703e-07,
      "learning_rate": 0.2593130301953831,
      "loss": 0.0,
      "num_input_tokens_seen": 5457984,
      "step": 9605
    },
    {
      "epoch": 168.60176991150442,
      "grad_norm": 1.221045749844052e-06,
      "learning_rate": 0.2592726851267718,
      "loss": 0.0,
      "num_input_tokens_seen": 5460992,
      "step": 9610
    },
    {
      "epoch": 168.69026548672565,
      "grad_norm": 3.6759220733983966e-07,
      "learning_rate": 0.2592323232069393,
      "loss": 0.0,
      "num_input_tokens_seen": 5464464,
      "step": 9615
    },
    {
      "epoch": 168.7787610619469,
      "grad_norm": 1.3608339486381738e-06,
      "learning_rate": 0.25919194444210986,
      "loss": 0.0,
      "num_input_tokens_seen": 5467280,
      "step": 9620
    },
    {
      "epoch": 168.86725663716814,
      "grad_norm": 5.421613877842901e-07,
      "learning_rate": 0.2591515488385103,
      "loss": 0.0,
      "num_input_tokens_seen": 5470064,
      "step": 9625
    },
    {
      "epoch": 168.95575221238937,
      "grad_norm": 9.259190392185701e-07,
      "learning_rate": 0.2591111364023704,
      "loss": 0.0,
      "num_input_tokens_seen": 5473632,
      "step": 9630
    },
    {
      "epoch": 169.0353982300885,
      "grad_norm": 5.97123460011062e-07,
      "learning_rate": 0.259070707139922,
      "loss": 0.0,
      "num_input_tokens_seen": 5476040,
      "step": 9635
    },
    {
      "epoch": 169.12389380530973,
      "grad_norm": 2.086968464709571e-07,
      "learning_rate": 0.25903026105739985,
      "loss": 0.0,
      "num_input_tokens_seen": 5479496,
      "step": 9640
    },
    {
      "epoch": 169.21238938053096,
      "grad_norm": 4.2667556954256725e-07,
      "learning_rate": 0.2589897981610413,
      "loss": 0.0,
      "num_input_tokens_seen": 5482344,
      "step": 9645
    },
    {
      "epoch": 169.30088495575222,
      "grad_norm": 6.6142064270025e-07,
      "learning_rate": 0.2589493184570863,
      "loss": 0.0,
      "num_input_tokens_seen": 5485160,
      "step": 9650
    },
    {
      "epoch": 169.38938053097345,
      "grad_norm": 8.537323878954339e-07,
      "learning_rate": 0.25890882195177717,
      "loss": 0.0,
      "num_input_tokens_seen": 5488504,
      "step": 9655
    },
    {
      "epoch": 169.47787610619469,
      "grad_norm": 4.222642076001648e-07,
      "learning_rate": 0.25886830865135907,
      "loss": 0.0,
      "num_input_tokens_seen": 5491368,
      "step": 9660
    },
    {
      "epoch": 169.56637168141592,
      "grad_norm": 4.42291366198333e-07,
      "learning_rate": 0.25882777856207967,
      "loss": 0.0,
      "num_input_tokens_seen": 5494088,
      "step": 9665
    },
    {
      "epoch": 169.65486725663717,
      "grad_norm": 6.075916303416307e-07,
      "learning_rate": 0.2587872316901892,
      "loss": 0.0,
      "num_input_tokens_seen": 5496600,
      "step": 9670
    },
    {
      "epoch": 169.7433628318584,
      "grad_norm": 8.180441000149585e-07,
      "learning_rate": 0.25874666804194046,
      "loss": 0.0,
      "num_input_tokens_seen": 5499688,
      "step": 9675
    },
    {
      "epoch": 169.83185840707964,
      "grad_norm": 1.3343995988179813e-06,
      "learning_rate": 0.258706087623589,
      "loss": 0.0,
      "num_input_tokens_seen": 5502408,
      "step": 9680
    },
    {
      "epoch": 169.9203539823009,
      "grad_norm": 2.598568471512408e-07,
      "learning_rate": 0.25866549044139264,
      "loss": 0.0,
      "num_input_tokens_seen": 5504648,
      "step": 9685
    },
    {
      "epoch": 170.0,
      "grad_norm": 1.9071860890562675e-07,
      "learning_rate": 0.25862487650161214,
      "loss": 0.0,
      "num_input_tokens_seen": 5507520,
      "step": 9690
    },
    {
      "epoch": 170.08849557522123,
      "grad_norm": 5.159791385267454e-07,
      "learning_rate": 0.2585842458105106,
      "loss": 0.0,
      "num_input_tokens_seen": 5510368,
      "step": 9695
    },
    {
      "epoch": 170.1769911504425,
      "grad_norm": 1.3275857781991363e-06,
      "learning_rate": 0.2585435983743538,
      "loss": 0.0,
      "num_input_tokens_seen": 5513312,
      "step": 9700
    },
    {
      "epoch": 170.26548672566372,
      "grad_norm": 8.732565106583934e-07,
      "learning_rate": 0.2585029341994101,
      "loss": 0.0,
      "num_input_tokens_seen": 5516144,
      "step": 9705
    },
    {
      "epoch": 170.35398230088495,
      "grad_norm": 5.096934501125361e-07,
      "learning_rate": 0.2584622532919504,
      "loss": 0.0,
      "num_input_tokens_seen": 5519584,
      "step": 9710
    },
    {
      "epoch": 170.44247787610618,
      "grad_norm": 3.5953905808128184e-07,
      "learning_rate": 0.2584215556582482,
      "loss": 0.0,
      "num_input_tokens_seen": 5522768,
      "step": 9715
    },
    {
      "epoch": 170.53097345132744,
      "grad_norm": 1.2492635050875833e-06,
      "learning_rate": 0.25838084130457967,
      "loss": 0.0,
      "num_input_tokens_seen": 5525888,
      "step": 9720
    },
    {
      "epoch": 170.61946902654867,
      "grad_norm": 9.357937074128131e-07,
      "learning_rate": 0.2583401102372234,
      "loss": 0.0,
      "num_input_tokens_seen": 5528976,
      "step": 9725
    },
    {
      "epoch": 170.7079646017699,
      "grad_norm": 6.92861135576095e-07,
      "learning_rate": 0.2582993624624606,
      "loss": 0.0,
      "num_input_tokens_seen": 5531600,
      "step": 9730
    },
    {
      "epoch": 170.79646017699116,
      "grad_norm": 1.4500195675282157e-06,
      "learning_rate": 0.25825859798657513,
      "loss": 0.0,
      "num_input_tokens_seen": 5534096,
      "step": 9735
    },
    {
      "epoch": 170.8849557522124,
      "grad_norm": 8.545809464521881e-07,
      "learning_rate": 0.25821781681585343,
      "loss": 0.0,
      "num_input_tokens_seen": 5537040,
      "step": 9740
    },
    {
      "epoch": 170.97345132743362,
      "grad_norm": 1.7535298866278026e-06,
      "learning_rate": 0.2581770189565844,
      "loss": 0.0,
      "num_input_tokens_seen": 5539408,
      "step": 9745
    },
    {
      "epoch": 171.05309734513276,
      "grad_norm": 1.0923268973783706e-06,
      "learning_rate": 0.25813620441505963,
      "loss": 0.0,
      "num_input_tokens_seen": 5541880,
      "step": 9750
    },
    {
      "epoch": 171.141592920354,
      "grad_norm": 5.707390755560482e-07,
      "learning_rate": 0.2580953731975732,
      "loss": 0.0,
      "num_input_tokens_seen": 5544680,
      "step": 9755
    },
    {
      "epoch": 171.23008849557522,
      "grad_norm": 4.5402796899907116e-07,
      "learning_rate": 0.2580545253104218,
      "loss": 0.0,
      "num_input_tokens_seen": 5547368,
      "step": 9760
    },
    {
      "epoch": 171.31858407079645,
      "grad_norm": 1.0678242006179062e-06,
      "learning_rate": 0.2580136607599047,
      "loss": 0.0,
      "num_input_tokens_seen": 5550840,
      "step": 9765
    },
    {
      "epoch": 171.4070796460177,
      "grad_norm": 6.227437552297488e-07,
      "learning_rate": 0.2579727795523238,
      "loss": 0.0,
      "num_input_tokens_seen": 5553928,
      "step": 9770
    },
    {
      "epoch": 171.49557522123894,
      "grad_norm": 5.642093015012506e-07,
      "learning_rate": 0.25793188169398334,
      "loss": 0.0,
      "num_input_tokens_seen": 5557528,
      "step": 9775
    },
    {
      "epoch": 171.58407079646017,
      "grad_norm": 8.428304454355384e-07,
      "learning_rate": 0.25789096719119037,
      "loss": 0.0,
      "num_input_tokens_seen": 5560232,
      "step": 9780
    },
    {
      "epoch": 171.67256637168143,
      "grad_norm": 1.4495470850306447e-06,
      "learning_rate": 0.2578500360502544,
      "loss": 0.0,
      "num_input_tokens_seen": 5563320,
      "step": 9785
    },
    {
      "epoch": 171.76106194690266,
      "grad_norm": 3.746903018964076e-07,
      "learning_rate": 0.2578090882774876,
      "loss": 0.0,
      "num_input_tokens_seen": 5566072,
      "step": 9790
    },
    {
      "epoch": 171.8495575221239,
      "grad_norm": 2.463800115037884e-07,
      "learning_rate": 0.25776812387920456,
      "loss": 0.0,
      "num_input_tokens_seen": 5568776,
      "step": 9795
    },
    {
      "epoch": 171.93805309734512,
      "grad_norm": 5.784648351436772e-07,
      "learning_rate": 0.2577271428617225,
      "loss": 0.0,
      "num_input_tokens_seen": 5571144,
      "step": 9800
    },
    {
      "epoch": 171.93805309734512,
      "eval_loss": 0.46453505754470825,
      "eval_runtime": 0.9478,
      "eval_samples_per_second": 26.378,
      "eval_steps_per_second": 13.716,
      "num_input_tokens_seen": 5571144,
      "step": 9800
    },
    {
      "epoch": 172.01769911504425,
      "grad_norm": 3.9363231962852296e-07,
      "learning_rate": 0.25768614523136124,
      "loss": 0.0,
      "num_input_tokens_seen": 5573536,
      "step": 9805
    },
    {
      "epoch": 172.10619469026548,
      "grad_norm": 4.1034496689462685e-07,
      "learning_rate": 0.25764513099444314,
      "loss": 0.0,
      "num_input_tokens_seen": 5577200,
      "step": 9810
    },
    {
      "epoch": 172.1946902654867,
      "grad_norm": 4.3027793594774266e-07,
      "learning_rate": 0.25760410015729307,
      "loss": 0.0,
      "num_input_tokens_seen": 5579760,
      "step": 9815
    },
    {
      "epoch": 172.28318584070797,
      "grad_norm": 1.1344837957949494e-06,
      "learning_rate": 0.2575630527262385,
      "loss": 0.0,
      "num_input_tokens_seen": 5582272,
      "step": 9820
    },
    {
      "epoch": 172.3716814159292,
      "grad_norm": 1.294854087063868e-06,
      "learning_rate": 0.25752198870760945,
      "loss": 0.0,
      "num_input_tokens_seen": 5584960,
      "step": 9825
    },
    {
      "epoch": 172.46017699115043,
      "grad_norm": 6.653154400737549e-07,
      "learning_rate": 0.2574809081077386,
      "loss": 0.0,
      "num_input_tokens_seen": 5587952,
      "step": 9830
    },
    {
      "epoch": 172.5486725663717,
      "grad_norm": 3.500682623780449e-07,
      "learning_rate": 0.257439810932961,
      "loss": 0.0,
      "num_input_tokens_seen": 5590816,
      "step": 9835
    },
    {
      "epoch": 172.63716814159292,
      "grad_norm": 4.631496040019556e-07,
      "learning_rate": 0.2573986971896144,
      "loss": 0.0,
      "num_input_tokens_seen": 5593600,
      "step": 9840
    },
    {
      "epoch": 172.72566371681415,
      "grad_norm": 7.678991096327081e-07,
      "learning_rate": 0.257357566884039,
      "loss": 0.0,
      "num_input_tokens_seen": 5596416,
      "step": 9845
    },
    {
      "epoch": 172.81415929203538,
      "grad_norm": 5.57580676741054e-07,
      "learning_rate": 0.25731642002257765,
      "loss": 0.0,
      "num_input_tokens_seen": 5599328,
      "step": 9850
    },
    {
      "epoch": 172.90265486725664,
      "grad_norm": 5.268072413855407e-07,
      "learning_rate": 0.25727525661157574,
      "loss": 0.0,
      "num_input_tokens_seen": 5602256,
      "step": 9855
    },
    {
      "epoch": 172.99115044247787,
      "grad_norm": 9.959489943867084e-07,
      "learning_rate": 0.2572340766573811,
      "loss": 0.0,
      "num_input_tokens_seen": 5605344,
      "step": 9860
    },
    {
      "epoch": 173.07079646017698,
      "grad_norm": 5.619958756142296e-07,
      "learning_rate": 0.25719288016634434,
      "loss": 0.0,
      "num_input_tokens_seen": 5608000,
      "step": 9865
    },
    {
      "epoch": 173.15929203539824,
      "grad_norm": 6.375897214638826e-07,
      "learning_rate": 0.25715166714481835,
      "loss": 0.0,
      "num_input_tokens_seen": 5610912,
      "step": 9870
    },
    {
      "epoch": 173.24778761061947,
      "grad_norm": 8.576348022870661e-07,
      "learning_rate": 0.2571104375991587,
      "loss": 0.0,
      "num_input_tokens_seen": 5613472,
      "step": 9875
    },
    {
      "epoch": 173.3362831858407,
      "grad_norm": 3.8613978858847986e-07,
      "learning_rate": 0.2570691915357236,
      "loss": 0.0,
      "num_input_tokens_seen": 5616848,
      "step": 9880
    },
    {
      "epoch": 173.42477876106196,
      "grad_norm": 1.133868977376551e-06,
      "learning_rate": 0.2570279289608736,
      "loss": 0.0,
      "num_input_tokens_seen": 5619792,
      "step": 9885
    },
    {
      "epoch": 173.5132743362832,
      "grad_norm": 1.475868316447304e-06,
      "learning_rate": 0.256986649880972,
      "loss": 0.0,
      "num_input_tokens_seen": 5622464,
      "step": 9890
    },
    {
      "epoch": 173.60176991150442,
      "grad_norm": 6.175059752422385e-07,
      "learning_rate": 0.25694535430238447,
      "loss": 0.0,
      "num_input_tokens_seen": 5625072,
      "step": 9895
    },
    {
      "epoch": 173.69026548672565,
      "grad_norm": 6.138965318314149e-07,
      "learning_rate": 0.25690404223147933,
      "loss": 0.0,
      "num_input_tokens_seen": 5627920,
      "step": 9900
    },
    {
      "epoch": 173.7787610619469,
      "grad_norm": 6.891769999128883e-07,
      "learning_rate": 0.2568627136746275,
      "loss": 0.0,
      "num_input_tokens_seen": 5631152,
      "step": 9905
    },
    {
      "epoch": 173.86725663716814,
      "grad_norm": 7.149875500545022e-07,
      "learning_rate": 0.25682136863820226,
      "loss": 0.0,
      "num_input_tokens_seen": 5633968,
      "step": 9910
    },
    {
      "epoch": 173.95575221238937,
      "grad_norm": 7.502736707465374e-07,
      "learning_rate": 0.25678000712857957,
      "loss": 0.0,
      "num_input_tokens_seen": 5636880,
      "step": 9915
    },
    {
      "epoch": 174.0353982300885,
      "grad_norm": 6.72746693908266e-07,
      "learning_rate": 0.2567386291521379,
      "loss": 0.0,
      "num_input_tokens_seen": 5638976,
      "step": 9920
    },
    {
      "epoch": 174.12389380530973,
      "grad_norm": 3.2097449320644955e-07,
      "learning_rate": 0.2566972347152583,
      "loss": 0.0,
      "num_input_tokens_seen": 5641904,
      "step": 9925
    },
    {
      "epoch": 174.21238938053096,
      "grad_norm": 4.426210864494351e-07,
      "learning_rate": 0.2566558238243242,
      "loss": 0.0,
      "num_input_tokens_seen": 5644704,
      "step": 9930
    },
    {
      "epoch": 174.30088495575222,
      "grad_norm": 1.5719176360562415e-07,
      "learning_rate": 0.25661439648572176,
      "loss": 0.0,
      "num_input_tokens_seen": 5647760,
      "step": 9935
    },
    {
      "epoch": 174.38938053097345,
      "grad_norm": 6.890215900057228e-07,
      "learning_rate": 0.25657295270583963,
      "loss": 0.0,
      "num_input_tokens_seen": 5650896,
      "step": 9940
    },
    {
      "epoch": 174.47787610619469,
      "grad_norm": 8.40659254208731e-07,
      "learning_rate": 0.25653149249106894,
      "loss": 0.0,
      "num_input_tokens_seen": 5653824,
      "step": 9945
    },
    {
      "epoch": 174.56637168141592,
      "grad_norm": 1.111713004320336e-06,
      "learning_rate": 0.25649001584780323,
      "loss": 0.0,
      "num_input_tokens_seen": 5656496,
      "step": 9950
    },
    {
      "epoch": 174.65486725663717,
      "grad_norm": 1.3682382586921449e-06,
      "learning_rate": 0.2564485227824389,
      "loss": 0.0,
      "num_input_tokens_seen": 5659728,
      "step": 9955
    },
    {
      "epoch": 174.7433628318584,
      "grad_norm": 6.153516665108327e-07,
      "learning_rate": 0.25640701330137466,
      "loss": 0.0,
      "num_input_tokens_seen": 5662688,
      "step": 9960
    },
    {
      "epoch": 174.83185840707964,
      "grad_norm": 4.057859825934429e-07,
      "learning_rate": 0.2563654874110117,
      "loss": 0.0,
      "num_input_tokens_seen": 5665344,
      "step": 9965
    },
    {
      "epoch": 174.9203539823009,
      "grad_norm": 6.175017688292428e-07,
      "learning_rate": 0.256323945117754,
      "loss": 0.0,
      "num_input_tokens_seen": 5668368,
      "step": 9970
    },
    {
      "epoch": 175.0,
      "grad_norm": 1.9634464933915297e-06,
      "learning_rate": 0.2562823864280078,
      "loss": 0.0,
      "num_input_tokens_seen": 5670672,
      "step": 9975
    },
    {
      "epoch": 175.08849557522123,
      "grad_norm": 2.096967932629923e-07,
      "learning_rate": 0.25624081134818194,
      "loss": 0.0,
      "num_input_tokens_seen": 5673520,
      "step": 9980
    },
    {
      "epoch": 175.1769911504425,
      "grad_norm": 8.006753660083632e-07,
      "learning_rate": 0.2561992198846879,
      "loss": 0.0,
      "num_input_tokens_seen": 5676432,
      "step": 9985
    },
    {
      "epoch": 175.26548672566372,
      "grad_norm": 5.378751097850909e-07,
      "learning_rate": 0.25615761204393955,
      "loss": 0.0,
      "num_input_tokens_seen": 5679040,
      "step": 9990
    },
    {
      "epoch": 175.35398230088495,
      "grad_norm": 1.0169601409870666e-06,
      "learning_rate": 0.2561159878323534,
      "loss": 0.0,
      "num_input_tokens_seen": 5682224,
      "step": 9995
    },
    {
      "epoch": 175.44247787610618,
      "grad_norm": 6.552631930389907e-07,
      "learning_rate": 0.2560743472563483,
      "loss": 0.0,
      "num_input_tokens_seen": 5684752,
      "step": 10000
    },
    {
      "epoch": 175.44247787610618,
      "eval_loss": 0.46158474683761597,
      "eval_runtime": 0.9334,
      "eval_samples_per_second": 26.783,
      "eval_steps_per_second": 13.927,
      "num_input_tokens_seen": 5684752,
      "step": 10000
    },
    {
      "epoch": 175.53097345132744,
      "grad_norm": 1.7069710338546429e-06,
      "learning_rate": 0.25603269032234593,
      "loss": 0.0,
      "num_input_tokens_seen": 5687408,
      "step": 10005
    },
    {
      "epoch": 175.61946902654867,
      "grad_norm": 6.926238143023511e-07,
      "learning_rate": 0.2559910170367702,
      "loss": 0.0,
      "num_input_tokens_seen": 5690576,
      "step": 10010
    },
    {
      "epoch": 175.7079646017699,
      "grad_norm": 1.1514555353642209e-07,
      "learning_rate": 0.2559493274060477,
      "loss": 0.0,
      "num_input_tokens_seen": 5693824,
      "step": 10015
    },
    {
      "epoch": 175.79646017699116,
      "grad_norm": 4.4904780338583805e-07,
      "learning_rate": 0.2559076214366074,
      "loss": 0.0,
      "num_input_tokens_seen": 5696208,
      "step": 10020
    },
    {
      "epoch": 175.8849557522124,
      "grad_norm": 1.7226157922323182e-07,
      "learning_rate": 0.25586589913488106,
      "loss": 0.0,
      "num_input_tokens_seen": 5699024,
      "step": 10025
    },
    {
      "epoch": 175.97345132743362,
      "grad_norm": 9.8963880645897e-07,
      "learning_rate": 0.2558241605073026,
      "loss": 0.0,
      "num_input_tokens_seen": 5702400,
      "step": 10030
    },
    {
      "epoch": 176.05309734513276,
      "grad_norm": 3.4393423220535624e-07,
      "learning_rate": 0.25578240556030873,
      "loss": 0.0,
      "num_input_tokens_seen": 5705040,
      "step": 10035
    },
    {
      "epoch": 176.141592920354,
      "grad_norm": 7.927820320219325e-07,
      "learning_rate": 0.2557406343003386,
      "loss": 0.0,
      "num_input_tokens_seen": 5707824,
      "step": 10040
    },
    {
      "epoch": 176.23008849557522,
      "grad_norm": 7.080615205268259e-07,
      "learning_rate": 0.25569884673383375,
      "loss": 0.0,
      "num_input_tokens_seen": 5710928,
      "step": 10045
    },
    {
      "epoch": 176.31858407079645,
      "grad_norm": 8.161716209542647e-07,
      "learning_rate": 0.25565704286723856,
      "loss": 0.0,
      "num_input_tokens_seen": 5713984,
      "step": 10050
    },
    {
      "epoch": 176.4070796460177,
      "grad_norm": 8.949215271059074e-07,
      "learning_rate": 0.25561522270699955,
      "loss": 0.0,
      "num_input_tokens_seen": 5716688,
      "step": 10055
    },
    {
      "epoch": 176.49557522123894,
      "grad_norm": 1.1462577731435886e-06,
      "learning_rate": 0.25557338625956594,
      "loss": 0.0,
      "num_input_tokens_seen": 5719504,
      "step": 10060
    },
    {
      "epoch": 176.58407079646017,
      "grad_norm": 7.322515784835559e-07,
      "learning_rate": 0.25553153353138947,
      "loss": 0.0,
      "num_input_tokens_seen": 5723040,
      "step": 10065
    },
    {
      "epoch": 176.67256637168143,
      "grad_norm": 1.1235404144827044e-06,
      "learning_rate": 0.2554896645289243,
      "loss": 0.0,
      "num_input_tokens_seen": 5726336,
      "step": 10070
    },
    {
      "epoch": 176.76106194690266,
      "grad_norm": 5.732352974519017e-07,
      "learning_rate": 0.2554477792586272,
      "loss": 0.0,
      "num_input_tokens_seen": 5728752,
      "step": 10075
    },
    {
      "epoch": 176.8495575221239,
      "grad_norm": 1.2697099691649782e-06,
      "learning_rate": 0.25540587772695744,
      "loss": 0.0,
      "num_input_tokens_seen": 5731760,
      "step": 10080
    },
    {
      "epoch": 176.93805309734512,
      "grad_norm": 1.6324030411851709e-06,
      "learning_rate": 0.2553639599403767,
      "loss": 0.0,
      "num_input_tokens_seen": 5734512,
      "step": 10085
    },
    {
      "epoch": 177.01769911504425,
      "grad_norm": 4.800507440450019e-07,
      "learning_rate": 0.2553220259053493,
      "loss": 0.0,
      "num_input_tokens_seen": 5736544,
      "step": 10090
    },
    {
      "epoch": 177.10619469026548,
      "grad_norm": 1.3244165302239708e-06,
      "learning_rate": 0.2552800756283419,
      "loss": 0.0,
      "num_input_tokens_seen": 5739264,
      "step": 10095
    },
    {
      "epoch": 177.1946902654867,
      "grad_norm": 1.3003855201532133e-06,
      "learning_rate": 0.25523810911582373,
      "loss": 0.0,
      "num_input_tokens_seen": 5741616,
      "step": 10100
    },
    {
      "epoch": 177.28318584070797,
      "grad_norm": 1.2432545304363884e-07,
      "learning_rate": 0.25519612637426675,
      "loss": 0.0,
      "num_input_tokens_seen": 5744560,
      "step": 10105
    },
    {
      "epoch": 177.3716814159292,
      "grad_norm": 2.2067681015869312e-07,
      "learning_rate": 0.25515412741014504,
      "loss": 0.0,
      "num_input_tokens_seen": 5747488,
      "step": 10110
    },
    {
      "epoch": 177.46017699115043,
      "grad_norm": 5.107368110657262e-07,
      "learning_rate": 0.2551121122299355,
      "loss": 0.0,
      "num_input_tokens_seen": 5750416,
      "step": 10115
    },
    {
      "epoch": 177.5486725663717,
      "grad_norm": 9.266151437259396e-07,
      "learning_rate": 0.2550700808401173,
      "loss": 0.0,
      "num_input_tokens_seen": 5753440,
      "step": 10120
    },
    {
      "epoch": 177.63716814159292,
      "grad_norm": 6.896650575072272e-07,
      "learning_rate": 0.2550280332471722,
      "loss": 0.0,
      "num_input_tokens_seen": 5756464,
      "step": 10125
    },
    {
      "epoch": 177.72566371681415,
      "grad_norm": 7.001908102211019e-07,
      "learning_rate": 0.2549859694575845,
      "loss": 0.0,
      "num_input_tokens_seen": 5759648,
      "step": 10130
    },
    {
      "epoch": 177.81415929203538,
      "grad_norm": 7.504602308472386e-07,
      "learning_rate": 0.254943889477841,
      "loss": 0.0,
      "num_input_tokens_seen": 5762496,
      "step": 10135
    },
    {
      "epoch": 177.90265486725664,
      "grad_norm": 1.500200141890673e-07,
      "learning_rate": 0.25490179331443097,
      "loss": 0.0,
      "num_input_tokens_seen": 5765104,
      "step": 10140
    },
    {
      "epoch": 177.99115044247787,
      "grad_norm": 1.0296648724761326e-06,
      "learning_rate": 0.25485968097384615,
      "loss": 0.0,
      "num_input_tokens_seen": 5768032,
      "step": 10145
    },
    {
      "epoch": 178.07079646017698,
      "grad_norm": 5.137089260642824e-07,
      "learning_rate": 0.25481755246258075,
      "loss": 0.0,
      "num_input_tokens_seen": 5770576,
      "step": 10150
    },
    {
      "epoch": 178.15929203539824,
      "grad_norm": 1.202865405502962e-06,
      "learning_rate": 0.2547754077871315,
      "loss": 0.0,
      "num_input_tokens_seen": 5773232,
      "step": 10155
    },
    {
      "epoch": 178.24778761061947,
      "grad_norm": 1.043835737846166e-07,
      "learning_rate": 0.25473324695399774,
      "loss": 0.0,
      "num_input_tokens_seen": 5776272,
      "step": 10160
    },
    {
      "epoch": 178.3362831858407,
      "grad_norm": 4.298837836813618e-07,
      "learning_rate": 0.25469106996968105,
      "loss": 0.0,
      "num_input_tokens_seen": 5779056,
      "step": 10165
    },
    {
      "epoch": 178.42477876106196,
      "grad_norm": 1.0468470463820267e-06,
      "learning_rate": 0.2546488768406858,
      "loss": 0.0,
      "num_input_tokens_seen": 5782176,
      "step": 10170
    },
    {
      "epoch": 178.5132743362832,
      "grad_norm": 4.6743105031055165e-07,
      "learning_rate": 0.25460666757351863,
      "loss": 0.0,
      "num_input_tokens_seen": 5784656,
      "step": 10175
    },
    {
      "epoch": 178.60176991150442,
      "grad_norm": 6.569048878191097e-07,
      "learning_rate": 0.25456444217468877,
      "loss": 0.0,
      "num_input_tokens_seen": 5787488,
      "step": 10180
    },
    {
      "epoch": 178.69026548672565,
      "grad_norm": 5.913069571761298e-07,
      "learning_rate": 0.25452220065070785,
      "loss": 0.0,
      "num_input_tokens_seen": 5790464,
      "step": 10185
    },
    {
      "epoch": 178.7787610619469,
      "grad_norm": 3.1503839181823423e-07,
      "learning_rate": 0.2544799430080901,
      "loss": 0.0,
      "num_input_tokens_seen": 5793568,
      "step": 10190
    },
    {
      "epoch": 178.86725663716814,
      "grad_norm": 7.537661304013454e-07,
      "learning_rate": 0.2544376692533522,
      "loss": 0.0,
      "num_input_tokens_seen": 5796240,
      "step": 10195
    },
    {
      "epoch": 178.95575221238937,
      "grad_norm": 1.484410461216612e-07,
      "learning_rate": 0.2543953793930132,
      "loss": 0.0,
      "num_input_tokens_seen": 5799088,
      "step": 10200
    },
    {
      "epoch": 178.95575221238937,
      "eval_loss": 0.4657347798347473,
      "eval_runtime": 0.943,
      "eval_samples_per_second": 26.511,
      "eval_steps_per_second": 13.786,
      "num_input_tokens_seen": 5799088,
      "step": 10200
    },
    {
      "epoch": 179.0353982300885,
      "grad_norm": 6.438323225665954e-07,
      "learning_rate": 0.2543530734335948,
      "loss": 0.0,
      "num_input_tokens_seen": 5801504,
      "step": 10205
    },
    {
      "epoch": 179.12389380530973,
      "grad_norm": 3.868573799081787e-07,
      "learning_rate": 0.2543107513816211,
      "loss": 0.0,
      "num_input_tokens_seen": 5804720,
      "step": 10210
    },
    {
      "epoch": 179.21238938053096,
      "grad_norm": 1.1935974271182204e-06,
      "learning_rate": 0.25426841324361865,
      "loss": 0.0,
      "num_input_tokens_seen": 5807760,
      "step": 10215
    },
    {
      "epoch": 179.30088495575222,
      "grad_norm": 2.2406334210245404e-07,
      "learning_rate": 0.2542260590261166,
      "loss": 0.0,
      "num_input_tokens_seen": 5810400,
      "step": 10220
    },
    {
      "epoch": 179.38938053097345,
      "grad_norm": 4.5803599846294674e-07,
      "learning_rate": 0.2541836887356465,
      "loss": 0.0,
      "num_input_tokens_seen": 5813040,
      "step": 10225
    },
    {
      "epoch": 179.47787610619469,
      "grad_norm": 8.532252877557767e-07,
      "learning_rate": 0.2541413023787423,
      "loss": 0.0,
      "num_input_tokens_seen": 5815760,
      "step": 10230
    },
    {
      "epoch": 179.56637168141592,
      "grad_norm": 3.699542787671817e-07,
      "learning_rate": 0.2540988999619405,
      "loss": 0.0,
      "num_input_tokens_seen": 5818896,
      "step": 10235
    },
    {
      "epoch": 179.65486725663717,
      "grad_norm": 5.080584060124238e-07,
      "learning_rate": 0.25405648149178023,
      "loss": 0.0,
      "num_input_tokens_seen": 5821712,
      "step": 10240
    },
    {
      "epoch": 179.7433628318584,
      "grad_norm": 1.140921654041449e-06,
      "learning_rate": 0.2540140469748028,
      "loss": 0.0,
      "num_input_tokens_seen": 5824880,
      "step": 10245
    },
    {
      "epoch": 179.83185840707964,
      "grad_norm": 9.539871825836599e-07,
      "learning_rate": 0.25397159641755224,
      "loss": 0.0,
      "num_input_tokens_seen": 5827120,
      "step": 10250
    },
    {
      "epoch": 179.9203539823009,
      "grad_norm": 5.078371145827987e-07,
      "learning_rate": 0.2539291298265749,
      "loss": 0.0,
      "num_input_tokens_seen": 5829744,
      "step": 10255
    },
    {
      "epoch": 180.0,
      "grad_norm": 3.6598241877072724e-06,
      "learning_rate": 0.2538866472084197,
      "loss": 0.0,
      "num_input_tokens_seen": 5832608,
      "step": 10260
    },
    {
      "epoch": 180.08849557522123,
      "grad_norm": 3.426382875204581e-07,
      "learning_rate": 0.25384414856963794,
      "loss": 0.0,
      "num_input_tokens_seen": 5835344,
      "step": 10265
    },
    {
      "epoch": 180.1769911504425,
      "grad_norm": 2.6591936830300256e-07,
      "learning_rate": 0.25380163391678356,
      "loss": 0.0,
      "num_input_tokens_seen": 5838000,
      "step": 10270
    },
    {
      "epoch": 180.26548672566372,
      "grad_norm": 8.167453984242456e-07,
      "learning_rate": 0.2537591032564127,
      "loss": 0.0,
      "num_input_tokens_seen": 5841456,
      "step": 10275
    },
    {
      "epoch": 180.35398230088495,
      "grad_norm": 7.224388127724524e-07,
      "learning_rate": 0.25371655659508424,
      "loss": 0.0,
      "num_input_tokens_seen": 5844736,
      "step": 10280
    },
    {
      "epoch": 180.44247787610618,
      "grad_norm": 3.1182091220216535e-07,
      "learning_rate": 0.25367399393935935,
      "loss": 0.0,
      "num_input_tokens_seen": 5847664,
      "step": 10285
    },
    {
      "epoch": 180.53097345132744,
      "grad_norm": 5.951781076873885e-07,
      "learning_rate": 0.25363141529580174,
      "loss": 0.0,
      "num_input_tokens_seen": 5850640,
      "step": 10290
    },
    {
      "epoch": 180.61946902654867,
      "grad_norm": 6.274538577599742e-07,
      "learning_rate": 0.2535888206709776,
      "loss": 0.0,
      "num_input_tokens_seen": 5853648,
      "step": 10295
    },
    {
      "epoch": 180.7079646017699,
      "grad_norm": 3.1341474482360354e-07,
      "learning_rate": 0.2535462100714555,
      "loss": 0.0,
      "num_input_tokens_seen": 5856320,
      "step": 10300
    },
    {
      "epoch": 180.79646017699116,
      "grad_norm": 1.0557417908785283e-06,
      "learning_rate": 0.2535035835038066,
      "loss": 0.0,
      "num_input_tokens_seen": 5859488,
      "step": 10305
    },
    {
      "epoch": 180.8849557522124,
      "grad_norm": 5.690814077752293e-07,
      "learning_rate": 0.2534609409746044,
      "loss": 0.0,
      "num_input_tokens_seen": 5862000,
      "step": 10310
    },
    {
      "epoch": 180.97345132743362,
      "grad_norm": 1.494498746978934e-07,
      "learning_rate": 0.253418282490425,
      "loss": 0.0,
      "num_input_tokens_seen": 5864784,
      "step": 10315
    },
    {
      "epoch": 181.05309734513276,
      "grad_norm": 8.032748723962868e-07,
      "learning_rate": 0.2533756080578467,
      "loss": 0.0,
      "num_input_tokens_seen": 5867224,
      "step": 10320
    },
    {
      "epoch": 181.141592920354,
      "grad_norm": 3.888093544901494e-07,
      "learning_rate": 0.25333291768345056,
      "loss": 0.0,
      "num_input_tokens_seen": 5870120,
      "step": 10325
    },
    {
      "epoch": 181.23008849557522,
      "grad_norm": 3.0545811569027137e-07,
      "learning_rate": 0.25329021137381996,
      "loss": 0.0,
      "num_input_tokens_seen": 5873176,
      "step": 10330
    },
    {
      "epoch": 181.31858407079645,
      "grad_norm": 5.146350190443627e-07,
      "learning_rate": 0.25324748913554074,
      "loss": 0.0,
      "num_input_tokens_seen": 5875640,
      "step": 10335
    },
    {
      "epoch": 181.4070796460177,
      "grad_norm": 3.896085729593324e-07,
      "learning_rate": 0.2532047509752013,
      "loss": 0.0,
      "num_input_tokens_seen": 5878280,
      "step": 10340
    },
    {
      "epoch": 181.49557522123894,
      "grad_norm": 5.762699402112048e-07,
      "learning_rate": 0.25316199689939217,
      "loss": 0.0,
      "num_input_tokens_seen": 5881016,
      "step": 10345
    },
    {
      "epoch": 181.58407079646017,
      "grad_norm": 3.710241571752704e-07,
      "learning_rate": 0.2531192269147068,
      "loss": 0.0,
      "num_input_tokens_seen": 5884168,
      "step": 10350
    },
    {
      "epoch": 181.67256637168143,
      "grad_norm": 2.6285010790161323e-07,
      "learning_rate": 0.2530764410277407,
      "loss": 0.0,
      "num_input_tokens_seen": 5886712,
      "step": 10355
    },
    {
      "epoch": 181.76106194690266,
      "grad_norm": 3.969768727074552e-07,
      "learning_rate": 0.25303363924509203,
      "loss": 0.0,
      "num_input_tokens_seen": 5889544,
      "step": 10360
    },
    {
      "epoch": 181.8495575221239,
      "grad_norm": 7.360694098679232e-07,
      "learning_rate": 0.25299082157336145,
      "loss": 0.0,
      "num_input_tokens_seen": 5892392,
      "step": 10365
    },
    {
      "epoch": 181.93805309734512,
      "grad_norm": 1.5729705182820908e-07,
      "learning_rate": 0.2529479880191519,
      "loss": 0.0,
      "num_input_tokens_seen": 5895608,
      "step": 10370
    },
    {
      "epoch": 182.01769911504425,
      "grad_norm": 3.913989132797724e-07,
      "learning_rate": 0.2529051385890689,
      "loss": 0.0,
      "num_input_tokens_seen": 5898128,
      "step": 10375
    },
    {
      "epoch": 182.10619469026548,
      "grad_norm": 3.455609487446054e-07,
      "learning_rate": 0.2528622732897203,
      "loss": 0.0,
      "num_input_tokens_seen": 5900848,
      "step": 10380
    },
    {
      "epoch": 182.1946902654867,
      "grad_norm": 4.55203974070173e-07,
      "learning_rate": 0.25281939212771654,
      "loss": 0.0,
      "num_input_tokens_seen": 5903488,
      "step": 10385
    },
    {
      "epoch": 182.28318584070797,
      "grad_norm": 2.2679968481043034e-07,
      "learning_rate": 0.2527764951096704,
      "loss": 0.0,
      "num_input_tokens_seen": 5906112,
      "step": 10390
    },
    {
      "epoch": 182.3716814159292,
      "grad_norm": 1.0591805903459317e-06,
      "learning_rate": 0.2527335822421971,
      "loss": 0.0,
      "num_input_tokens_seen": 5909056,
      "step": 10395
    },
    {
      "epoch": 182.46017699115043,
      "grad_norm": 5.844327688464546e-07,
      "learning_rate": 0.25269065353191444,
      "loss": 0.0,
      "num_input_tokens_seen": 5911888,
      "step": 10400
    },
    {
      "epoch": 182.46017699115043,
      "eval_loss": 0.4707694351673126,
      "eval_runtime": 0.942,
      "eval_samples_per_second": 26.54,
      "eval_steps_per_second": 13.801,
      "num_input_tokens_seen": 5911888,
      "step": 10400
    },
    {
      "epoch": 182.5486725663717,
      "grad_norm": 4.958702106705459e-07,
      "learning_rate": 0.2526477089854425,
      "loss": 0.0,
      "num_input_tokens_seen": 5914992,
      "step": 10405
    },
    {
      "epoch": 182.63716814159292,
      "grad_norm": 1.3877202036383096e-06,
      "learning_rate": 0.25260474860940385,
      "loss": 0.0,
      "num_input_tokens_seen": 5918080,
      "step": 10410
    },
    {
      "epoch": 182.72566371681415,
      "grad_norm": 9.782097549759783e-07,
      "learning_rate": 0.2525617724104236,
      "loss": 0.0,
      "num_input_tokens_seen": 5920704,
      "step": 10415
    },
    {
      "epoch": 182.81415929203538,
      "grad_norm": 4.6713273604837013e-07,
      "learning_rate": 0.25251878039512915,
      "loss": 0.0,
      "num_input_tokens_seen": 5923680,
      "step": 10420
    },
    {
      "epoch": 182.90265486725664,
      "grad_norm": 1.989144209346705e-07,
      "learning_rate": 0.25247577257015047,
      "loss": 0.0,
      "num_input_tokens_seen": 5926752,
      "step": 10425
    },
    {
      "epoch": 182.99115044247787,
      "grad_norm": 6.125210916252399e-07,
      "learning_rate": 0.2524327489421198,
      "loss": 0.0,
      "num_input_tokens_seen": 5929520,
      "step": 10430
    },
    {
      "epoch": 183.07079646017698,
      "grad_norm": 7.330514222303464e-07,
      "learning_rate": 0.25238970951767203,
      "loss": 0.0,
      "num_input_tokens_seen": 5932064,
      "step": 10435
    },
    {
      "epoch": 183.15929203539824,
      "grad_norm": 6.831447763033793e-07,
      "learning_rate": 0.25234665430344433,
      "loss": 0.0,
      "num_input_tokens_seen": 5935312,
      "step": 10440
    },
    {
      "epoch": 183.24778761061947,
      "grad_norm": 6.221484909474384e-07,
      "learning_rate": 0.2523035833060764,
      "loss": 0.0,
      "num_input_tokens_seen": 5938096,
      "step": 10445
    },
    {
      "epoch": 183.3362831858407,
      "grad_norm": 4.300223110931256e-07,
      "learning_rate": 0.2522604965322103,
      "loss": 0.0,
      "num_input_tokens_seen": 5941376,
      "step": 10450
    },
    {
      "epoch": 183.42477876106196,
      "grad_norm": 4.1622288904363813e-07,
      "learning_rate": 0.25221739398849047,
      "loss": 0.0,
      "num_input_tokens_seen": 5944368,
      "step": 10455
    },
    {
      "epoch": 183.5132743362832,
      "grad_norm": 5.485777592184604e-07,
      "learning_rate": 0.252174275681564,
      "loss": 0.0,
      "num_input_tokens_seen": 5946816,
      "step": 10460
    },
    {
      "epoch": 183.60176991150442,
      "grad_norm": 8.706378480383137e-07,
      "learning_rate": 0.2521311416180802,
      "loss": 0.0,
      "num_input_tokens_seen": 5949456,
      "step": 10465
    },
    {
      "epoch": 183.69026548672565,
      "grad_norm": 7.577494898214354e-07,
      "learning_rate": 0.25208799180469094,
      "loss": 0.0,
      "num_input_tokens_seen": 5952544,
      "step": 10470
    },
    {
      "epoch": 183.7787610619469,
      "grad_norm": 9.773308420335525e-07,
      "learning_rate": 0.2520448262480504,
      "loss": 0.0,
      "num_input_tokens_seen": 5955488,
      "step": 10475
    },
    {
      "epoch": 183.86725663716814,
      "grad_norm": 4.0792343725115643e-07,
      "learning_rate": 0.25200164495481525,
      "loss": 0.0,
      "num_input_tokens_seen": 5958112,
      "step": 10480
    },
    {
      "epoch": 183.95575221238937,
      "grad_norm": 8.518703680238104e-07,
      "learning_rate": 0.25195844793164474,
      "loss": 0.0,
      "num_input_tokens_seen": 5960688,
      "step": 10485
    },
    {
      "epoch": 184.0353982300885,
      "grad_norm": 3.7727272683696356e-07,
      "learning_rate": 0.2519152351852001,
      "loss": 0.0,
      "num_input_tokens_seen": 5963120,
      "step": 10490
    },
    {
      "epoch": 184.12389380530973,
      "grad_norm": 7.621957820447278e-07,
      "learning_rate": 0.25187200672214555,
      "loss": 0.0,
      "num_input_tokens_seen": 5965872,
      "step": 10495
    },
    {
      "epoch": 184.21238938053096,
      "grad_norm": 3.7039231415292306e-07,
      "learning_rate": 0.2518287625491473,
      "loss": 0.0,
      "num_input_tokens_seen": 5968512,
      "step": 10500
    },
    {
      "epoch": 184.30088495575222,
      "grad_norm": 6.50580147976143e-07,
      "learning_rate": 0.25178550267287425,
      "loss": 0.0,
      "num_input_tokens_seen": 5972032,
      "step": 10505
    },
    {
      "epoch": 184.38938053097345,
      "grad_norm": 3.7202289604465477e-07,
      "learning_rate": 0.2517422270999976,
      "loss": 0.0,
      "num_input_tokens_seen": 5975248,
      "step": 10510
    },
    {
      "epoch": 184.47787610619469,
      "grad_norm": 7.721676524852228e-07,
      "learning_rate": 0.2516989358371909,
      "loss": 0.0,
      "num_input_tokens_seen": 5977968,
      "step": 10515
    },
    {
      "epoch": 184.56637168141592,
      "grad_norm": 1.0812516393343685e-06,
      "learning_rate": 0.25165562889113025,
      "loss": 0.0,
      "num_input_tokens_seen": 5980656,
      "step": 10520
    },
    {
      "epoch": 184.65486725663717,
      "grad_norm": 3.520802636103326e-07,
      "learning_rate": 0.2516123062684942,
      "loss": 0.0,
      "num_input_tokens_seen": 5983440,
      "step": 10525
    },
    {
      "epoch": 184.7433628318584,
      "grad_norm": 6.168631330183416e-07,
      "learning_rate": 0.25156896797596356,
      "loss": 0.0,
      "num_input_tokens_seen": 5986064,
      "step": 10530
    },
    {
      "epoch": 184.83185840707964,
      "grad_norm": 8.278086056634493e-07,
      "learning_rate": 0.2515256140202216,
      "loss": 0.0,
      "num_input_tokens_seen": 5988640,
      "step": 10535
    },
    {
      "epoch": 184.9203539823009,
      "grad_norm": 9.347334071208024e-07,
      "learning_rate": 0.25148224440795425,
      "loss": 0.0,
      "num_input_tokens_seen": 5991168,
      "step": 10540
    },
    {
      "epoch": 185.0,
      "grad_norm": 4.119514869671548e-06,
      "learning_rate": 0.2514388591458494,
      "loss": 0.0,
      "num_input_tokens_seen": 5993944,
      "step": 10545
    },
    {
      "epoch": 185.08849557522123,
      "grad_norm": 9.773356168807368e-07,
      "learning_rate": 0.2513954582405977,
      "loss": 0.0,
      "num_input_tokens_seen": 5996888,
      "step": 10550
    },
    {
      "epoch": 185.1769911504425,
      "grad_norm": 8.215463935812295e-07,
      "learning_rate": 0.2513520416988922,
      "loss": 0.0,
      "num_input_tokens_seen": 6000120,
      "step": 10555
    },
    {
      "epoch": 185.26548672566372,
      "grad_norm": 5.190559022594243e-07,
      "learning_rate": 0.2513086095274281,
      "loss": 0.0,
      "num_input_tokens_seen": 6003112,
      "step": 10560
    },
    {
      "epoch": 185.35398230088495,
      "grad_norm": 2.912716183800512e-07,
      "learning_rate": 0.25126516173290336,
      "loss": 0.0,
      "num_input_tokens_seen": 6005704,
      "step": 10565
    },
    {
      "epoch": 185.44247787610618,
      "grad_norm": 4.3505318103598256e-07,
      "learning_rate": 0.2512216983220181,
      "loss": 0.0,
      "num_input_tokens_seen": 6008600,
      "step": 10570
    },
    {
      "epoch": 185.53097345132744,
      "grad_norm": 3.2258657256534207e-07,
      "learning_rate": 0.25117821930147494,
      "loss": 0.0,
      "num_input_tokens_seen": 6011480,
      "step": 10575
    },
    {
      "epoch": 185.61946902654867,
      "grad_norm": 6.446313705055218e-07,
      "learning_rate": 0.2511347246779788,
      "loss": 0.0,
      "num_input_tokens_seen": 6014408,
      "step": 10580
    },
    {
      "epoch": 185.7079646017699,
      "grad_norm": 5.882522486899688e-07,
      "learning_rate": 0.25109121445823723,
      "loss": 0.0,
      "num_input_tokens_seen": 6017176,
      "step": 10585
    },
    {
      "epoch": 185.79646017699116,
      "grad_norm": 4.255907697370276e-07,
      "learning_rate": 0.25104768864896004,
      "loss": 0.0,
      "num_input_tokens_seen": 6019752,
      "step": 10590
    },
    {
      "epoch": 185.8849557522124,
      "grad_norm": 2.937666749858181e-07,
      "learning_rate": 0.2510041472568594,
      "loss": 0.0,
      "num_input_tokens_seen": 6022696,
      "step": 10595
    },
    {
      "epoch": 185.97345132743362,
      "grad_norm": 3.41511196211286e-07,
      "learning_rate": 0.25096059028864987,
      "loss": 0.0,
      "num_input_tokens_seen": 6025544,
      "step": 10600
    },
    {
      "epoch": 185.97345132743362,
      "eval_loss": 0.47358882427215576,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.588,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 6025544,
      "step": 10600
    },
    {
      "epoch": 186.05309734513276,
      "grad_norm": 7.141936180232733e-07,
      "learning_rate": 0.25091701775104863,
      "loss": 0.0,
      "num_input_tokens_seen": 6027568,
      "step": 10605
    },
    {
      "epoch": 186.141592920354,
      "grad_norm": 9.398655151926505e-07,
      "learning_rate": 0.250873429650775,
      "loss": 0.0,
      "num_input_tokens_seen": 6030480,
      "step": 10610
    },
    {
      "epoch": 186.23008849557522,
      "grad_norm": 5.027687279834936e-07,
      "learning_rate": 0.25082982599455095,
      "loss": 0.0,
      "num_input_tokens_seen": 6033408,
      "step": 10615
    },
    {
      "epoch": 186.31858407079645,
      "grad_norm": 3.0971995101936045e-07,
      "learning_rate": 0.2507862067891006,
      "loss": 0.0,
      "num_input_tokens_seen": 6036192,
      "step": 10620
    },
    {
      "epoch": 186.4070796460177,
      "grad_norm": 4.382305860417546e-07,
      "learning_rate": 0.25074257204115064,
      "loss": 0.0,
      "num_input_tokens_seen": 6038880,
      "step": 10625
    },
    {
      "epoch": 186.49557522123894,
      "grad_norm": 4.243970295192412e-07,
      "learning_rate": 0.25069892175742997,
      "loss": 0.0,
      "num_input_tokens_seen": 6041856,
      "step": 10630
    },
    {
      "epoch": 186.58407079646017,
      "grad_norm": 6.796063303227129e-07,
      "learning_rate": 0.25065525594467014,
      "loss": 0.0,
      "num_input_tokens_seen": 6044752,
      "step": 10635
    },
    {
      "epoch": 186.67256637168143,
      "grad_norm": 8.313458579323196e-07,
      "learning_rate": 0.2506115746096049,
      "loss": 0.0,
      "num_input_tokens_seen": 6047632,
      "step": 10640
    },
    {
      "epoch": 186.76106194690266,
      "grad_norm": 1.4733642217379384e-07,
      "learning_rate": 0.25056787775897055,
      "loss": 0.0,
      "num_input_tokens_seen": 6050464,
      "step": 10645
    },
    {
      "epoch": 186.8495575221239,
      "grad_norm": 5.794728963337548e-07,
      "learning_rate": 0.2505241653995056,
      "loss": 0.0,
      "num_input_tokens_seen": 6053968,
      "step": 10650
    },
    {
      "epoch": 186.93805309734512,
      "grad_norm": 2.1404898120636062e-07,
      "learning_rate": 0.25048043753795113,
      "loss": 0.0,
      "num_input_tokens_seen": 6056832,
      "step": 10655
    },
    {
      "epoch": 187.01769911504425,
      "grad_norm": 6.169322546156764e-07,
      "learning_rate": 0.2504366941810504,
      "loss": 0.0,
      "num_input_tokens_seen": 6059416,
      "step": 10660
    },
    {
      "epoch": 187.10619469026548,
      "grad_norm": 2.6862571189667506e-07,
      "learning_rate": 0.2503929353355493,
      "loss": 0.0,
      "num_input_tokens_seen": 6061800,
      "step": 10665
    },
    {
      "epoch": 187.1946902654867,
      "grad_norm": 2.2368506336079008e-07,
      "learning_rate": 0.250349161008196,
      "loss": 0.0,
      "num_input_tokens_seen": 6065224,
      "step": 10670
    },
    {
      "epoch": 187.28318584070797,
      "grad_norm": 3.9494253201155516e-07,
      "learning_rate": 0.2503053712057409,
      "loss": 0.0,
      "num_input_tokens_seen": 6068200,
      "step": 10675
    },
    {
      "epoch": 187.3716814159292,
      "grad_norm": 7.574935523280146e-08,
      "learning_rate": 0.25026156593493715,
      "loss": 0.0,
      "num_input_tokens_seen": 6071112,
      "step": 10680
    },
    {
      "epoch": 187.46017699115043,
      "grad_norm": 4.361864682778105e-07,
      "learning_rate": 0.2502177452025399,
      "loss": 0.0,
      "num_input_tokens_seen": 6074088,
      "step": 10685
    },
    {
      "epoch": 187.5486725663717,
      "grad_norm": 1.7279249675539177e-07,
      "learning_rate": 0.25017390901530695,
      "loss": 0.0,
      "num_input_tokens_seen": 6076520,
      "step": 10690
    },
    {
      "epoch": 187.63716814159292,
      "grad_norm": 1.795526429759775e-07,
      "learning_rate": 0.2501300573799984,
      "loss": 0.0,
      "num_input_tokens_seen": 6079496,
      "step": 10695
    },
    {
      "epoch": 187.72566371681415,
      "grad_norm": 3.1657737054047175e-07,
      "learning_rate": 0.2500861903033766,
      "loss": 0.0,
      "num_input_tokens_seen": 6082168,
      "step": 10700
    },
    {
      "epoch": 187.81415929203538,
      "grad_norm": 1.2217277571835439e-06,
      "learning_rate": 0.25004230779220654,
      "loss": 0.0,
      "num_input_tokens_seen": 6085144,
      "step": 10705
    },
    {
      "epoch": 187.90265486725664,
      "grad_norm": 3.095866816238413e-07,
      "learning_rate": 0.24999840985325542,
      "loss": 0.0,
      "num_input_tokens_seen": 6088328,
      "step": 10710
    },
    {
      "epoch": 187.99115044247787,
      "grad_norm": 8.483309557050234e-07,
      "learning_rate": 0.24995449649329285,
      "loss": 0.0,
      "num_input_tokens_seen": 6090984,
      "step": 10715
    },
    {
      "epoch": 188.07079646017698,
      "grad_norm": 4.945620162288833e-07,
      "learning_rate": 0.2499105677190908,
      "loss": 0.0,
      "num_input_tokens_seen": 6093456,
      "step": 10720
    },
    {
      "epoch": 188.15929203539824,
      "grad_norm": 4.1252650362366694e-07,
      "learning_rate": 0.24986662353742364,
      "loss": 0.0,
      "num_input_tokens_seen": 6096960,
      "step": 10725
    },
    {
      "epoch": 188.24778761061947,
      "grad_norm": 1.4448303886638314e-07,
      "learning_rate": 0.24982266395506814,
      "loss": 0.0,
      "num_input_tokens_seen": 6099568,
      "step": 10730
    },
    {
      "epoch": 188.3362831858407,
      "grad_norm": 7.48989066323702e-07,
      "learning_rate": 0.2497786889788034,
      "loss": 0.0,
      "num_input_tokens_seen": 6102288,
      "step": 10735
    },
    {
      "epoch": 188.42477876106196,
      "grad_norm": 2.1444556352889776e-07,
      "learning_rate": 0.24973469861541095,
      "loss": 0.0,
      "num_input_tokens_seen": 6105888,
      "step": 10740
    },
    {
      "epoch": 188.5132743362832,
      "grad_norm": 3.073825212140946e-07,
      "learning_rate": 0.24969069287167456,
      "loss": 0.0,
      "num_input_tokens_seen": 6108912,
      "step": 10745
    },
    {
      "epoch": 188.60176991150442,
      "grad_norm": 4.427854491950711e-07,
      "learning_rate": 0.2496466717543806,
      "loss": 0.0,
      "num_input_tokens_seen": 6111760,
      "step": 10750
    },
    {
      "epoch": 188.69026548672565,
      "grad_norm": 4.917750402455567e-07,
      "learning_rate": 0.24960263527031762,
      "loss": 0.0,
      "num_input_tokens_seen": 6114768,
      "step": 10755
    },
    {
      "epoch": 188.7787610619469,
      "grad_norm": 6.384632911249355e-07,
      "learning_rate": 0.24955858342627657,
      "loss": 0.0,
      "num_input_tokens_seen": 6116992,
      "step": 10760
    },
    {
      "epoch": 188.86725663716814,
      "grad_norm": 5.255380983726354e-07,
      "learning_rate": 0.24951451622905083,
      "loss": 0.0,
      "num_input_tokens_seen": 6119472,
      "step": 10765
    },
    {
      "epoch": 188.95575221238937,
      "grad_norm": 1.154272922576638e-06,
      "learning_rate": 0.24947043368543612,
      "loss": 0.0,
      "num_input_tokens_seen": 6122544,
      "step": 10770
    },
    {
      "epoch": 189.0353982300885,
      "grad_norm": 4.429795126270619e-07,
      "learning_rate": 0.2494263358022305,
      "loss": 0.0,
      "num_input_tokens_seen": 6125072,
      "step": 10775
    },
    {
      "epoch": 189.12389380530973,
      "grad_norm": 2.4528475250917836e-07,
      "learning_rate": 0.24938222258623444,
      "loss": 0.0,
      "num_input_tokens_seen": 6128112,
      "step": 10780
    },
    {
      "epoch": 189.21238938053096,
      "grad_norm": 1.6990765061564161e-07,
      "learning_rate": 0.24933809404425075,
      "loss": 0.0,
      "num_input_tokens_seen": 6130800,
      "step": 10785
    },
    {
      "epoch": 189.30088495575222,
      "grad_norm": 3.0469183798231825e-07,
      "learning_rate": 0.24929395018308453,
      "loss": 0.0,
      "num_input_tokens_seen": 6133456,
      "step": 10790
    },
    {
      "epoch": 189.38938053097345,
      "grad_norm": 3.0293321628960257e-07,
      "learning_rate": 0.24924979100954348,
      "loss": 0.0,
      "num_input_tokens_seen": 6136384,
      "step": 10795
    },
    {
      "epoch": 189.47787610619469,
      "grad_norm": 2.8457790790525905e-07,
      "learning_rate": 0.24920561653043735,
      "loss": 0.0,
      "num_input_tokens_seen": 6139264,
      "step": 10800
    },
    {
      "epoch": 189.47787610619469,
      "eval_loss": 0.47310909628868103,
      "eval_runtime": 0.9433,
      "eval_samples_per_second": 26.504,
      "eval_steps_per_second": 13.782,
      "num_input_tokens_seen": 6139264,
      "step": 10800
    },
    {
      "epoch": 189.56637168141592,
      "grad_norm": 3.9557707509629836e-07,
      "learning_rate": 0.24916142675257846,
      "loss": 0.0,
      "num_input_tokens_seen": 6142000,
      "step": 10805
    },
    {
      "epoch": 189.65486725663717,
      "grad_norm": 3.168356101923564e-07,
      "learning_rate": 0.24911722168278144,
      "loss": 0.0,
      "num_input_tokens_seen": 6144832,
      "step": 10810
    },
    {
      "epoch": 189.7433628318584,
      "grad_norm": 3.8287581105578283e-07,
      "learning_rate": 0.24907300132786328,
      "loss": 0.0,
      "num_input_tokens_seen": 6147456,
      "step": 10815
    },
    {
      "epoch": 189.83185840707964,
      "grad_norm": 1.1402554491724004e-06,
      "learning_rate": 0.24902876569464322,
      "loss": 0.0,
      "num_input_tokens_seen": 6150800,
      "step": 10820
    },
    {
      "epoch": 189.9203539823009,
      "grad_norm": 6.70919177991891e-07,
      "learning_rate": 0.24898451478994305,
      "loss": 0.0,
      "num_input_tokens_seen": 6154256,
      "step": 10825
    },
    {
      "epoch": 190.0,
      "grad_norm": 7.349685802182648e-07,
      "learning_rate": 0.2489402486205868,
      "loss": 0.0,
      "num_input_tokens_seen": 6156512,
      "step": 10830
    },
    {
      "epoch": 190.08849557522123,
      "grad_norm": 4.452994630810281e-07,
      "learning_rate": 0.24889596719340085,
      "loss": 0.0,
      "num_input_tokens_seen": 6159904,
      "step": 10835
    },
    {
      "epoch": 190.1769911504425,
      "grad_norm": 8.583847943555156e-08,
      "learning_rate": 0.24885167051521392,
      "loss": 0.0,
      "num_input_tokens_seen": 6162944,
      "step": 10840
    },
    {
      "epoch": 190.26548672566372,
      "grad_norm": 1.063407239598746e-06,
      "learning_rate": 0.24880735859285716,
      "loss": 0.0,
      "num_input_tokens_seen": 6165680,
      "step": 10845
    },
    {
      "epoch": 190.35398230088495,
      "grad_norm": 6.24448659891641e-07,
      "learning_rate": 0.24876303143316406,
      "loss": 0.0,
      "num_input_tokens_seen": 6168448,
      "step": 10850
    },
    {
      "epoch": 190.44247787610618,
      "grad_norm": 4.4572138335752243e-07,
      "learning_rate": 0.24871868904297031,
      "loss": 0.0,
      "num_input_tokens_seen": 6171136,
      "step": 10855
    },
    {
      "epoch": 190.53097345132744,
      "grad_norm": 1.3379678875935497e-07,
      "learning_rate": 0.24867433142911416,
      "loss": 0.0,
      "num_input_tokens_seen": 6173744,
      "step": 10860
    },
    {
      "epoch": 190.61946902654867,
      "grad_norm": 3.1760416163706395e-07,
      "learning_rate": 0.24862995859843612,
      "loss": 0.0,
      "num_input_tokens_seen": 6176416,
      "step": 10865
    },
    {
      "epoch": 190.7079646017699,
      "grad_norm": 5.91529044413619e-07,
      "learning_rate": 0.24858557055777897,
      "loss": 0.0,
      "num_input_tokens_seen": 6179328,
      "step": 10870
    },
    {
      "epoch": 190.79646017699116,
      "grad_norm": 5.314742566042696e-07,
      "learning_rate": 0.24854116731398793,
      "loss": 0.0,
      "num_input_tokens_seen": 6182416,
      "step": 10875
    },
    {
      "epoch": 190.8849557522124,
      "grad_norm": 5.482298206516134e-07,
      "learning_rate": 0.24849674887391052,
      "loss": 0.0,
      "num_input_tokens_seen": 6185072,
      "step": 10880
    },
    {
      "epoch": 190.97345132743362,
      "grad_norm": 4.6992704483272973e-07,
      "learning_rate": 0.2484523152443967,
      "loss": 0.0,
      "num_input_tokens_seen": 6187904,
      "step": 10885
    },
    {
      "epoch": 191.05309734513276,
      "grad_norm": 8.088842378128902e-07,
      "learning_rate": 0.24840786643229862,
      "loss": 0.0,
      "num_input_tokens_seen": 6190096,
      "step": 10890
    },
    {
      "epoch": 191.141592920354,
      "grad_norm": 1.7810582164656807e-07,
      "learning_rate": 0.2483634024444709,
      "loss": 0.0,
      "num_input_tokens_seen": 6193072,
      "step": 10895
    },
    {
      "epoch": 191.23008849557522,
      "grad_norm": 2.264718119704412e-07,
      "learning_rate": 0.24831892328777033,
      "loss": 0.0,
      "num_input_tokens_seen": 6195648,
      "step": 10900
    },
    {
      "epoch": 191.31858407079645,
      "grad_norm": 3.506106622808147e-07,
      "learning_rate": 0.2482744289690563,
      "loss": 0.0,
      "num_input_tokens_seen": 6198896,
      "step": 10905
    },
    {
      "epoch": 191.4070796460177,
      "grad_norm": 3.3023960099853866e-07,
      "learning_rate": 0.2482299194951903,
      "loss": 0.0,
      "num_input_tokens_seen": 6201488,
      "step": 10910
    },
    {
      "epoch": 191.49557522123894,
      "grad_norm": 1.1667943908832967e-06,
      "learning_rate": 0.2481853948730363,
      "loss": 0.0,
      "num_input_tokens_seen": 6204720,
      "step": 10915
    },
    {
      "epoch": 191.58407079646017,
      "grad_norm": 2.918106076776894e-07,
      "learning_rate": 0.24814085510946052,
      "loss": 0.0,
      "num_input_tokens_seen": 6207168,
      "step": 10920
    },
    {
      "epoch": 191.67256637168143,
      "grad_norm": 4.5092676259628206e-07,
      "learning_rate": 0.24809630021133158,
      "loss": 0.0,
      "num_input_tokens_seen": 6210000,
      "step": 10925
    },
    {
      "epoch": 191.76106194690266,
      "grad_norm": 2.7836446747642185e-07,
      "learning_rate": 0.24805173018552037,
      "loss": 0.0,
      "num_input_tokens_seen": 6212704,
      "step": 10930
    },
    {
      "epoch": 191.8495575221239,
      "grad_norm": 3.0935845529711514e-07,
      "learning_rate": 0.2480071450389002,
      "loss": 0.0,
      "num_input_tokens_seen": 6215680,
      "step": 10935
    },
    {
      "epoch": 191.93805309734512,
      "grad_norm": 6.661985594291764e-07,
      "learning_rate": 0.24796254477834662,
      "loss": 0.0,
      "num_input_tokens_seen": 6218800,
      "step": 10940
    },
    {
      "epoch": 192.01769911504425,
      "grad_norm": 6.033015438333678e-07,
      "learning_rate": 0.24791792941073754,
      "loss": 0.0,
      "num_input_tokens_seen": 6221184,
      "step": 10945
    },
    {
      "epoch": 192.10619469026548,
      "grad_norm": 4.358778085133963e-07,
      "learning_rate": 0.2478732989429533,
      "loss": 0.0,
      "num_input_tokens_seen": 6223936,
      "step": 10950
    },
    {
      "epoch": 192.1946902654867,
      "grad_norm": 4.037546830204519e-07,
      "learning_rate": 0.24782865338187632,
      "loss": 0.0,
      "num_input_tokens_seen": 6227248,
      "step": 10955
    },
    {
      "epoch": 192.28318584070797,
      "grad_norm": 2.5660742153377214e-07,
      "learning_rate": 0.2477839927343916,
      "loss": 0.0,
      "num_input_tokens_seen": 6229952,
      "step": 10960
    },
    {
      "epoch": 192.3716814159292,
      "grad_norm": 3.144311904179631e-07,
      "learning_rate": 0.2477393170073864,
      "loss": 0.0,
      "num_input_tokens_seen": 6232816,
      "step": 10965
    },
    {
      "epoch": 192.46017699115043,
      "grad_norm": 4.99984196267178e-07,
      "learning_rate": 0.2476946262077503,
      "loss": 0.0,
      "num_input_tokens_seen": 6235552,
      "step": 10970
    },
    {
      "epoch": 192.5486725663717,
      "grad_norm": 2.242415746422921e-07,
      "learning_rate": 0.24764992034237507,
      "loss": 0.0,
      "num_input_tokens_seen": 6238512,
      "step": 10975
    },
    {
      "epoch": 192.63716814159292,
      "grad_norm": 3.632996481428563e-07,
      "learning_rate": 0.24760519941815498,
      "loss": 0.0,
      "num_input_tokens_seen": 6241136,
      "step": 10980
    },
    {
      "epoch": 192.72566371681415,
      "grad_norm": 1.8644517751909007e-07,
      "learning_rate": 0.2475604634419866,
      "loss": 0.0,
      "num_input_tokens_seen": 6244112,
      "step": 10985
    },
    {
      "epoch": 192.81415929203538,
      "grad_norm": 4.404055005124974e-07,
      "learning_rate": 0.24751571242076872,
      "loss": 0.0,
      "num_input_tokens_seen": 6247248,
      "step": 10990
    },
    {
      "epoch": 192.90265486725664,
      "grad_norm": 1.6823021553591389e-07,
      "learning_rate": 0.2474709463614025,
      "loss": 0.0,
      "num_input_tokens_seen": 6250080,
      "step": 10995
    },
    {
      "epoch": 192.99115044247787,
      "grad_norm": 5.242434326646617e-07,
      "learning_rate": 0.24742616527079145,
      "loss": 0.0,
      "num_input_tokens_seen": 6252832,
      "step": 11000
    },
    {
      "epoch": 192.99115044247787,
      "eval_loss": 0.47574570775032043,
      "eval_runtime": 0.9371,
      "eval_samples_per_second": 26.679,
      "eval_steps_per_second": 13.873,
      "num_input_tokens_seen": 6252832,
      "step": 11000
    },
    {
      "epoch": 193.07079646017698,
      "grad_norm": 1.8304254467693681e-07,
      "learning_rate": 0.24738136915584139,
      "loss": 0.0,
      "num_input_tokens_seen": 6254912,
      "step": 11005
    },
    {
      "epoch": 193.15929203539824,
      "grad_norm": 5.626996539831453e-07,
      "learning_rate": 0.24733655802346047,
      "loss": 0.0,
      "num_input_tokens_seen": 6257696,
      "step": 11010
    },
    {
      "epoch": 193.24778761061947,
      "grad_norm": 5.216946874497808e-07,
      "learning_rate": 0.24729173188055906,
      "loss": 0.0,
      "num_input_tokens_seen": 6260304,
      "step": 11015
    },
    {
      "epoch": 193.3362831858407,
      "grad_norm": 7.933174401841825e-07,
      "learning_rate": 0.24724689073404996,
      "loss": 0.0,
      "num_input_tokens_seen": 6263312,
      "step": 11020
    },
    {
      "epoch": 193.42477876106196,
      "grad_norm": 2.512471439786168e-07,
      "learning_rate": 0.24720203459084822,
      "loss": 0.0,
      "num_input_tokens_seen": 6266928,
      "step": 11025
    },
    {
      "epoch": 193.5132743362832,
      "grad_norm": 3.270527884069452e-07,
      "learning_rate": 0.24715716345787123,
      "loss": 0.0,
      "num_input_tokens_seen": 6269808,
      "step": 11030
    },
    {
      "epoch": 193.60176991150442,
      "grad_norm": 1.2024506190755346e-07,
      "learning_rate": 0.2471122773420387,
      "loss": 0.0,
      "num_input_tokens_seen": 6272512,
      "step": 11035
    },
    {
      "epoch": 193.69026548672565,
      "grad_norm": 3.6714033058160567e-07,
      "learning_rate": 0.24706737625027259,
      "loss": 0.0,
      "num_input_tokens_seen": 6275904,
      "step": 11040
    },
    {
      "epoch": 193.7787610619469,
      "grad_norm": 2.5506221845716937e-07,
      "learning_rate": 0.24702246018949725,
      "loss": 0.0,
      "num_input_tokens_seen": 6279312,
      "step": 11045
    },
    {
      "epoch": 193.86725663716814,
      "grad_norm": 5.016304953642248e-07,
      "learning_rate": 0.2469775291666393,
      "loss": 0.0,
      "num_input_tokens_seen": 6281760,
      "step": 11050
    },
    {
      "epoch": 193.95575221238937,
      "grad_norm": 5.275255148262659e-07,
      "learning_rate": 0.24693258318862765,
      "loss": 0.0,
      "num_input_tokens_seen": 6284656,
      "step": 11055
    },
    {
      "epoch": 194.0353982300885,
      "grad_norm": 2.295623602321939e-07,
      "learning_rate": 0.2468876222623935,
      "loss": 0.0,
      "num_input_tokens_seen": 6286984,
      "step": 11060
    },
    {
      "epoch": 194.12389380530973,
      "grad_norm": 5.454732558973774e-07,
      "learning_rate": 0.2468426463948705,
      "loss": 0.0,
      "num_input_tokens_seen": 6289928,
      "step": 11065
    },
    {
      "epoch": 194.21238938053096,
      "grad_norm": 4.812580982616055e-07,
      "learning_rate": 0.24679765559299438,
      "loss": 0.0,
      "num_input_tokens_seen": 6292840,
      "step": 11070
    },
    {
      "epoch": 194.30088495575222,
      "grad_norm": 3.1782917631062446e-07,
      "learning_rate": 0.24675264986370332,
      "loss": 0.0,
      "num_input_tokens_seen": 6295624,
      "step": 11075
    },
    {
      "epoch": 194.38938053097345,
      "grad_norm": 3.3239402341678215e-07,
      "learning_rate": 0.2467076292139378,
      "loss": 0.0,
      "num_input_tokens_seen": 6298104,
      "step": 11080
    },
    {
      "epoch": 194.47787610619469,
      "grad_norm": 4.2444514747330686e-07,
      "learning_rate": 0.24666259365064055,
      "loss": 0.0,
      "num_input_tokens_seen": 6300776,
      "step": 11085
    },
    {
      "epoch": 194.56637168141592,
      "grad_norm": 5.164253593648027e-07,
      "learning_rate": 0.24661754318075663,
      "loss": 0.0,
      "num_input_tokens_seen": 6303560,
      "step": 11090
    },
    {
      "epoch": 194.65486725663717,
      "grad_norm": 1.2986433262085484e-07,
      "learning_rate": 0.2465724778112334,
      "loss": 0.0,
      "num_input_tokens_seen": 6306280,
      "step": 11095
    },
    {
      "epoch": 194.7433628318584,
      "grad_norm": 3.0520570248881995e-07,
      "learning_rate": 0.24652739754902042,
      "loss": 0.0,
      "num_input_tokens_seen": 6309400,
      "step": 11100
    },
    {
      "epoch": 194.83185840707964,
      "grad_norm": 3.675654909329751e-07,
      "learning_rate": 0.24648230240106975,
      "loss": 0.0,
      "num_input_tokens_seen": 6312760,
      "step": 11105
    },
    {
      "epoch": 194.9203539823009,
      "grad_norm": 1.2132759366068058e-06,
      "learning_rate": 0.2464371923743356,
      "loss": 0.0,
      "num_input_tokens_seen": 6315624,
      "step": 11110
    },
    {
      "epoch": 195.0,
      "grad_norm": 8.689604555911501e-07,
      "learning_rate": 0.24639206747577444,
      "loss": 0.0,
      "num_input_tokens_seen": 6318352,
      "step": 11115
    },
    {
      "epoch": 195.08849557522123,
      "grad_norm": 1.115298076115323e-07,
      "learning_rate": 0.24634692771234515,
      "loss": 0.0,
      "num_input_tokens_seen": 6321360,
      "step": 11120
    },
    {
      "epoch": 195.1769911504425,
      "grad_norm": 1.6852425233082613e-07,
      "learning_rate": 0.2463017730910088,
      "loss": 0.0,
      "num_input_tokens_seen": 6324256,
      "step": 11125
    },
    {
      "epoch": 195.26548672566372,
      "grad_norm": 3.0951372309573344e-07,
      "learning_rate": 0.2462566036187289,
      "loss": 0.0,
      "num_input_tokens_seen": 6326896,
      "step": 11130
    },
    {
      "epoch": 195.35398230088495,
      "grad_norm": 3.50245045410702e-07,
      "learning_rate": 0.24621141930247106,
      "loss": 0.0,
      "num_input_tokens_seen": 6329600,
      "step": 11135
    },
    {
      "epoch": 195.44247787610618,
      "grad_norm": 1.501759356870025e-07,
      "learning_rate": 0.2461662201492033,
      "loss": 0.0,
      "num_input_tokens_seen": 6332768,
      "step": 11140
    },
    {
      "epoch": 195.53097345132744,
      "grad_norm": 4.48300284006109e-07,
      "learning_rate": 0.24612100616589586,
      "loss": 0.0,
      "num_input_tokens_seen": 6335536,
      "step": 11145
    },
    {
      "epoch": 195.61946902654867,
      "grad_norm": 1.1534446286987077e-07,
      "learning_rate": 0.24607577735952135,
      "loss": 0.0,
      "num_input_tokens_seen": 6338064,
      "step": 11150
    },
    {
      "epoch": 195.7079646017699,
      "grad_norm": 6.801561198699346e-07,
      "learning_rate": 0.24603053373705464,
      "loss": 0.0,
      "num_input_tokens_seen": 6341056,
      "step": 11155
    },
    {
      "epoch": 195.79646017699116,
      "grad_norm": 3.6943600889571826e-07,
      "learning_rate": 0.2459852753054728,
      "loss": 0.0,
      "num_input_tokens_seen": 6343856,
      "step": 11160
    },
    {
      "epoch": 195.8849557522124,
      "grad_norm": 2.777157419586729e-07,
      "learning_rate": 0.24594000207175526,
      "loss": 0.0,
      "num_input_tokens_seen": 6346928,
      "step": 11165
    },
    {
      "epoch": 195.97345132743362,
      "grad_norm": 7.308519798243651e-07,
      "learning_rate": 0.2458947140428838,
      "loss": 0.0,
      "num_input_tokens_seen": 6349888,
      "step": 11170
    },
    {
      "epoch": 196.05309734513276,
      "grad_norm": 4.2593634930199187e-07,
      "learning_rate": 0.24584941122584233,
      "loss": 0.0,
      "num_input_tokens_seen": 6352344,
      "step": 11175
    },
    {
      "epoch": 196.141592920354,
      "grad_norm": 2.902134212945384e-07,
      "learning_rate": 0.24580409362761713,
      "loss": 0.0,
      "num_input_tokens_seen": 6355608,
      "step": 11180
    },
    {
      "epoch": 196.23008849557522,
      "grad_norm": 6.046739713383431e-07,
      "learning_rate": 0.2457587612551967,
      "loss": 0.0,
      "num_input_tokens_seen": 6358552,
      "step": 11185
    },
    {
      "epoch": 196.31858407079645,
      "grad_norm": 6.458273560383532e-07,
      "learning_rate": 0.24571341411557193,
      "loss": 0.0,
      "num_input_tokens_seen": 6360840,
      "step": 11190
    },
    {
      "epoch": 196.4070796460177,
      "grad_norm": 1.0938633465684688e-07,
      "learning_rate": 0.2456680522157359,
      "loss": 0.0,
      "num_input_tokens_seen": 6363720,
      "step": 11195
    },
    {
      "epoch": 196.49557522123894,
      "grad_norm": 4.265176585249719e-07,
      "learning_rate": 0.245622675562684,
      "loss": 0.0,
      "num_input_tokens_seen": 6366440,
      "step": 11200
    },
    {
      "epoch": 196.49557522123894,
      "eval_loss": 0.48474833369255066,
      "eval_runtime": 0.9172,
      "eval_samples_per_second": 27.256,
      "eval_steps_per_second": 14.173,
      "num_input_tokens_seen": 6366440,
      "step": 11200
    },
    {
      "epoch": 196.58407079646017,
      "grad_norm": 1.1758405662476434e-06,
      "learning_rate": 0.24557728416341384,
      "loss": 0.0,
      "num_input_tokens_seen": 6369384,
      "step": 11205
    },
    {
      "epoch": 196.67256637168143,
      "grad_norm": 6.129566827439703e-07,
      "learning_rate": 0.24553187802492538,
      "loss": 0.0,
      "num_input_tokens_seen": 6372072,
      "step": 11210
    },
    {
      "epoch": 196.76106194690266,
      "grad_norm": 3.5082680938103294e-07,
      "learning_rate": 0.24548645715422074,
      "loss": 0.0,
      "num_input_tokens_seen": 6374968,
      "step": 11215
    },
    {
      "epoch": 196.8495575221239,
      "grad_norm": 4.1631631120253587e-07,
      "learning_rate": 0.2454410215583045,
      "loss": 0.0,
      "num_input_tokens_seen": 6377592,
      "step": 11220
    },
    {
      "epoch": 196.93805309734512,
      "grad_norm": 5.586352926911786e-07,
      "learning_rate": 0.24539557124418332,
      "loss": 0.0,
      "num_input_tokens_seen": 6380888,
      "step": 11225
    },
    {
      "epoch": 197.01769911504425,
      "grad_norm": 2.568944807990192e-07,
      "learning_rate": 0.24535010621886624,
      "loss": 0.0,
      "num_input_tokens_seen": 6383104,
      "step": 11230
    },
    {
      "epoch": 197.10619469026548,
      "grad_norm": 3.545693516571191e-07,
      "learning_rate": 0.2453046264893646,
      "loss": 0.0,
      "num_input_tokens_seen": 6386400,
      "step": 11235
    },
    {
      "epoch": 197.1946902654867,
      "grad_norm": 5.338868618309789e-07,
      "learning_rate": 0.24525913206269184,
      "loss": 0.0,
      "num_input_tokens_seen": 6389392,
      "step": 11240
    },
    {
      "epoch": 197.28318584070797,
      "grad_norm": 1.4497980771466246e-07,
      "learning_rate": 0.2452136229458638,
      "loss": 0.0,
      "num_input_tokens_seen": 6391968,
      "step": 11245
    },
    {
      "epoch": 197.3716814159292,
      "grad_norm": 1.6952149906046543e-07,
      "learning_rate": 0.24516809914589857,
      "loss": 0.0,
      "num_input_tokens_seen": 6394720,
      "step": 11250
    },
    {
      "epoch": 197.46017699115043,
      "grad_norm": 2.4893796535252477e-07,
      "learning_rate": 0.2451225606698165,
      "loss": 0.0,
      "num_input_tokens_seen": 6397808,
      "step": 11255
    },
    {
      "epoch": 197.5486725663717,
      "grad_norm": 8.085440583727177e-08,
      "learning_rate": 0.2450770075246402,
      "loss": 0.0,
      "num_input_tokens_seen": 6400448,
      "step": 11260
    },
    {
      "epoch": 197.63716814159292,
      "grad_norm": 1.0553313245509344e-07,
      "learning_rate": 0.24503143971739455,
      "loss": 0.0,
      "num_input_tokens_seen": 6403136,
      "step": 11265
    },
    {
      "epoch": 197.72566371681415,
      "grad_norm": 3.5066133818872913e-07,
      "learning_rate": 0.24498585725510663,
      "loss": 0.0,
      "num_input_tokens_seen": 6406288,
      "step": 11270
    },
    {
      "epoch": 197.81415929203538,
      "grad_norm": 6.866970920782478e-07,
      "learning_rate": 0.24494026014480583,
      "loss": 0.0,
      "num_input_tokens_seen": 6408896,
      "step": 11275
    },
    {
      "epoch": 197.90265486725664,
      "grad_norm": 7.089213909239334e-07,
      "learning_rate": 0.24489464839352387,
      "loss": 0.0,
      "num_input_tokens_seen": 6411728,
      "step": 11280
    },
    {
      "epoch": 197.99115044247787,
      "grad_norm": 3.9694168663118035e-07,
      "learning_rate": 0.2448490220082946,
      "loss": 0.0,
      "num_input_tokens_seen": 6414592,
      "step": 11285
    },
    {
      "epoch": 198.07079646017698,
      "grad_norm": 5.000907208341232e-07,
      "learning_rate": 0.24480338099615415,
      "loss": 0.0,
      "num_input_tokens_seen": 6416920,
      "step": 11290
    },
    {
      "epoch": 198.15929203539824,
      "grad_norm": 3.50579909991211e-07,
      "learning_rate": 0.244757725364141,
      "loss": 0.0,
      "num_input_tokens_seen": 6419336,
      "step": 11295
    },
    {
      "epoch": 198.24778761061947,
      "grad_norm": 3.625450801791885e-07,
      "learning_rate": 0.24471205511929583,
      "loss": 0.0,
      "num_input_tokens_seen": 6422376,
      "step": 11300
    },
    {
      "epoch": 198.3362831858407,
      "grad_norm": 2.760376105470641e-07,
      "learning_rate": 0.24466637026866145,
      "loss": 0.0,
      "num_input_tokens_seen": 6425320,
      "step": 11305
    },
    {
      "epoch": 198.42477876106196,
      "grad_norm": 3.347263941577694e-07,
      "learning_rate": 0.2446206708192832,
      "loss": 0.0,
      "num_input_tokens_seen": 6428056,
      "step": 11310
    },
    {
      "epoch": 198.5132743362832,
      "grad_norm": 1.5345820258971798e-07,
      "learning_rate": 0.2445749567782084,
      "loss": 0.0,
      "num_input_tokens_seen": 6431480,
      "step": 11315
    },
    {
      "epoch": 198.60176991150442,
      "grad_norm": 1.6261822111118818e-07,
      "learning_rate": 0.2445292281524868,
      "loss": 0.0,
      "num_input_tokens_seen": 6433976,
      "step": 11320
    },
    {
      "epoch": 198.69026548672565,
      "grad_norm": 4.3261323412480124e-07,
      "learning_rate": 0.24448348494917022,
      "loss": 0.0,
      "num_input_tokens_seen": 6437080,
      "step": 11325
    },
    {
      "epoch": 198.7787610619469,
      "grad_norm": 6.27820270437951e-07,
      "learning_rate": 0.24443772717531295,
      "loss": 0.0,
      "num_input_tokens_seen": 6439896,
      "step": 11330
    },
    {
      "epoch": 198.86725663716814,
      "grad_norm": 6.888078587508062e-07,
      "learning_rate": 0.24439195483797138,
      "loss": 0.0,
      "num_input_tokens_seen": 6442888,
      "step": 11335
    },
    {
      "epoch": 198.95575221238937,
      "grad_norm": 2.819663222908275e-07,
      "learning_rate": 0.24434616794420416,
      "loss": 0.0,
      "num_input_tokens_seen": 6445448,
      "step": 11340
    },
    {
      "epoch": 199.0353982300885,
      "grad_norm": 4.622407061560807e-07,
      "learning_rate": 0.24430036650107223,
      "loss": 0.0,
      "num_input_tokens_seen": 6447712,
      "step": 11345
    },
    {
      "epoch": 199.12389380530973,
      "grad_norm": 9.098673814378344e-08,
      "learning_rate": 0.2442545505156387,
      "loss": 0.0,
      "num_input_tokens_seen": 6450336,
      "step": 11350
    },
    {
      "epoch": 199.21238938053096,
      "grad_norm": 3.381737769814208e-07,
      "learning_rate": 0.24420871999496904,
      "loss": 0.0,
      "num_input_tokens_seen": 6453584,
      "step": 11355
    },
    {
      "epoch": 199.30088495575222,
      "grad_norm": 3.6445760542846983e-07,
      "learning_rate": 0.24416287494613084,
      "loss": 0.0,
      "num_input_tokens_seen": 6456752,
      "step": 11360
    },
    {
      "epoch": 199.38938053097345,
      "grad_norm": 8.263517088380468e-07,
      "learning_rate": 0.24411701537619399,
      "loss": 0.0,
      "num_input_tokens_seen": 6459536,
      "step": 11365
    },
    {
      "epoch": 199.47787610619469,
      "grad_norm": 2.625263846312009e-07,
      "learning_rate": 0.24407114129223062,
      "loss": 0.0,
      "num_input_tokens_seen": 6462160,
      "step": 11370
    },
    {
      "epoch": 199.56637168141592,
      "grad_norm": 4.878698405264004e-07,
      "learning_rate": 0.2440252527013151,
      "loss": 0.0,
      "num_input_tokens_seen": 6465072,
      "step": 11375
    },
    {
      "epoch": 199.65486725663717,
      "grad_norm": 4.358525700354221e-07,
      "learning_rate": 0.24397934961052403,
      "loss": 0.0,
      "num_input_tokens_seen": 6467920,
      "step": 11380
    },
    {
      "epoch": 199.7433628318584,
      "grad_norm": 7.192636530817254e-07,
      "learning_rate": 0.24393343202693618,
      "loss": 0.0,
      "num_input_tokens_seen": 6470416,
      "step": 11385
    },
    {
      "epoch": 199.83185840707964,
      "grad_norm": 1.615396740817232e-07,
      "learning_rate": 0.2438874999576327,
      "loss": 0.0,
      "num_input_tokens_seen": 6473280,
      "step": 11390
    },
    {
      "epoch": 199.9203539823009,
      "grad_norm": 3.636731094047718e-07,
      "learning_rate": 0.24384155340969688,
      "loss": 0.0,
      "num_input_tokens_seen": 6476512,
      "step": 11395
    },
    {
      "epoch": 200.0,
      "grad_norm": 3.462122322162031e-07,
      "learning_rate": 0.24379559239021423,
      "loss": 0.0,
      "num_input_tokens_seen": 6478776,
      "step": 11400
    },
    {
      "epoch": 200.0,
      "eval_loss": 0.48118656873703003,
      "eval_runtime": 0.9116,
      "eval_samples_per_second": 27.425,
      "eval_steps_per_second": 14.261,
      "num_input_tokens_seen": 6478776,
      "step": 11400
    },
    {
      "epoch": 200.08849557522123,
      "grad_norm": 4.416397132445127e-07,
      "learning_rate": 0.2437496169062725,
      "loss": 0.0,
      "num_input_tokens_seen": 6481576,
      "step": 11405
    },
    {
      "epoch": 200.1769911504425,
      "grad_norm": 5.445846227303264e-07,
      "learning_rate": 0.24370362696496176,
      "loss": 0.0,
      "num_input_tokens_seen": 6484520,
      "step": 11410
    },
    {
      "epoch": 200.26548672566372,
      "grad_norm": 2.674276231573458e-07,
      "learning_rate": 0.24365762257337417,
      "loss": 0.0,
      "num_input_tokens_seen": 6487592,
      "step": 11415
    },
    {
      "epoch": 200.35398230088495,
      "grad_norm": 1.491927292818218e-07,
      "learning_rate": 0.2436116037386042,
      "loss": 0.0,
      "num_input_tokens_seen": 6490152,
      "step": 11420
    },
    {
      "epoch": 200.44247787610618,
      "grad_norm": 1.8416470481952274e-07,
      "learning_rate": 0.24356557046774852,
      "loss": 0.0,
      "num_input_tokens_seen": 6493016,
      "step": 11425
    },
    {
      "epoch": 200.53097345132744,
      "grad_norm": 4.906454478259548e-07,
      "learning_rate": 0.24351952276790606,
      "loss": 0.0,
      "num_input_tokens_seen": 6496104,
      "step": 11430
    },
    {
      "epoch": 200.61946902654867,
      "grad_norm": 2.184129783699973e-07,
      "learning_rate": 0.24347346064617797,
      "loss": 0.0,
      "num_input_tokens_seen": 6498632,
      "step": 11435
    },
    {
      "epoch": 200.7079646017699,
      "grad_norm": 1.2045755681810988e-07,
      "learning_rate": 0.24342738410966758,
      "loss": 0.0,
      "num_input_tokens_seen": 6501512,
      "step": 11440
    },
    {
      "epoch": 200.79646017699116,
      "grad_norm": 2.598614514681685e-07,
      "learning_rate": 0.24338129316548046,
      "loss": 0.0,
      "num_input_tokens_seen": 6504200,
      "step": 11445
    },
    {
      "epoch": 200.8849557522124,
      "grad_norm": 5.688012265636644e-07,
      "learning_rate": 0.24333518782072444,
      "loss": 0.0,
      "num_input_tokens_seen": 6507496,
      "step": 11450
    },
    {
      "epoch": 200.97345132743362,
      "grad_norm": 4.67436024109702e-07,
      "learning_rate": 0.24328906808250952,
      "loss": 0.0,
      "num_input_tokens_seen": 6510568,
      "step": 11455
    },
    {
      "epoch": 201.05309734513276,
      "grad_norm": 3.0688272545376094e-07,
      "learning_rate": 0.243242933957948,
      "loss": 0.0,
      "num_input_tokens_seen": 6512936,
      "step": 11460
    },
    {
      "epoch": 201.141592920354,
      "grad_norm": 2.7619387310551247e-07,
      "learning_rate": 0.24319678545415427,
      "loss": 0.0,
      "num_input_tokens_seen": 6515416,
      "step": 11465
    },
    {
      "epoch": 201.23008849557522,
      "grad_norm": 2.7301692284709134e-07,
      "learning_rate": 0.24315062257824507,
      "loss": 0.0,
      "num_input_tokens_seen": 6518744,
      "step": 11470
    },
    {
      "epoch": 201.31858407079645,
      "grad_norm": 2.3796430070888164e-07,
      "learning_rate": 0.24310444533733921,
      "loss": 0.0,
      "num_input_tokens_seen": 6521080,
      "step": 11475
    },
    {
      "epoch": 201.4070796460177,
      "grad_norm": 1.3575788671005284e-07,
      "learning_rate": 0.2430582537385579,
      "loss": 0.0,
      "num_input_tokens_seen": 6523720,
      "step": 11480
    },
    {
      "epoch": 201.49557522123894,
      "grad_norm": 4.96651011872018e-07,
      "learning_rate": 0.2430120477890244,
      "loss": 0.0,
      "num_input_tokens_seen": 6526760,
      "step": 11485
    },
    {
      "epoch": 201.58407079646017,
      "grad_norm": 1.0637005942726319e-07,
      "learning_rate": 0.24296582749586426,
      "loss": 0.0,
      "num_input_tokens_seen": 6529992,
      "step": 11490
    },
    {
      "epoch": 201.67256637168143,
      "grad_norm": 8.248828748946835e-07,
      "learning_rate": 0.24291959286620526,
      "loss": 0.0,
      "num_input_tokens_seen": 6532584,
      "step": 11495
    },
    {
      "epoch": 201.76106194690266,
      "grad_norm": 5.344727469491772e-07,
      "learning_rate": 0.24287334390717738,
      "loss": 0.0,
      "num_input_tokens_seen": 6535768,
      "step": 11500
    },
    {
      "epoch": 201.8495575221239,
      "grad_norm": 1.5034217426546093e-07,
      "learning_rate": 0.24282708062591268,
      "loss": 0.0,
      "num_input_tokens_seen": 6538728,
      "step": 11505
    },
    {
      "epoch": 201.93805309734512,
      "grad_norm": 2.912214114303424e-08,
      "learning_rate": 0.24278080302954563,
      "loss": 0.0,
      "num_input_tokens_seen": 6541320,
      "step": 11510
    },
    {
      "epoch": 202.01769911504425,
      "grad_norm": 2.110160863821875e-07,
      "learning_rate": 0.24273451112521283,
      "loss": 0.0,
      "num_input_tokens_seen": 6543928,
      "step": 11515
    },
    {
      "epoch": 202.10619469026548,
      "grad_norm": 3.68499087244345e-07,
      "learning_rate": 0.242688204920053,
      "loss": 0.0,
      "num_input_tokens_seen": 6546680,
      "step": 11520
    },
    {
      "epoch": 202.1946902654867,
      "grad_norm": 2.1557346485678863e-07,
      "learning_rate": 0.24264188442120715,
      "loss": 0.0,
      "num_input_tokens_seen": 6549400,
      "step": 11525
    },
    {
      "epoch": 202.28318584070797,
      "grad_norm": 1.2224849399444793e-07,
      "learning_rate": 0.24259554963581853,
      "loss": 0.0,
      "num_input_tokens_seen": 6552088,
      "step": 11530
    },
    {
      "epoch": 202.3716814159292,
      "grad_norm": 3.830428454421053e-07,
      "learning_rate": 0.24254920057103257,
      "loss": 0.0,
      "num_input_tokens_seen": 6555048,
      "step": 11535
    },
    {
      "epoch": 202.46017699115043,
      "grad_norm": 6.217058512447693e-07,
      "learning_rate": 0.24250283723399685,
      "loss": 0.0,
      "num_input_tokens_seen": 6558280,
      "step": 11540
    },
    {
      "epoch": 202.5486725663717,
      "grad_norm": 4.879277639702195e-07,
      "learning_rate": 0.24245645963186108,
      "loss": 0.0,
      "num_input_tokens_seen": 6561240,
      "step": 11545
    },
    {
      "epoch": 202.63716814159292,
      "grad_norm": 5.153269171387365e-07,
      "learning_rate": 0.2424100677717774,
      "loss": 0.0,
      "num_input_tokens_seen": 6564040,
      "step": 11550
    },
    {
      "epoch": 202.72566371681415,
      "grad_norm": 5.82323707476462e-07,
      "learning_rate": 0.24236366166090004,
      "loss": 0.0,
      "num_input_tokens_seen": 6566536,
      "step": 11555
    },
    {
      "epoch": 202.81415929203538,
      "grad_norm": 1.086180887455157e-07,
      "learning_rate": 0.24231724130638527,
      "loss": 0.0,
      "num_input_tokens_seen": 6569368,
      "step": 11560
    },
    {
      "epoch": 202.90265486725664,
      "grad_norm": 4.842882503908186e-07,
      "learning_rate": 0.2422708067153917,
      "loss": 0.0,
      "num_input_tokens_seen": 6572024,
      "step": 11565
    },
    {
      "epoch": 202.99115044247787,
      "grad_norm": 3.9859395428720745e-07,
      "learning_rate": 0.24222435789508026,
      "loss": 0.0,
      "num_input_tokens_seen": 6574904,
      "step": 11570
    },
    {
      "epoch": 203.07079646017698,
      "grad_norm": 2.642480581016571e-07,
      "learning_rate": 0.24217789485261387,
      "loss": 0.0,
      "num_input_tokens_seen": 6577192,
      "step": 11575
    },
    {
      "epoch": 203.15929203539824,
      "grad_norm": 1.8290812420218572e-07,
      "learning_rate": 0.2421314175951577,
      "loss": 0.0,
      "num_input_tokens_seen": 6580680,
      "step": 11580
    },
    {
      "epoch": 203.24778761061947,
      "grad_norm": 8.232578352362907e-08,
      "learning_rate": 0.2420849261298791,
      "loss": 0.0,
      "num_input_tokens_seen": 6583640,
      "step": 11585
    },
    {
      "epoch": 203.3362831858407,
      "grad_norm": 3.479897259239806e-07,
      "learning_rate": 0.24203842046394775,
      "loss": 0.0,
      "num_input_tokens_seen": 6586648,
      "step": 11590
    },
    {
      "epoch": 203.42477876106196,
      "grad_norm": 2.8679508545792487e-07,
      "learning_rate": 0.24199190060453535,
      "loss": 0.0,
      "num_input_tokens_seen": 6589480,
      "step": 11595
    },
    {
      "epoch": 203.5132743362832,
      "grad_norm": 4.1507590253786475e-07,
      "learning_rate": 0.2419453665588158,
      "loss": 0.0,
      "num_input_tokens_seen": 6592280,
      "step": 11600
    },
    {
      "epoch": 203.5132743362832,
      "eval_loss": 0.487193763256073,
      "eval_runtime": 0.933,
      "eval_samples_per_second": 26.795,
      "eval_steps_per_second": 13.933,
      "num_input_tokens_seen": 6592280,
      "step": 11600
    },
    {
      "epoch": 203.60176991150442,
      "grad_norm": 3.4720304142865643e-07,
      "learning_rate": 0.24189881833396523,
      "loss": 0.0,
      "num_input_tokens_seen": 6594760,
      "step": 11605
    },
    {
      "epoch": 203.69026548672565,
      "grad_norm": 4.1194010691469884e-07,
      "learning_rate": 0.24185225593716203,
      "loss": 0.0,
      "num_input_tokens_seen": 6598040,
      "step": 11610
    },
    {
      "epoch": 203.7787610619469,
      "grad_norm": 9.797552280588206e-08,
      "learning_rate": 0.2418056793755867,
      "loss": 0.0,
      "num_input_tokens_seen": 6600792,
      "step": 11615
    },
    {
      "epoch": 203.86725663716814,
      "grad_norm": 2.943382355624635e-07,
      "learning_rate": 0.24175908865642187,
      "loss": 0.0,
      "num_input_tokens_seen": 6604136,
      "step": 11620
    },
    {
      "epoch": 203.95575221238937,
      "grad_norm": 3.446189396072441e-07,
      "learning_rate": 0.24171248378685248,
      "loss": 0.0,
      "num_input_tokens_seen": 6606392,
      "step": 11625
    },
    {
      "epoch": 204.0353982300885,
      "grad_norm": 6.046129215064866e-07,
      "learning_rate": 0.24166586477406554,
      "loss": 0.0,
      "num_input_tokens_seen": 6608488,
      "step": 11630
    },
    {
      "epoch": 204.12389380530973,
      "grad_norm": 1.3305665902407782e-07,
      "learning_rate": 0.24161923162525034,
      "loss": 0.0,
      "num_input_tokens_seen": 6611128,
      "step": 11635
    },
    {
      "epoch": 204.21238938053096,
      "grad_norm": 1.0226756330666831e-07,
      "learning_rate": 0.2415725843475982,
      "loss": 0.0,
      "num_input_tokens_seen": 6613832,
      "step": 11640
    },
    {
      "epoch": 204.30088495575222,
      "grad_norm": 6.238636842681444e-07,
      "learning_rate": 0.24152592294830286,
      "loss": 0.0,
      "num_input_tokens_seen": 6617144,
      "step": 11645
    },
    {
      "epoch": 204.38938053097345,
      "grad_norm": 5.202887791710964e-07,
      "learning_rate": 0.24147924743455995,
      "loss": 0.0,
      "num_input_tokens_seen": 6619848,
      "step": 11650
    },
    {
      "epoch": 204.47787610619469,
      "grad_norm": 2.871303195206565e-07,
      "learning_rate": 0.24143255781356754,
      "loss": 0.0,
      "num_input_tokens_seen": 6622664,
      "step": 11655
    },
    {
      "epoch": 204.56637168141592,
      "grad_norm": 4.174555101599253e-07,
      "learning_rate": 0.24138585409252566,
      "loss": 0.0,
      "num_input_tokens_seen": 6625672,
      "step": 11660
    },
    {
      "epoch": 204.65486725663717,
      "grad_norm": 1.121352752875282e-07,
      "learning_rate": 0.24133913627863662,
      "loss": 0.0,
      "num_input_tokens_seen": 6628312,
      "step": 11665
    },
    {
      "epoch": 204.7433628318584,
      "grad_norm": 3.0508829240716295e-07,
      "learning_rate": 0.241292404379105,
      "loss": 0.0,
      "num_input_tokens_seen": 6631224,
      "step": 11670
    },
    {
      "epoch": 204.83185840707964,
      "grad_norm": 7.1947698643271e-07,
      "learning_rate": 0.24124565840113735,
      "loss": 0.0,
      "num_input_tokens_seen": 6634568,
      "step": 11675
    },
    {
      "epoch": 204.9203539823009,
      "grad_norm": 1.4557302563389385e-07,
      "learning_rate": 0.2411988983519425,
      "loss": 0.0,
      "num_input_tokens_seen": 6637528,
      "step": 11680
    },
    {
      "epoch": 205.0,
      "grad_norm": 4.814737053493445e-07,
      "learning_rate": 0.24115212423873145,
      "loss": 0.0,
      "num_input_tokens_seen": 6639624,
      "step": 11685
    },
    {
      "epoch": 205.08849557522123,
      "grad_norm": 2.769475315744785e-07,
      "learning_rate": 0.24110533606871737,
      "loss": 0.0,
      "num_input_tokens_seen": 6642280,
      "step": 11690
    },
    {
      "epoch": 205.1769911504425,
      "grad_norm": 4.091311041065637e-07,
      "learning_rate": 0.24105853384911552,
      "loss": 0.0,
      "num_input_tokens_seen": 6645784,
      "step": 11695
    },
    {
      "epoch": 205.26548672566372,
      "grad_norm": 3.298055162304081e-07,
      "learning_rate": 0.24101171758714346,
      "loss": 0.0,
      "num_input_tokens_seen": 6648008,
      "step": 11700
    },
    {
      "epoch": 205.35398230088495,
      "grad_norm": 2.610799185731594e-07,
      "learning_rate": 0.24096488729002086,
      "loss": 0.0,
      "num_input_tokens_seen": 6650904,
      "step": 11705
    },
    {
      "epoch": 205.44247787610618,
      "grad_norm": 5.6814862148257816e-08,
      "learning_rate": 0.24091804296496946,
      "loss": 0.0,
      "num_input_tokens_seen": 6653848,
      "step": 11710
    },
    {
      "epoch": 205.53097345132744,
      "grad_norm": 3.9590460687577433e-07,
      "learning_rate": 0.2408711846192133,
      "loss": 0.0,
      "num_input_tokens_seen": 6656632,
      "step": 11715
    },
    {
      "epoch": 205.61946902654867,
      "grad_norm": 3.9307300880864204e-07,
      "learning_rate": 0.24082431225997855,
      "loss": 0.0,
      "num_input_tokens_seen": 6659528,
      "step": 11720
    },
    {
      "epoch": 205.7079646017699,
      "grad_norm": 1.100673472365088e-07,
      "learning_rate": 0.24077742589449344,
      "loss": 0.0,
      "num_input_tokens_seen": 6663000,
      "step": 11725
    },
    {
      "epoch": 205.79646017699116,
      "grad_norm": 2.5154989202746947e-07,
      "learning_rate": 0.24073052552998844,
      "loss": 0.0,
      "num_input_tokens_seen": 6665848,
      "step": 11730
    },
    {
      "epoch": 205.8849557522124,
      "grad_norm": 4.735847767278756e-07,
      "learning_rate": 0.2406836111736963,
      "loss": 0.0,
      "num_input_tokens_seen": 6668472,
      "step": 11735
    },
    {
      "epoch": 205.97345132743362,
      "grad_norm": 2.0581003923325625e-07,
      "learning_rate": 0.2406366828328517,
      "loss": 0.0,
      "num_input_tokens_seen": 6671320,
      "step": 11740
    },
    {
      "epoch": 206.05309734513276,
      "grad_norm": 1.9081984703461785e-07,
      "learning_rate": 0.2405897405146915,
      "loss": 0.0,
      "num_input_tokens_seen": 6673536,
      "step": 11745
    },
    {
      "epoch": 206.141592920354,
      "grad_norm": 4.60091484910663e-07,
      "learning_rate": 0.240542784226455,
      "loss": 0.0,
      "num_input_tokens_seen": 6676256,
      "step": 11750
    },
    {
      "epoch": 206.23008849557522,
      "grad_norm": 4.968562734575244e-07,
      "learning_rate": 0.24049581397538328,
      "loss": 0.0,
      "num_input_tokens_seen": 6679232,
      "step": 11755
    },
    {
      "epoch": 206.31858407079645,
      "grad_norm": 2.0141001755291654e-07,
      "learning_rate": 0.24044882976871984,
      "loss": 0.0,
      "num_input_tokens_seen": 6682224,
      "step": 11760
    },
    {
      "epoch": 206.4070796460177,
      "grad_norm": 3.458565700498184e-08,
      "learning_rate": 0.2404018316137102,
      "loss": 0.0,
      "num_input_tokens_seen": 6684880,
      "step": 11765
    },
    {
      "epoch": 206.49557522123894,
      "grad_norm": 1.6193952490084484e-07,
      "learning_rate": 0.24035481951760204,
      "loss": 0.0,
      "num_input_tokens_seen": 6688224,
      "step": 11770
    },
    {
      "epoch": 206.58407079646017,
      "grad_norm": 1.4019437344359176e-07,
      "learning_rate": 0.2403077934876452,
      "loss": 0.0,
      "num_input_tokens_seen": 6691040,
      "step": 11775
    },
    {
      "epoch": 206.67256637168143,
      "grad_norm": 5.110609890834894e-08,
      "learning_rate": 0.2402607535310918,
      "loss": 0.0,
      "num_input_tokens_seen": 6694016,
      "step": 11780
    },
    {
      "epoch": 206.76106194690266,
      "grad_norm": 1.9071717360930052e-07,
      "learning_rate": 0.2402136996551959,
      "loss": 0.0,
      "num_input_tokens_seen": 6696736,
      "step": 11785
    },
    {
      "epoch": 206.8495575221239,
      "grad_norm": 2.2265949439770338e-07,
      "learning_rate": 0.24016663186721376,
      "loss": 0.0,
      "num_input_tokens_seen": 6700016,
      "step": 11790
    },
    {
      "epoch": 206.93805309734512,
      "grad_norm": 5.683350536855869e-07,
      "learning_rate": 0.24011955017440395,
      "loss": 0.0,
      "num_input_tokens_seen": 6702592,
      "step": 11795
    },
    {
      "epoch": 207.01769911504425,
      "grad_norm": 3.197432079105056e-07,
      "learning_rate": 0.24007245458402696,
      "loss": 0.0,
      "num_input_tokens_seen": 6704968,
      "step": 11800
    },
    {
      "epoch": 207.01769911504425,
      "eval_loss": 0.4892892837524414,
      "eval_runtime": 0.9367,
      "eval_samples_per_second": 26.69,
      "eval_steps_per_second": 13.879,
      "num_input_tokens_seen": 6704968,
      "step": 11800
    },
    {
      "epoch": 207.10619469026548,
      "grad_norm": 1.49448098341054e-07,
      "learning_rate": 0.2400253451033456,
      "loss": 0.0,
      "num_input_tokens_seen": 6708008,
      "step": 11805
    },
    {
      "epoch": 207.1946902654867,
      "grad_norm": 2.9182066896282777e-07,
      "learning_rate": 0.23997822173962463,
      "loss": 0.0,
      "num_input_tokens_seen": 6711368,
      "step": 11810
    },
    {
      "epoch": 207.28318584070797,
      "grad_norm": 2.892212762617419e-07,
      "learning_rate": 0.23993108450013118,
      "loss": 0.0,
      "num_input_tokens_seen": 6714280,
      "step": 11815
    },
    {
      "epoch": 207.3716814159292,
      "grad_norm": 4.6587470592385216e-07,
      "learning_rate": 0.2398839333921343,
      "loss": 0.0,
      "num_input_tokens_seen": 6716840,
      "step": 11820
    },
    {
      "epoch": 207.46017699115043,
      "grad_norm": 6.0321710293465e-08,
      "learning_rate": 0.23983676842290536,
      "loss": 0.0,
      "num_input_tokens_seen": 6719288,
      "step": 11825
    },
    {
      "epoch": 207.5486725663717,
      "grad_norm": 1.510777565272292e-07,
      "learning_rate": 0.2397895895997178,
      "loss": 0.0,
      "num_input_tokens_seen": 6722488,
      "step": 11830
    },
    {
      "epoch": 207.63716814159292,
      "grad_norm": 3.27690344192888e-07,
      "learning_rate": 0.23974239692984714,
      "loss": 0.0,
      "num_input_tokens_seen": 6725464,
      "step": 11835
    },
    {
      "epoch": 207.72566371681415,
      "grad_norm": 2.2574933211672032e-07,
      "learning_rate": 0.2396951904205711,
      "loss": 0.0,
      "num_input_tokens_seen": 6728376,
      "step": 11840
    },
    {
      "epoch": 207.81415929203538,
      "grad_norm": 1.3871260762243764e-07,
      "learning_rate": 0.23964797007916952,
      "loss": 0.0,
      "num_input_tokens_seen": 6730856,
      "step": 11845
    },
    {
      "epoch": 207.90265486725664,
      "grad_norm": 1.014870392168632e-07,
      "learning_rate": 0.23960073591292436,
      "loss": 0.0,
      "num_input_tokens_seen": 6733608,
      "step": 11850
    },
    {
      "epoch": 207.99115044247787,
      "grad_norm": 2.5451330998293997e-07,
      "learning_rate": 0.2395534879291197,
      "loss": 0.0,
      "num_input_tokens_seen": 6736904,
      "step": 11855
    },
    {
      "epoch": 208.07079646017698,
      "grad_norm": 4.804853688256117e-07,
      "learning_rate": 0.23950622613504186,
      "loss": 0.0,
      "num_input_tokens_seen": 6739352,
      "step": 11860
    },
    {
      "epoch": 208.15929203539824,
      "grad_norm": 1.5932776875615673e-07,
      "learning_rate": 0.2394589505379791,
      "loss": 0.0,
      "num_input_tokens_seen": 6742152,
      "step": 11865
    },
    {
      "epoch": 208.24778761061947,
      "grad_norm": 7.910145427558746e-07,
      "learning_rate": 0.23941166114522197,
      "loss": 0.0,
      "num_input_tokens_seen": 6745224,
      "step": 11870
    },
    {
      "epoch": 208.3362831858407,
      "grad_norm": 3.9410454633070913e-07,
      "learning_rate": 0.23936435796406308,
      "loss": 0.0,
      "num_input_tokens_seen": 6748200,
      "step": 11875
    },
    {
      "epoch": 208.42477876106196,
      "grad_norm": 2.309620299456583e-07,
      "learning_rate": 0.23931704100179715,
      "loss": 0.0,
      "num_input_tokens_seen": 6750888,
      "step": 11880
    },
    {
      "epoch": 208.5132743362832,
      "grad_norm": 2.068713769176611e-07,
      "learning_rate": 0.2392697102657211,
      "loss": 0.0,
      "num_input_tokens_seen": 6753848,
      "step": 11885
    },
    {
      "epoch": 208.60176991150442,
      "grad_norm": 3.2528348015148367e-07,
      "learning_rate": 0.23922236576313388,
      "loss": 0.0,
      "num_input_tokens_seen": 6756264,
      "step": 11890
    },
    {
      "epoch": 208.69026548672565,
      "grad_norm": 2.4071962911875744e-07,
      "learning_rate": 0.2391750075013366,
      "loss": 0.0,
      "num_input_tokens_seen": 6759224,
      "step": 11895
    },
    {
      "epoch": 208.7787610619469,
      "grad_norm": 2.5311922513537866e-07,
      "learning_rate": 0.2391276354876326,
      "loss": 0.0,
      "num_input_tokens_seen": 6761736,
      "step": 11900
    },
    {
      "epoch": 208.86725663716814,
      "grad_norm": 4.633786545582552e-07,
      "learning_rate": 0.23908024972932707,
      "loss": 0.0,
      "num_input_tokens_seen": 6764824,
      "step": 11905
    },
    {
      "epoch": 208.95575221238937,
      "grad_norm": 1.3194527070936601e-07,
      "learning_rate": 0.2390328502337276,
      "loss": 0.0,
      "num_input_tokens_seen": 6768040,
      "step": 11910
    },
    {
      "epoch": 209.0353982300885,
      "grad_norm": 1.6709616090793133e-07,
      "learning_rate": 0.23898543700814376,
      "loss": 0.0,
      "num_input_tokens_seen": 6770584,
      "step": 11915
    },
    {
      "epoch": 209.12389380530973,
      "grad_norm": 6.593207046989846e-08,
      "learning_rate": 0.2389380100598873,
      "loss": 0.0,
      "num_input_tokens_seen": 6773256,
      "step": 11920
    },
    {
      "epoch": 209.21238938053096,
      "grad_norm": 2.7999178087156906e-07,
      "learning_rate": 0.23889056939627207,
      "loss": 0.0,
      "num_input_tokens_seen": 6775960,
      "step": 11925
    },
    {
      "epoch": 209.30088495575222,
      "grad_norm": 2.3850557795412897e-07,
      "learning_rate": 0.23884311502461386,
      "loss": 0.0,
      "num_input_tokens_seen": 6778840,
      "step": 11930
    },
    {
      "epoch": 209.38938053097345,
      "grad_norm": 8.749361057880378e-08,
      "learning_rate": 0.23879564695223088,
      "loss": 0.0,
      "num_input_tokens_seen": 6781656,
      "step": 11935
    },
    {
      "epoch": 209.47787610619469,
      "grad_norm": 6.715486478015009e-08,
      "learning_rate": 0.23874816518644332,
      "loss": 0.0,
      "num_input_tokens_seen": 6784824,
      "step": 11940
    },
    {
      "epoch": 209.56637168141592,
      "grad_norm": 2.8082251901651034e-07,
      "learning_rate": 0.23870066973457335,
      "loss": 0.0,
      "num_input_tokens_seen": 6787592,
      "step": 11945
    },
    {
      "epoch": 209.65486725663717,
      "grad_norm": 4.257181274169852e-07,
      "learning_rate": 0.23865316060394545,
      "loss": 0.0,
      "num_input_tokens_seen": 6790600,
      "step": 11950
    },
    {
      "epoch": 209.7433628318584,
      "grad_norm": 2.644700884957274e-07,
      "learning_rate": 0.2386056378018861,
      "loss": 0.0,
      "num_input_tokens_seen": 6793096,
      "step": 11955
    },
    {
      "epoch": 209.83185840707964,
      "grad_norm": 3.445805418778036e-07,
      "learning_rate": 0.2385581013357239,
      "loss": 0.0,
      "num_input_tokens_seen": 6796776,
      "step": 11960
    },
    {
      "epoch": 209.9203539823009,
      "grad_norm": 5.839167442900361e-07,
      "learning_rate": 0.23851055121278958,
      "loss": 0.0,
      "num_input_tokens_seen": 6799464,
      "step": 11965
    },
    {
      "epoch": 210.0,
      "grad_norm": 4.547521825770673e-07,
      "learning_rate": 0.23846298744041594,
      "loss": 0.0,
      "num_input_tokens_seen": 6802096,
      "step": 11970
    },
    {
      "epoch": 210.08849557522123,
      "grad_norm": 4.2158333712905005e-07,
      "learning_rate": 0.23841541002593802,
      "loss": 0.0,
      "num_input_tokens_seen": 6804736,
      "step": 11975
    },
    {
      "epoch": 210.1769911504425,
      "grad_norm": 2.382707009473961e-07,
      "learning_rate": 0.23836781897669276,
      "loss": 0.0,
      "num_input_tokens_seen": 6807408,
      "step": 11980
    },
    {
      "epoch": 210.26548672566372,
      "grad_norm": 2.783126262784208e-07,
      "learning_rate": 0.23832021430001926,
      "loss": 0.0,
      "num_input_tokens_seen": 6810368,
      "step": 11985
    },
    {
      "epoch": 210.35398230088495,
      "grad_norm": 1.1486446283015539e-07,
      "learning_rate": 0.2382725960032588,
      "loss": 0.0,
      "num_input_tokens_seen": 6813776,
      "step": 11990
    },
    {
      "epoch": 210.44247787610618,
      "grad_norm": 3.9150376096586115e-07,
      "learning_rate": 0.23822496409375482,
      "loss": 0.0,
      "num_input_tokens_seen": 6816432,
      "step": 11995
    },
    {
      "epoch": 210.53097345132744,
      "grad_norm": 3.304718632080039e-07,
      "learning_rate": 0.2381773185788526,
      "loss": 0.0,
      "num_input_tokens_seen": 6819568,
      "step": 12000
    },
    {
      "epoch": 210.53097345132744,
      "eval_loss": 0.49621883034706116,
      "eval_runtime": 0.9242,
      "eval_samples_per_second": 27.05,
      "eval_steps_per_second": 14.066,
      "num_input_tokens_seen": 6819568,
      "step": 12000
    },
    {
      "epoch": 210.61946902654867,
      "grad_norm": 2.0781872933639534e-07,
      "learning_rate": 0.2381296594658998,
      "loss": 0.0,
      "num_input_tokens_seen": 6822048,
      "step": 12005
    },
    {
      "epoch": 210.7079646017699,
      "grad_norm": 5.349688763089944e-07,
      "learning_rate": 0.238081986762246,
      "loss": 0.0,
      "num_input_tokens_seen": 6824960,
      "step": 12010
    },
    {
      "epoch": 210.79646017699116,
      "grad_norm": 2.2202142702099081e-07,
      "learning_rate": 0.23803430047524293,
      "loss": 0.0,
      "num_input_tokens_seen": 6827856,
      "step": 12015
    },
    {
      "epoch": 210.8849557522124,
      "grad_norm": 4.923164809156333e-08,
      "learning_rate": 0.23798660061224441,
      "loss": 0.0,
      "num_input_tokens_seen": 6831024,
      "step": 12020
    },
    {
      "epoch": 210.97345132743362,
      "grad_norm": 1.6533751079350623e-07,
      "learning_rate": 0.23793888718060632,
      "loss": 0.0,
      "num_input_tokens_seen": 6833952,
      "step": 12025
    },
    {
      "epoch": 211.05309734513276,
      "grad_norm": 2.785128856430674e-07,
      "learning_rate": 0.23789116018768675,
      "loss": 0.0,
      "num_input_tokens_seen": 6836016,
      "step": 12030
    },
    {
      "epoch": 211.141592920354,
      "grad_norm": 1.1303449554134204e-07,
      "learning_rate": 0.2378434196408458,
      "loss": 0.0,
      "num_input_tokens_seen": 6838512,
      "step": 12035
    },
    {
      "epoch": 211.23008849557522,
      "grad_norm": 1.3228979867108137e-07,
      "learning_rate": 0.23779566554744563,
      "loss": 0.0,
      "num_input_tokens_seen": 6841424,
      "step": 12040
    },
    {
      "epoch": 211.31858407079645,
      "grad_norm": 3.3929444498426165e-07,
      "learning_rate": 0.23774789791485051,
      "loss": 0.0,
      "num_input_tokens_seen": 6844480,
      "step": 12045
    },
    {
      "epoch": 211.4070796460177,
      "grad_norm": 1.4902263956173556e-07,
      "learning_rate": 0.2377001167504268,
      "loss": 0.0,
      "num_input_tokens_seen": 6847456,
      "step": 12050
    },
    {
      "epoch": 211.49557522123894,
      "grad_norm": 1.856122935350868e-07,
      "learning_rate": 0.23765232206154302,
      "loss": 0.0,
      "num_input_tokens_seen": 6850112,
      "step": 12055
    },
    {
      "epoch": 211.58407079646017,
      "grad_norm": 1.5214016002573771e-07,
      "learning_rate": 0.23760451385556966,
      "loss": 0.0,
      "num_input_tokens_seen": 6853136,
      "step": 12060
    },
    {
      "epoch": 211.67256637168143,
      "grad_norm": 9.710696957654363e-08,
      "learning_rate": 0.23755669213987932,
      "loss": 0.0,
      "num_input_tokens_seen": 6856240,
      "step": 12065
    },
    {
      "epoch": 211.76106194690266,
      "grad_norm": 5.941586778135388e-08,
      "learning_rate": 0.23750885692184676,
      "loss": 0.0,
      "num_input_tokens_seen": 6859168,
      "step": 12070
    },
    {
      "epoch": 211.8495575221239,
      "grad_norm": 1.0096700009398774e-07,
      "learning_rate": 0.23746100820884875,
      "loss": 0.0,
      "num_input_tokens_seen": 6862240,
      "step": 12075
    },
    {
      "epoch": 211.93805309734512,
      "grad_norm": 3.790553364524385e-07,
      "learning_rate": 0.23741314600826421,
      "loss": 0.0,
      "num_input_tokens_seen": 6865344,
      "step": 12080
    },
    {
      "epoch": 212.01769911504425,
      "grad_norm": 1.1807700417421074e-07,
      "learning_rate": 0.23736527032747406,
      "loss": 0.0,
      "num_input_tokens_seen": 6868176,
      "step": 12085
    },
    {
      "epoch": 212.10619469026548,
      "grad_norm": 6.57779537505121e-08,
      "learning_rate": 0.23731738117386128,
      "loss": 0.0,
      "num_input_tokens_seen": 6871296,
      "step": 12090
    },
    {
      "epoch": 212.1946902654867,
      "grad_norm": 2.130368557118345e-07,
      "learning_rate": 0.237269478554811,
      "loss": 0.0,
      "num_input_tokens_seen": 6873888,
      "step": 12095
    },
    {
      "epoch": 212.28318584070797,
      "grad_norm": 7.235352228462943e-08,
      "learning_rate": 0.23722156247771053,
      "loss": 0.0,
      "num_input_tokens_seen": 6876848,
      "step": 12100
    },
    {
      "epoch": 212.3716814159292,
      "grad_norm": 3.394007137558219e-07,
      "learning_rate": 0.23717363294994895,
      "loss": 0.0,
      "num_input_tokens_seen": 6880000,
      "step": 12105
    },
    {
      "epoch": 212.46017699115043,
      "grad_norm": 3.5401970421844453e-07,
      "learning_rate": 0.2371256899789177,
      "loss": 0.0,
      "num_input_tokens_seen": 6882928,
      "step": 12110
    },
    {
      "epoch": 212.5486725663717,
      "grad_norm": 1.1584475601011945e-07,
      "learning_rate": 0.23707773357201017,
      "loss": 0.0,
      "num_input_tokens_seen": 6886208,
      "step": 12115
    },
    {
      "epoch": 212.63716814159292,
      "grad_norm": 3.043722074380639e-07,
      "learning_rate": 0.2370297637366218,
      "loss": 0.0,
      "num_input_tokens_seen": 6888736,
      "step": 12120
    },
    {
      "epoch": 212.72566371681415,
      "grad_norm": 3.242463435526588e-07,
      "learning_rate": 0.23698178048015026,
      "loss": 0.0,
      "num_input_tokens_seen": 6891696,
      "step": 12125
    },
    {
      "epoch": 212.81415929203538,
      "grad_norm": 4.4308862356956524e-07,
      "learning_rate": 0.236933783809995,
      "loss": 0.0,
      "num_input_tokens_seen": 6894224,
      "step": 12130
    },
    {
      "epoch": 212.90265486725664,
      "grad_norm": 3.022300063548755e-07,
      "learning_rate": 0.23688577373355785,
      "loss": 0.0,
      "num_input_tokens_seen": 6896720,
      "step": 12135
    },
    {
      "epoch": 212.99115044247787,
      "grad_norm": 3.8581887906730117e-07,
      "learning_rate": 0.23683775025824247,
      "loss": 0.0,
      "num_input_tokens_seen": 6899408,
      "step": 12140
    },
    {
      "epoch": 213.07079646017698,
      "grad_norm": 1.3239019835964427e-07,
      "learning_rate": 0.2367897133914548,
      "loss": 0.0,
      "num_input_tokens_seen": 6902216,
      "step": 12145
    },
    {
      "epoch": 213.15929203539824,
      "grad_norm": 2.5055794594663894e-07,
      "learning_rate": 0.2367416631406026,
      "loss": 0.0,
      "num_input_tokens_seen": 6905544,
      "step": 12150
    },
    {
      "epoch": 213.24778761061947,
      "grad_norm": 3.6806500247621443e-07,
      "learning_rate": 0.23669359951309588,
      "loss": 0.0,
      "num_input_tokens_seen": 6907976,
      "step": 12155
    },
    {
      "epoch": 213.3362831858407,
      "grad_norm": 3.078547763379902e-07,
      "learning_rate": 0.23664552251634666,
      "loss": 0.0,
      "num_input_tokens_seen": 6911176,
      "step": 12160
    },
    {
      "epoch": 213.42477876106196,
      "grad_norm": 1.072717807915069e-07,
      "learning_rate": 0.23659743215776907,
      "loss": 0.0,
      "num_input_tokens_seen": 6913848,
      "step": 12165
    },
    {
      "epoch": 213.5132743362832,
      "grad_norm": 1.450858491125473e-07,
      "learning_rate": 0.23654932844477908,
      "loss": 0.0,
      "num_input_tokens_seen": 6916952,
      "step": 12170
    },
    {
      "epoch": 213.60176991150442,
      "grad_norm": 4.917617957289622e-07,
      "learning_rate": 0.23650121138479507,
      "loss": 0.0,
      "num_input_tokens_seen": 6919560,
      "step": 12175
    },
    {
      "epoch": 213.69026548672565,
      "grad_norm": 3.639090380147536e-07,
      "learning_rate": 0.23645308098523724,
      "loss": 0.0,
      "num_input_tokens_seen": 6922472,
      "step": 12180
    },
    {
      "epoch": 213.7787610619469,
      "grad_norm": 3.5976827916783805e-07,
      "learning_rate": 0.23640493725352785,
      "loss": 0.0,
      "num_input_tokens_seen": 6925528,
      "step": 12185
    },
    {
      "epoch": 213.86725663716814,
      "grad_norm": 1.562461733328746e-07,
      "learning_rate": 0.2363567801970913,
      "loss": 0.0,
      "num_input_tokens_seen": 6928072,
      "step": 12190
    },
    {
      "epoch": 213.95575221238937,
      "grad_norm": 2.707745068164513e-07,
      "learning_rate": 0.236308609823354,
      "loss": 0.0,
      "num_input_tokens_seen": 6930616,
      "step": 12195
    },
    {
      "epoch": 214.0353982300885,
      "grad_norm": 5.0929379824538046e-08,
      "learning_rate": 0.23626042613974452,
      "loss": 0.0,
      "num_input_tokens_seen": 6933264,
      "step": 12200
    },
    {
      "epoch": 214.0353982300885,
      "eval_loss": 0.4928906261920929,
      "eval_runtime": 0.9317,
      "eval_samples_per_second": 26.831,
      "eval_steps_per_second": 13.952,
      "num_input_tokens_seen": 6933264,
      "step": 12200
    },
    {
      "epoch": 214.12389380530973,
      "grad_norm": 4.867177949563484e-07,
      "learning_rate": 0.23621222915369325,
      "loss": 0.0,
      "num_input_tokens_seen": 6935952,
      "step": 12205
    },
    {
      "epoch": 214.21238938053096,
      "grad_norm": 2.3095871881650964e-07,
      "learning_rate": 0.23616401887263283,
      "loss": 0.0,
      "num_input_tokens_seen": 6938960,
      "step": 12210
    },
    {
      "epoch": 214.30088495575222,
      "grad_norm": 2.0928104049744434e-07,
      "learning_rate": 0.23611579530399793,
      "loss": 0.0,
      "num_input_tokens_seen": 6941936,
      "step": 12215
    },
    {
      "epoch": 214.38938053097345,
      "grad_norm": 3.0740426382180885e-07,
      "learning_rate": 0.23606755845522517,
      "loss": 0.0,
      "num_input_tokens_seen": 6944496,
      "step": 12220
    },
    {
      "epoch": 214.47787610619469,
      "grad_norm": 1.9969229470007122e-07,
      "learning_rate": 0.23601930833375329,
      "loss": 0.0,
      "num_input_tokens_seen": 6947264,
      "step": 12225
    },
    {
      "epoch": 214.56637168141592,
      "grad_norm": 1.302764616184504e-07,
      "learning_rate": 0.23597104494702312,
      "loss": 0.0,
      "num_input_tokens_seen": 6949744,
      "step": 12230
    },
    {
      "epoch": 214.65486725663717,
      "grad_norm": 1.2968605744845263e-07,
      "learning_rate": 0.23592276830247744,
      "loss": 0.0,
      "num_input_tokens_seen": 6952432,
      "step": 12235
    },
    {
      "epoch": 214.7433628318584,
      "grad_norm": 5.0475485124934494e-08,
      "learning_rate": 0.2358744784075611,
      "loss": 0.0,
      "num_input_tokens_seen": 6955008,
      "step": 12240
    },
    {
      "epoch": 214.83185840707964,
      "grad_norm": 4.78043091334257e-07,
      "learning_rate": 0.235826175269721,
      "loss": 0.0,
      "num_input_tokens_seen": 6957856,
      "step": 12245
    },
    {
      "epoch": 214.9203539823009,
      "grad_norm": 1.6798730939626694e-07,
      "learning_rate": 0.23577785889640612,
      "loss": 0.0,
      "num_input_tokens_seen": 6960784,
      "step": 12250
    },
    {
      "epoch": 215.0,
      "grad_norm": 1.3436637402719498e-07,
      "learning_rate": 0.23572952929506744,
      "loss": 0.0,
      "num_input_tokens_seen": 6963416,
      "step": 12255
    },
    {
      "epoch": 215.08849557522123,
      "grad_norm": 1.5469250058686157e-07,
      "learning_rate": 0.23568118647315803,
      "loss": 0.0,
      "num_input_tokens_seen": 6966216,
      "step": 12260
    },
    {
      "epoch": 215.1769911504425,
      "grad_norm": 1.9714543952886743e-07,
      "learning_rate": 0.23563283043813296,
      "loss": 0.0,
      "num_input_tokens_seen": 6968904,
      "step": 12265
    },
    {
      "epoch": 215.26548672566372,
      "grad_norm": 3.8524711953868973e-07,
      "learning_rate": 0.23558446119744922,
      "loss": 0.0,
      "num_input_tokens_seen": 6971272,
      "step": 12270
    },
    {
      "epoch": 215.35398230088495,
      "grad_norm": 1.973434962110332e-07,
      "learning_rate": 0.23553607875856608,
      "loss": 0.0,
      "num_input_tokens_seen": 6973976,
      "step": 12275
    },
    {
      "epoch": 215.44247787610618,
      "grad_norm": 2.0153437674252928e-07,
      "learning_rate": 0.2354876831289447,
      "loss": 0.0,
      "num_input_tokens_seen": 6977736,
      "step": 12280
    },
    {
      "epoch": 215.53097345132744,
      "grad_norm": 1.9678871865380643e-07,
      "learning_rate": 0.23543927431604827,
      "loss": 0.0,
      "num_input_tokens_seen": 6980936,
      "step": 12285
    },
    {
      "epoch": 215.61946902654867,
      "grad_norm": 2.2862732862449775e-07,
      "learning_rate": 0.23539085232734203,
      "loss": 0.0,
      "num_input_tokens_seen": 6984104,
      "step": 12290
    },
    {
      "epoch": 215.7079646017699,
      "grad_norm": 4.569604925563908e-07,
      "learning_rate": 0.2353424171702933,
      "loss": 0.0,
      "num_input_tokens_seen": 6986696,
      "step": 12295
    },
    {
      "epoch": 215.79646017699116,
      "grad_norm": 2.44468139953824e-07,
      "learning_rate": 0.23529396885237133,
      "loss": 0.0,
      "num_input_tokens_seen": 6989560,
      "step": 12300
    },
    {
      "epoch": 215.8849557522124,
      "grad_norm": 5.362099386729824e-07,
      "learning_rate": 0.2352455073810475,
      "loss": 0.0,
      "num_input_tokens_seen": 6992344,
      "step": 12305
    },
    {
      "epoch": 215.97345132743362,
      "grad_norm": 2.0994924909700785e-07,
      "learning_rate": 0.23519703276379517,
      "loss": 0.0,
      "num_input_tokens_seen": 6995064,
      "step": 12310
    },
    {
      "epoch": 216.05309734513276,
      "grad_norm": 5.53833388039493e-08,
      "learning_rate": 0.2351485450080897,
      "loss": 0.0,
      "num_input_tokens_seen": 6997408,
      "step": 12315
    },
    {
      "epoch": 216.141592920354,
      "grad_norm": 1.0898384772417558e-07,
      "learning_rate": 0.2351000441214086,
      "loss": 0.0,
      "num_input_tokens_seen": 7000240,
      "step": 12320
    },
    {
      "epoch": 216.23008849557522,
      "grad_norm": 4.014283092601545e-07,
      "learning_rate": 0.23505153011123125,
      "loss": 0.0,
      "num_input_tokens_seen": 7002960,
      "step": 12325
    },
    {
      "epoch": 216.31858407079645,
      "grad_norm": 2.1063546284949553e-07,
      "learning_rate": 0.23500300298503912,
      "loss": 0.0,
      "num_input_tokens_seen": 7005824,
      "step": 12330
    },
    {
      "epoch": 216.4070796460177,
      "grad_norm": 3.056090918107657e-07,
      "learning_rate": 0.23495446275031576,
      "loss": 0.0,
      "num_input_tokens_seen": 7008576,
      "step": 12335
    },
    {
      "epoch": 216.49557522123894,
      "grad_norm": 1.0643684333899728e-07,
      "learning_rate": 0.2349059094145466,
      "loss": 0.0,
      "num_input_tokens_seen": 7011344,
      "step": 12340
    },
    {
      "epoch": 216.58407079646017,
      "grad_norm": 2.2202651450697886e-07,
      "learning_rate": 0.2348573429852192,
      "loss": 0.0,
      "num_input_tokens_seen": 7014416,
      "step": 12345
    },
    {
      "epoch": 216.67256637168143,
      "grad_norm": 2.74347371487238e-07,
      "learning_rate": 0.23480876346982313,
      "loss": 0.0,
      "num_input_tokens_seen": 7017344,
      "step": 12350
    },
    {
      "epoch": 216.76106194690266,
      "grad_norm": 7.482920949541949e-08,
      "learning_rate": 0.23476017087585,
      "loss": 0.0,
      "num_input_tokens_seen": 7020400,
      "step": 12355
    },
    {
      "epoch": 216.8495575221239,
      "grad_norm": 2.5974929940275615e-07,
      "learning_rate": 0.23471156521079334,
      "loss": 0.0,
      "num_input_tokens_seen": 7023536,
      "step": 12360
    },
    {
      "epoch": 216.93805309734512,
      "grad_norm": 5.025373184253112e-07,
      "learning_rate": 0.23466294648214875,
      "loss": 0.0,
      "num_input_tokens_seen": 7026192,
      "step": 12365
    },
    {
      "epoch": 217.01769911504425,
      "grad_norm": 1.2951481664913445e-07,
      "learning_rate": 0.2346143146974139,
      "loss": 0.0,
      "num_input_tokens_seen": 7028648,
      "step": 12370
    },
    {
      "epoch": 217.10619469026548,
      "grad_norm": 8.34119617820761e-08,
      "learning_rate": 0.23456566986408836,
      "loss": 0.0,
      "num_input_tokens_seen": 7031272,
      "step": 12375
    },
    {
      "epoch": 217.1946902654867,
      "grad_norm": 1.7626679493787378e-07,
      "learning_rate": 0.23451701198967384,
      "loss": 0.0,
      "num_input_tokens_seen": 7033928,
      "step": 12380
    },
    {
      "epoch": 217.28318584070797,
      "grad_norm": 1.2963946005584148e-07,
      "learning_rate": 0.23446834108167397,
      "loss": 0.0,
      "num_input_tokens_seen": 7036984,
      "step": 12385
    },
    {
      "epoch": 217.3716814159292,
      "grad_norm": 1.960343922746688e-07,
      "learning_rate": 0.23441965714759438,
      "loss": 0.0,
      "num_input_tokens_seen": 7039480,
      "step": 12390
    },
    {
      "epoch": 217.46017699115043,
      "grad_norm": 6.779532668588217e-08,
      "learning_rate": 0.23437096019494277,
      "loss": 0.0,
      "num_input_tokens_seen": 7042536,
      "step": 12395
    },
    {
      "epoch": 217.5486725663717,
      "grad_norm": 9.813528123459037e-08,
      "learning_rate": 0.23432225023122885,
      "loss": 0.0,
      "num_input_tokens_seen": 7045688,
      "step": 12400
    },
    {
      "epoch": 217.5486725663717,
      "eval_loss": 0.49578797817230225,
      "eval_runtime": 0.9371,
      "eval_samples_per_second": 26.677,
      "eval_steps_per_second": 13.872,
      "num_input_tokens_seen": 7045688,
      "step": 12400
    },
    {
      "epoch": 217.63716814159292,
      "grad_norm": 1.702029379657688e-07,
      "learning_rate": 0.23427352726396428,
      "loss": 0.0,
      "num_input_tokens_seen": 7048344,
      "step": 12405
    },
    {
      "epoch": 217.72566371681415,
      "grad_norm": 1.6143071945862175e-07,
      "learning_rate": 0.2342247913006628,
      "loss": 0.0,
      "num_input_tokens_seen": 7051480,
      "step": 12410
    },
    {
      "epoch": 217.81415929203538,
      "grad_norm": 1.7043353750523238e-07,
      "learning_rate": 0.23417604234883999,
      "loss": 0.0,
      "num_input_tokens_seen": 7054264,
      "step": 12415
    },
    {
      "epoch": 217.90265486725664,
      "grad_norm": 1.3728536885082576e-07,
      "learning_rate": 0.23412728041601363,
      "loss": 0.0,
      "num_input_tokens_seen": 7057272,
      "step": 12420
    },
    {
      "epoch": 217.99115044247787,
      "grad_norm": 3.205842347142607e-07,
      "learning_rate": 0.23407850550970347,
      "loss": 0.0,
      "num_input_tokens_seen": 7060104,
      "step": 12425
    },
    {
      "epoch": 218.07079646017698,
      "grad_norm": 1.9919704641324643e-07,
      "learning_rate": 0.23402971763743116,
      "loss": 0.0,
      "num_input_tokens_seen": 7062728,
      "step": 12430
    },
    {
      "epoch": 218.15929203539824,
      "grad_norm": 3.390366885014373e-07,
      "learning_rate": 0.23398091680672037,
      "loss": 0.0,
      "num_input_tokens_seen": 7065416,
      "step": 12435
    },
    {
      "epoch": 218.24778761061947,
      "grad_norm": 2.4981272872537374e-07,
      "learning_rate": 0.23393210302509687,
      "loss": 0.0,
      "num_input_tokens_seen": 7068200,
      "step": 12440
    },
    {
      "epoch": 218.3362831858407,
      "grad_norm": 4.092896119800571e-07,
      "learning_rate": 0.23388327630008832,
      "loss": 0.0,
      "num_input_tokens_seen": 7070984,
      "step": 12445
    },
    {
      "epoch": 218.42477876106196,
      "grad_norm": 2.1598164323677338e-07,
      "learning_rate": 0.23383443663922443,
      "loss": 0.0,
      "num_input_tokens_seen": 7074312,
      "step": 12450
    },
    {
      "epoch": 218.5132743362832,
      "grad_norm": 2.551159923314117e-07,
      "learning_rate": 0.23378558405003685,
      "loss": 0.0,
      "num_input_tokens_seen": 7077048,
      "step": 12455
    },
    {
      "epoch": 218.60176991150442,
      "grad_norm": 3.0948223184168455e-07,
      "learning_rate": 0.2337367185400593,
      "loss": 0.0,
      "num_input_tokens_seen": 7079784,
      "step": 12460
    },
    {
      "epoch": 218.69026548672565,
      "grad_norm": 8.587687716499204e-08,
      "learning_rate": 0.23368784011682747,
      "loss": 0.0,
      "num_input_tokens_seen": 7083080,
      "step": 12465
    },
    {
      "epoch": 218.7787610619469,
      "grad_norm": 1.4631768863182515e-07,
      "learning_rate": 0.23363894878787902,
      "loss": 0.0,
      "num_input_tokens_seen": 7085848,
      "step": 12470
    },
    {
      "epoch": 218.86725663716814,
      "grad_norm": 4.377039601877186e-07,
      "learning_rate": 0.23359004456075352,
      "loss": 0.0,
      "num_input_tokens_seen": 7088760,
      "step": 12475
    },
    {
      "epoch": 218.95575221238937,
      "grad_norm": 3.220252438040916e-07,
      "learning_rate": 0.23354112744299277,
      "loss": 0.0,
      "num_input_tokens_seen": 7091816,
      "step": 12480
    },
    {
      "epoch": 219.0353982300885,
      "grad_norm": 1.4506402123970474e-07,
      "learning_rate": 0.2334921974421403,
      "loss": 0.0,
      "num_input_tokens_seen": 7094424,
      "step": 12485
    },
    {
      "epoch": 219.12389380530973,
      "grad_norm": 2.0734843531045044e-07,
      "learning_rate": 0.23344325456574178,
      "loss": 0.0,
      "num_input_tokens_seen": 7097208,
      "step": 12490
    },
    {
      "epoch": 219.21238938053096,
      "grad_norm": 3.7821507703483803e-07,
      "learning_rate": 0.23339429882134477,
      "loss": 0.0,
      "num_input_tokens_seen": 7100344,
      "step": 12495
    },
    {
      "epoch": 219.30088495575222,
      "grad_norm": 2.0057552774233045e-07,
      "learning_rate": 0.23334533021649884,
      "loss": 0.0,
      "num_input_tokens_seen": 7103320,
      "step": 12500
    },
    {
      "epoch": 219.38938053097345,
      "grad_norm": 1.8468847429176094e-07,
      "learning_rate": 0.23329634875875566,
      "loss": 0.0,
      "num_input_tokens_seen": 7106216,
      "step": 12505
    },
    {
      "epoch": 219.47787610619469,
      "grad_norm": 1.6738044905650895e-07,
      "learning_rate": 0.23324735445566874,
      "loss": 0.0,
      "num_input_tokens_seen": 7109080,
      "step": 12510
    },
    {
      "epoch": 219.56637168141592,
      "grad_norm": 1.2254810144440853e-07,
      "learning_rate": 0.2331983473147936,
      "loss": 0.0,
      "num_input_tokens_seen": 7112168,
      "step": 12515
    },
    {
      "epoch": 219.65486725663717,
      "grad_norm": 8.626222438579134e-07,
      "learning_rate": 0.23314932734368776,
      "loss": 0.0,
      "num_input_tokens_seen": 7115144,
      "step": 12520
    },
    {
      "epoch": 219.7433628318584,
      "grad_norm": 3.243315802592406e-07,
      "learning_rate": 0.2331002945499107,
      "loss": 0.0,
      "num_input_tokens_seen": 7117848,
      "step": 12525
    },
    {
      "epoch": 219.83185840707964,
      "grad_norm": 1.2940510885073309e-07,
      "learning_rate": 0.23305124894102397,
      "loss": 0.0,
      "num_input_tokens_seen": 7120664,
      "step": 12530
    },
    {
      "epoch": 219.9203539823009,
      "grad_norm": 8.2855230232326e-08,
      "learning_rate": 0.23300219052459092,
      "loss": 0.0,
      "num_input_tokens_seen": 7123208,
      "step": 12535
    },
    {
      "epoch": 220.0,
      "grad_norm": 4.7215301890446426e-08,
      "learning_rate": 0.23295311930817708,
      "loss": 0.0,
      "num_input_tokens_seen": 7125552,
      "step": 12540
    },
    {
      "epoch": 220.08849557522123,
      "grad_norm": 5.516052326015597e-08,
      "learning_rate": 0.23290403529934972,
      "loss": 0.0,
      "num_input_tokens_seen": 7128336,
      "step": 12545
    },
    {
      "epoch": 220.1769911504425,
      "grad_norm": 2.0745031292790372e-07,
      "learning_rate": 0.23285493850567832,
      "loss": 0.0,
      "num_input_tokens_seen": 7131072,
      "step": 12550
    },
    {
      "epoch": 220.26548672566372,
      "grad_norm": 2.7110854716738686e-07,
      "learning_rate": 0.23280582893473414,
      "loss": 0.0,
      "num_input_tokens_seen": 7133808,
      "step": 12555
    },
    {
      "epoch": 220.35398230088495,
      "grad_norm": 2.0625986962841125e-07,
      "learning_rate": 0.2327567065940906,
      "loss": 0.0,
      "num_input_tokens_seen": 7137520,
      "step": 12560
    },
    {
      "epoch": 220.44247787610618,
      "grad_norm": 1.8880187724334974e-07,
      "learning_rate": 0.23270757149132285,
      "loss": 0.0,
      "num_input_tokens_seen": 7140544,
      "step": 12565
    },
    {
      "epoch": 220.53097345132744,
      "grad_norm": 1.3934661069470167e-07,
      "learning_rate": 0.23265842363400827,
      "loss": 0.0,
      "num_input_tokens_seen": 7143216,
      "step": 12570
    },
    {
      "epoch": 220.61946902654867,
      "grad_norm": 3.14423658664964e-07,
      "learning_rate": 0.23260926302972595,
      "loss": 0.0,
      "num_input_tokens_seen": 7146096,
      "step": 12575
    },
    {
      "epoch": 220.7079646017699,
      "grad_norm": 9.249797017218953e-08,
      "learning_rate": 0.2325600896860572,
      "loss": 0.0,
      "num_input_tokens_seen": 7149184,
      "step": 12580
    },
    {
      "epoch": 220.79646017699116,
      "grad_norm": 6.314449763067387e-08,
      "learning_rate": 0.23251090361058505,
      "loss": 0.0,
      "num_input_tokens_seen": 7152144,
      "step": 12585
    },
    {
      "epoch": 220.8849557522124,
      "grad_norm": 8.207099000401286e-08,
      "learning_rate": 0.23246170481089476,
      "loss": 0.0,
      "num_input_tokens_seen": 7154704,
      "step": 12590
    },
    {
      "epoch": 220.97345132743362,
      "grad_norm": 2.479816032519011e-07,
      "learning_rate": 0.23241249329457317,
      "loss": 0.0,
      "num_input_tokens_seen": 7157568,
      "step": 12595
    },
    {
      "epoch": 221.05309734513276,
      "grad_norm": 2.1852331144600612e-07,
      "learning_rate": 0.23236326906920957,
      "loss": 0.0,
      "num_input_tokens_seen": 7159888,
      "step": 12600
    },
    {
      "epoch": 221.05309734513276,
      "eval_loss": 0.5160678625106812,
      "eval_runtime": 0.9363,
      "eval_samples_per_second": 26.701,
      "eval_steps_per_second": 13.885,
      "num_input_tokens_seen": 7159888,
      "step": 12600
    },
    {
      "epoch": 221.141592920354,
      "grad_norm": 8.579769428251893e-08,
      "learning_rate": 0.2323140321423948,
      "loss": 0.0,
      "num_input_tokens_seen": 7162768,
      "step": 12605
    },
    {
      "epoch": 221.23008849557522,
      "grad_norm": 1.3107484164720518e-07,
      "learning_rate": 0.23226478252172184,
      "loss": 0.0,
      "num_input_tokens_seen": 7165776,
      "step": 12610
    },
    {
      "epoch": 221.31858407079645,
      "grad_norm": 2.907549401243159e-07,
      "learning_rate": 0.23221552021478561,
      "loss": 0.0,
      "num_input_tokens_seen": 7168688,
      "step": 12615
    },
    {
      "epoch": 221.4070796460177,
      "grad_norm": 2.2111785824563412e-07,
      "learning_rate": 0.232166245229183,
      "loss": 0.0,
      "num_input_tokens_seen": 7171600,
      "step": 12620
    },
    {
      "epoch": 221.49557522123894,
      "grad_norm": 4.810115328268694e-08,
      "learning_rate": 0.2321169575725128,
      "loss": 0.0,
      "num_input_tokens_seen": 7174656,
      "step": 12625
    },
    {
      "epoch": 221.58407079646017,
      "grad_norm": 1.5246392592871416e-07,
      "learning_rate": 0.23206765725237577,
      "loss": 0.0,
      "num_input_tokens_seen": 7177648,
      "step": 12630
    },
    {
      "epoch": 221.67256637168143,
      "grad_norm": 2.4809074261611386e-07,
      "learning_rate": 0.2320183442763747,
      "loss": 0.0,
      "num_input_tokens_seen": 7180208,
      "step": 12635
    },
    {
      "epoch": 221.76106194690266,
      "grad_norm": 2.0712660386834614e-07,
      "learning_rate": 0.23196901865211422,
      "loss": 0.0,
      "num_input_tokens_seen": 7183504,
      "step": 12640
    },
    {
      "epoch": 221.8495575221239,
      "grad_norm": 2.2103148467067513e-07,
      "learning_rate": 0.231919680387201,
      "loss": 0.0,
      "num_input_tokens_seen": 7186512,
      "step": 12645
    },
    {
      "epoch": 221.93805309734512,
      "grad_norm": 3.521013809404394e-07,
      "learning_rate": 0.23187032948924358,
      "loss": 0.0,
      "num_input_tokens_seen": 7188912,
      "step": 12650
    },
    {
      "epoch": 222.01769911504425,
      "grad_norm": 3.639564170043741e-07,
      "learning_rate": 0.23182096596585247,
      "loss": 0.0,
      "num_input_tokens_seen": 7191504,
      "step": 12655
    },
    {
      "epoch": 222.10619469026548,
      "grad_norm": 3.236130226014211e-07,
      "learning_rate": 0.23177158982464025,
      "loss": 0.0,
      "num_input_tokens_seen": 7194384,
      "step": 12660
    },
    {
      "epoch": 222.1946902654867,
      "grad_norm": 3.971410365011252e-07,
      "learning_rate": 0.23172220107322122,
      "loss": 0.0,
      "num_input_tokens_seen": 7196960,
      "step": 12665
    },
    {
      "epoch": 222.28318584070797,
      "grad_norm": 1.2077116195996496e-07,
      "learning_rate": 0.23167279971921184,
      "loss": 0.0,
      "num_input_tokens_seen": 7200176,
      "step": 12670
    },
    {
      "epoch": 222.3716814159292,
      "grad_norm": 1.7236239102658146e-07,
      "learning_rate": 0.23162338577023034,
      "loss": 0.0,
      "num_input_tokens_seen": 7203392,
      "step": 12675
    },
    {
      "epoch": 222.46017699115043,
      "grad_norm": 2.6268776309734676e-07,
      "learning_rate": 0.23157395923389704,
      "loss": 0.0,
      "num_input_tokens_seen": 7206432,
      "step": 12680
    },
    {
      "epoch": 222.5486725663717,
      "grad_norm": 2.8908664262417005e-07,
      "learning_rate": 0.2315245201178341,
      "loss": 0.0,
      "num_input_tokens_seen": 7209296,
      "step": 12685
    },
    {
      "epoch": 222.63716814159292,
      "grad_norm": 2.0516149845661857e-07,
      "learning_rate": 0.23147506842966564,
      "loss": 0.0,
      "num_input_tokens_seen": 7212336,
      "step": 12690
    },
    {
      "epoch": 222.72566371681415,
      "grad_norm": 1.4075017418235802e-07,
      "learning_rate": 0.23142560417701774,
      "loss": 0.0,
      "num_input_tokens_seen": 7214960,
      "step": 12695
    },
    {
      "epoch": 222.81415929203538,
      "grad_norm": 4.5211596244598695e-08,
      "learning_rate": 0.23137612736751845,
      "loss": 0.0,
      "num_input_tokens_seen": 7217520,
      "step": 12700
    },
    {
      "epoch": 222.90265486725664,
      "grad_norm": 2.22119666659637e-07,
      "learning_rate": 0.23132663800879766,
      "loss": 0.0,
      "num_input_tokens_seen": 7220528,
      "step": 12705
    },
    {
      "epoch": 222.99115044247787,
      "grad_norm": 2.241112611045537e-07,
      "learning_rate": 0.2312771361084873,
      "loss": 0.0,
      "num_input_tokens_seen": 7223104,
      "step": 12710
    },
    {
      "epoch": 223.07079646017698,
      "grad_norm": 1.6822690440676524e-07,
      "learning_rate": 0.23122762167422112,
      "loss": 0.0,
      "num_input_tokens_seen": 7225568,
      "step": 12715
    },
    {
      "epoch": 223.15929203539824,
      "grad_norm": 3.1796182042853616e-07,
      "learning_rate": 0.23117809471363493,
      "loss": 0.0,
      "num_input_tokens_seen": 7228464,
      "step": 12720
    },
    {
      "epoch": 223.24778761061947,
      "grad_norm": 1.1719263426357429e-07,
      "learning_rate": 0.23112855523436637,
      "loss": 0.0,
      "num_input_tokens_seen": 7231024,
      "step": 12725
    },
    {
      "epoch": 223.3362831858407,
      "grad_norm": 1.4325594577258016e-07,
      "learning_rate": 0.23107900324405511,
      "loss": 0.0,
      "num_input_tokens_seen": 7233744,
      "step": 12730
    },
    {
      "epoch": 223.42477876106196,
      "grad_norm": 5.29597173226648e-07,
      "learning_rate": 0.2310294387503426,
      "loss": 0.0,
      "num_input_tokens_seen": 7237136,
      "step": 12735
    },
    {
      "epoch": 223.5132743362832,
      "grad_norm": 2.1905952962697484e-07,
      "learning_rate": 0.23097986176087237,
      "loss": 0.0,
      "num_input_tokens_seen": 7240288,
      "step": 12740
    },
    {
      "epoch": 223.60176991150442,
      "grad_norm": 5.510364786687205e-08,
      "learning_rate": 0.23093027228328986,
      "loss": 0.0,
      "num_input_tokens_seen": 7243232,
      "step": 12745
    },
    {
      "epoch": 223.69026548672565,
      "grad_norm": 2.2768264784644998e-07,
      "learning_rate": 0.23088067032524226,
      "loss": 0.0,
      "num_input_tokens_seen": 7246096,
      "step": 12750
    },
    {
      "epoch": 223.7787610619469,
      "grad_norm": 3.597275508582243e-07,
      "learning_rate": 0.23083105589437888,
      "loss": 0.0,
      "num_input_tokens_seen": 7249152,
      "step": 12755
    },
    {
      "epoch": 223.86725663716814,
      "grad_norm": 7.595461681830784e-08,
      "learning_rate": 0.23078142899835094,
      "loss": 0.0,
      "num_input_tokens_seen": 7252032,
      "step": 12760
    },
    {
      "epoch": 223.95575221238937,
      "grad_norm": 1.969575293969683e-07,
      "learning_rate": 0.23073178964481147,
      "loss": 0.0,
      "num_input_tokens_seen": 7254384,
      "step": 12765
    },
    {
      "epoch": 224.0353982300885,
      "grad_norm": 4.378934193027817e-07,
      "learning_rate": 0.2306821378414155,
      "loss": 0.0,
      "num_input_tokens_seen": 7256824,
      "step": 12770
    },
    {
      "epoch": 224.12389380530973,
      "grad_norm": 2.5744733989085944e-07,
      "learning_rate": 0.2306324735958199,
      "loss": 0.0,
      "num_input_tokens_seen": 7260104,
      "step": 12775
    },
    {
      "epoch": 224.21238938053096,
      "grad_norm": 1.6714179196242185e-07,
      "learning_rate": 0.23058279691568362,
      "loss": 0.0,
      "num_input_tokens_seen": 7263144,
      "step": 12780
    },
    {
      "epoch": 224.30088495575222,
      "grad_norm": 1.7217217873621848e-07,
      "learning_rate": 0.23053310780866745,
      "loss": 0.0,
      "num_input_tokens_seen": 7266136,
      "step": 12785
    },
    {
      "epoch": 224.38938053097345,
      "grad_norm": 8.126347950110357e-08,
      "learning_rate": 0.23048340628243397,
      "loss": 0.0,
      "num_input_tokens_seen": 7268968,
      "step": 12790
    },
    {
      "epoch": 224.47787610619469,
      "grad_norm": 3.05055408489352e-07,
      "learning_rate": 0.23043369234464783,
      "loss": 0.0,
      "num_input_tokens_seen": 7271496,
      "step": 12795
    },
    {
      "epoch": 224.56637168141592,
      "grad_norm": 8.278031060626745e-08,
      "learning_rate": 0.2303839660029755,
      "loss": 0.0,
      "num_input_tokens_seen": 7274296,
      "step": 12800
    },
    {
      "epoch": 224.56637168141592,
      "eval_loss": 0.507752001285553,
      "eval_runtime": 0.9413,
      "eval_samples_per_second": 26.558,
      "eval_steps_per_second": 13.81,
      "num_input_tokens_seen": 7274296,
      "step": 12800
    },
    {
      "epoch": 224.65486725663717,
      "grad_norm": 1.308851835801761e-07,
      "learning_rate": 0.23033422726508548,
      "loss": 0.0,
      "num_input_tokens_seen": 7277256,
      "step": 12805
    },
    {
      "epoch": 224.7433628318584,
      "grad_norm": 1.4374191437127593e-07,
      "learning_rate": 0.23028447613864808,
      "loss": 0.0,
      "num_input_tokens_seen": 7280152,
      "step": 12810
    },
    {
      "epoch": 224.83185840707964,
      "grad_norm": 1.5049036505843105e-07,
      "learning_rate": 0.2302347126313355,
      "loss": 0.0,
      "num_input_tokens_seen": 7282760,
      "step": 12815
    },
    {
      "epoch": 224.9203539823009,
      "grad_norm": 1.5567887601264374e-07,
      "learning_rate": 0.23018493675082197,
      "loss": 0.0,
      "num_input_tokens_seen": 7285448,
      "step": 12820
    },
    {
      "epoch": 225.0,
      "grad_norm": 3.368114676050027e-07,
      "learning_rate": 0.2301351485047835,
      "loss": 0.0,
      "num_input_tokens_seen": 7288280,
      "step": 12825
    },
    {
      "epoch": 225.08849557522123,
      "grad_norm": 2.876383007333061e-07,
      "learning_rate": 0.23008534790089813,
      "loss": 0.0,
      "num_input_tokens_seen": 7291208,
      "step": 12830
    },
    {
      "epoch": 225.1769911504425,
      "grad_norm": 8.366114201407981e-08,
      "learning_rate": 0.2300355349468457,
      "loss": 0.0,
      "num_input_tokens_seen": 7293928,
      "step": 12835
    },
    {
      "epoch": 225.26548672566372,
      "grad_norm": 2.457952632539673e-07,
      "learning_rate": 0.22998570965030793,
      "loss": 0.0,
      "num_input_tokens_seen": 7296728,
      "step": 12840
    },
    {
      "epoch": 225.35398230088495,
      "grad_norm": 2.9464618478414195e-07,
      "learning_rate": 0.22993587201896862,
      "loss": 0.0,
      "num_input_tokens_seen": 7299848,
      "step": 12845
    },
    {
      "epoch": 225.44247787610618,
      "grad_norm": 4.159136324233259e-07,
      "learning_rate": 0.2298860220605133,
      "loss": 0.0,
      "num_input_tokens_seen": 7303112,
      "step": 12850
    },
    {
      "epoch": 225.53097345132744,
      "grad_norm": 2.878837790376565e-07,
      "learning_rate": 0.22983615978262942,
      "loss": 0.0,
      "num_input_tokens_seen": 7305960,
      "step": 12855
    },
    {
      "epoch": 225.61946902654867,
      "grad_norm": 1.3703025558697846e-07,
      "learning_rate": 0.22978628519300648,
      "loss": 0.0,
      "num_input_tokens_seen": 7308984,
      "step": 12860
    },
    {
      "epoch": 225.7079646017699,
      "grad_norm": 2.308554769570037e-07,
      "learning_rate": 0.22973639829933568,
      "loss": 0.0,
      "num_input_tokens_seen": 7311832,
      "step": 12865
    },
    {
      "epoch": 225.79646017699116,
      "grad_norm": 1.1249062481510919e-07,
      "learning_rate": 0.22968649910931027,
      "loss": 0.0,
      "num_input_tokens_seen": 7314552,
      "step": 12870
    },
    {
      "epoch": 225.8849557522124,
      "grad_norm": 1.3017057653996744e-07,
      "learning_rate": 0.22963658763062528,
      "loss": 0.0,
      "num_input_tokens_seen": 7317736,
      "step": 12875
    },
    {
      "epoch": 225.97345132743362,
      "grad_norm": 1.5673307984798157e-07,
      "learning_rate": 0.22958666387097765,
      "loss": 0.0,
      "num_input_tokens_seen": 7320104,
      "step": 12880
    },
    {
      "epoch": 226.05309734513276,
      "grad_norm": 2.300128443266658e-07,
      "learning_rate": 0.22953672783806633,
      "loss": 0.0,
      "num_input_tokens_seen": 7322248,
      "step": 12885
    },
    {
      "epoch": 226.141592920354,
      "grad_norm": 1.1761974860746705e-07,
      "learning_rate": 0.22948677953959207,
      "loss": 0.0,
      "num_input_tokens_seen": 7325208,
      "step": 12890
    },
    {
      "epoch": 226.23008849557522,
      "grad_norm": 4.197062608568558e-08,
      "learning_rate": 0.2294368189832575,
      "loss": 0.0,
      "num_input_tokens_seen": 7327992,
      "step": 12895
    },
    {
      "epoch": 226.31858407079645,
      "grad_norm": 3.5693901168087905e-07,
      "learning_rate": 0.2293868461767672,
      "loss": 0.0,
      "num_input_tokens_seen": 7330936,
      "step": 12900
    },
    {
      "epoch": 226.4070796460177,
      "grad_norm": 1.2009093097731238e-07,
      "learning_rate": 0.22933686112782758,
      "loss": 0.0,
      "num_input_tokens_seen": 7333400,
      "step": 12905
    },
    {
      "epoch": 226.49557522123894,
      "grad_norm": 2.52236162623376e-07,
      "learning_rate": 0.22928686384414698,
      "loss": 0.0,
      "num_input_tokens_seen": 7336056,
      "step": 12910
    },
    {
      "epoch": 226.58407079646017,
      "grad_norm": 2.0431423308764352e-07,
      "learning_rate": 0.22923685433343552,
      "loss": 0.0,
      "num_input_tokens_seen": 7339064,
      "step": 12915
    },
    {
      "epoch": 226.67256637168143,
      "grad_norm": 8.888812175200655e-08,
      "learning_rate": 0.22918683260340542,
      "loss": 0.0,
      "num_input_tokens_seen": 7342264,
      "step": 12920
    },
    {
      "epoch": 226.76106194690266,
      "grad_norm": 3.0201735512491723e-07,
      "learning_rate": 0.2291367986617706,
      "loss": 0.0,
      "num_input_tokens_seen": 7344872,
      "step": 12925
    },
    {
      "epoch": 226.8495575221239,
      "grad_norm": 1.9705346687715064e-07,
      "learning_rate": 0.22908675251624697,
      "loss": 0.0,
      "num_input_tokens_seen": 7347752,
      "step": 12930
    },
    {
      "epoch": 226.93805309734512,
      "grad_norm": 1.715404494007089e-07,
      "learning_rate": 0.22903669417455216,
      "loss": 0.0,
      "num_input_tokens_seen": 7350632,
      "step": 12935
    },
    {
      "epoch": 227.01769911504425,
      "grad_norm": 1.8676725943578276e-08,
      "learning_rate": 0.22898662364440592,
      "loss": 0.0,
      "num_input_tokens_seen": 7353528,
      "step": 12940
    },
    {
      "epoch": 227.10619469026548,
      "grad_norm": 9.794651134598098e-08,
      "learning_rate": 0.2289365409335297,
      "loss": 0.0,
      "num_input_tokens_seen": 7356520,
      "step": 12945
    },
    {
      "epoch": 227.1946902654867,
      "grad_norm": 1.692681337317481e-07,
      "learning_rate": 0.2288864460496469,
      "loss": 0.0,
      "num_input_tokens_seen": 7359640,
      "step": 12950
    },
    {
      "epoch": 227.28318584070797,
      "grad_norm": 1.4435357797992765e-07,
      "learning_rate": 0.22883633900048272,
      "loss": 0.0,
      "num_input_tokens_seen": 7362504,
      "step": 12955
    },
    {
      "epoch": 227.3716814159292,
      "grad_norm": 3.7982462686159124e-07,
      "learning_rate": 0.2287862197937644,
      "loss": 0.0,
      "num_input_tokens_seen": 7365048,
      "step": 12960
    },
    {
      "epoch": 227.46017699115043,
      "grad_norm": 1.5413110077133751e-07,
      "learning_rate": 0.2287360884372209,
      "loss": 0.0,
      "num_input_tokens_seen": 7367976,
      "step": 12965
    },
    {
      "epoch": 227.5486725663717,
      "grad_norm": 2.7868668439623434e-07,
      "learning_rate": 0.22868594493858307,
      "loss": 0.0,
      "num_input_tokens_seen": 7370664,
      "step": 12970
    },
    {
      "epoch": 227.63716814159292,
      "grad_norm": 1.2751722522352793e-07,
      "learning_rate": 0.2286357893055837,
      "loss": 0.0,
      "num_input_tokens_seen": 7373832,
      "step": 12975
    },
    {
      "epoch": 227.72566371681415,
      "grad_norm": 1.1333040106364933e-07,
      "learning_rate": 0.22858562154595746,
      "loss": 0.0,
      "num_input_tokens_seen": 7376888,
      "step": 12980
    },
    {
      "epoch": 227.81415929203538,
      "grad_norm": 2.826891147833521e-07,
      "learning_rate": 0.22853544166744078,
      "loss": 0.0,
      "num_input_tokens_seen": 7379784,
      "step": 12985
    },
    {
      "epoch": 227.90265486725664,
      "grad_norm": 2.091875188625636e-07,
      "learning_rate": 0.22848524967777206,
      "loss": 0.0,
      "num_input_tokens_seen": 7382792,
      "step": 12990
    },
    {
      "epoch": 227.99115044247787,
      "grad_norm": 2.9369294907155563e-07,
      "learning_rate": 0.22843504558469152,
      "loss": 0.0,
      "num_input_tokens_seen": 7385400,
      "step": 12995
    },
    {
      "epoch": 228.07079646017698,
      "grad_norm": 9.852700344481491e-08,
      "learning_rate": 0.2283848293959413,
      "loss": 0.0,
      "num_input_tokens_seen": 7387544,
      "step": 13000
    },
    {
      "epoch": 228.07079646017698,
      "eval_loss": 0.5155118703842163,
      "eval_runtime": 0.9364,
      "eval_samples_per_second": 26.697,
      "eval_steps_per_second": 13.882,
      "num_input_tokens_seen": 7387544,
      "step": 13000
    },
    {
      "epoch": 228.15929203539824,
      "grad_norm": 2.473234985700401e-07,
      "learning_rate": 0.22833460111926532,
      "loss": 0.0,
      "num_input_tokens_seen": 7390040,
      "step": 13005
    },
    {
      "epoch": 228.24778761061947,
      "grad_norm": 1.5035975309274363e-07,
      "learning_rate": 0.22828436076240946,
      "loss": 0.0,
      "num_input_tokens_seen": 7392728,
      "step": 13010
    },
    {
      "epoch": 228.3362831858407,
      "grad_norm": 4.0559064018452773e-07,
      "learning_rate": 0.22823410833312135,
      "loss": 0.0,
      "num_input_tokens_seen": 7395432,
      "step": 13015
    },
    {
      "epoch": 228.42477876106196,
      "grad_norm": 1.3695444067707285e-07,
      "learning_rate": 0.2281838438391506,
      "loss": 0.0,
      "num_input_tokens_seen": 7398024,
      "step": 13020
    },
    {
      "epoch": 228.5132743362832,
      "grad_norm": 1.201549935103685e-07,
      "learning_rate": 0.22813356728824863,
      "loss": 0.0,
      "num_input_tokens_seen": 7401032,
      "step": 13025
    },
    {
      "epoch": 228.60176991150442,
      "grad_norm": 1.7291259268858994e-07,
      "learning_rate": 0.2280832786881687,
      "loss": 0.0,
      "num_input_tokens_seen": 7403880,
      "step": 13030
    },
    {
      "epoch": 228.69026548672565,
      "grad_norm": 3.930601266688427e-08,
      "learning_rate": 0.22803297804666592,
      "loss": 0.0,
      "num_input_tokens_seen": 7407064,
      "step": 13035
    },
    {
      "epoch": 228.7787610619469,
      "grad_norm": 1.4642513868068363e-08,
      "learning_rate": 0.22798266537149728,
      "loss": 0.0,
      "num_input_tokens_seen": 7409832,
      "step": 13040
    },
    {
      "epoch": 228.86725663716814,
      "grad_norm": 2.330566530872602e-07,
      "learning_rate": 0.22793234067042167,
      "loss": 0.0,
      "num_input_tokens_seen": 7412776,
      "step": 13045
    },
    {
      "epoch": 228.95575221238937,
      "grad_norm": 1.506413696006348e-07,
      "learning_rate": 0.22788200395119979,
      "loss": 0.0,
      "num_input_tokens_seen": 7415768,
      "step": 13050
    },
    {
      "epoch": 229.0353982300885,
      "grad_norm": 3.2726933341109543e-07,
      "learning_rate": 0.2278316552215942,
      "loss": 0.0,
      "num_input_tokens_seen": 7418296,
      "step": 13055
    },
    {
      "epoch": 229.12389380530973,
      "grad_norm": 1.5190136082310346e-07,
      "learning_rate": 0.22778129448936918,
      "loss": 0.0,
      "num_input_tokens_seen": 7420952,
      "step": 13060
    },
    {
      "epoch": 229.21238938053096,
      "grad_norm": 1.734777299589041e-07,
      "learning_rate": 0.22773092176229118,
      "loss": 0.0,
      "num_input_tokens_seen": 7423912,
      "step": 13065
    },
    {
      "epoch": 229.30088495575222,
      "grad_norm": 2.0497672892361152e-07,
      "learning_rate": 0.22768053704812816,
      "loss": 0.0,
      "num_input_tokens_seen": 7426264,
      "step": 13070
    },
    {
      "epoch": 229.38938053097345,
      "grad_norm": 2.3342381894053688e-07,
      "learning_rate": 0.22763014035465018,
      "loss": 0.0,
      "num_input_tokens_seen": 7429368,
      "step": 13075
    },
    {
      "epoch": 229.47787610619469,
      "grad_norm": 4.9142450109229685e-08,
      "learning_rate": 0.22757973168962892,
      "loss": 0.0,
      "num_input_tokens_seen": 7432120,
      "step": 13080
    },
    {
      "epoch": 229.56637168141592,
      "grad_norm": 1.5448104306869936e-07,
      "learning_rate": 0.22752931106083818,
      "loss": 0.0,
      "num_input_tokens_seen": 7435144,
      "step": 13085
    },
    {
      "epoch": 229.65486725663717,
      "grad_norm": 1.029226694981844e-07,
      "learning_rate": 0.22747887847605341,
      "loss": 0.0,
      "num_input_tokens_seen": 7438040,
      "step": 13090
    },
    {
      "epoch": 229.7433628318584,
      "grad_norm": 1.1244738118421083e-07,
      "learning_rate": 0.22742843394305184,
      "loss": 0.0,
      "num_input_tokens_seen": 7441160,
      "step": 13095
    },
    {
      "epoch": 229.83185840707964,
      "grad_norm": 2.1851153064744722e-07,
      "learning_rate": 0.22737797746961272,
      "loss": 0.0,
      "num_input_tokens_seen": 7444408,
      "step": 13100
    },
    {
      "epoch": 229.9203539823009,
      "grad_norm": 2.03382512609096e-07,
      "learning_rate": 0.22732750906351712,
      "loss": 0.0,
      "num_input_tokens_seen": 7447208,
      "step": 13105
    },
    {
      "epoch": 230.0,
      "grad_norm": 2.6268194730505456e-08,
      "learning_rate": 0.22727702873254785,
      "loss": 0.0,
      "num_input_tokens_seen": 7449656,
      "step": 13110
    },
    {
      "epoch": 230.08849557522123,
      "grad_norm": 2.272135901648653e-07,
      "learning_rate": 0.22722653648448968,
      "loss": 0.0,
      "num_input_tokens_seen": 7452312,
      "step": 13115
    },
    {
      "epoch": 230.1769911504425,
      "grad_norm": 1.3316589786427357e-07,
      "learning_rate": 0.22717603232712902,
      "loss": 0.0,
      "num_input_tokens_seen": 7454904,
      "step": 13120
    },
    {
      "epoch": 230.26548672566372,
      "grad_norm": 5.073731657034841e-08,
      "learning_rate": 0.22712551626825436,
      "loss": 0.0,
      "num_input_tokens_seen": 7457400,
      "step": 13125
    },
    {
      "epoch": 230.35398230088495,
      "grad_norm": 1.4682116500353004e-07,
      "learning_rate": 0.2270749883156559,
      "loss": 0.0,
      "num_input_tokens_seen": 7460424,
      "step": 13130
    },
    {
      "epoch": 230.44247787610618,
      "grad_norm": 5.7583520174375735e-08,
      "learning_rate": 0.22702444847712563,
      "loss": 0.0,
      "num_input_tokens_seen": 7463832,
      "step": 13135
    },
    {
      "epoch": 230.53097345132744,
      "grad_norm": 1.5088373572780256e-07,
      "learning_rate": 0.22697389676045743,
      "loss": 0.0,
      "num_input_tokens_seen": 7466520,
      "step": 13140
    },
    {
      "epoch": 230.61946902654867,
      "grad_norm": 4.2896863305941224e-07,
      "learning_rate": 0.22692333317344704,
      "loss": 0.0,
      "num_input_tokens_seen": 7469576,
      "step": 13145
    },
    {
      "epoch": 230.7079646017699,
      "grad_norm": 1.4654298752247996e-07,
      "learning_rate": 0.22687275772389198,
      "loss": 0.0,
      "num_input_tokens_seen": 7472120,
      "step": 13150
    },
    {
      "epoch": 230.79646017699116,
      "grad_norm": 1.0079144630026349e-07,
      "learning_rate": 0.22682217041959168,
      "loss": 0.0,
      "num_input_tokens_seen": 7475128,
      "step": 13155
    },
    {
      "epoch": 230.8849557522124,
      "grad_norm": 7.909792287819073e-08,
      "learning_rate": 0.2267715712683473,
      "loss": 0.0,
      "num_input_tokens_seen": 7478616,
      "step": 13160
    },
    {
      "epoch": 230.97345132743362,
      "grad_norm": 1.0788301096908981e-07,
      "learning_rate": 0.22672096027796182,
      "loss": 0.0,
      "num_input_tokens_seen": 7481336,
      "step": 13165
    },
    {
      "epoch": 231.05309734513276,
      "grad_norm": 1.3944898569206998e-07,
      "learning_rate": 0.22667033745624016,
      "loss": 0.0,
      "num_input_tokens_seen": 7483656,
      "step": 13170
    },
    {
      "epoch": 231.141592920354,
      "grad_norm": 8.761520575717441e-08,
      "learning_rate": 0.22661970281098895,
      "loss": 0.0,
      "num_input_tokens_seen": 7486552,
      "step": 13175
    },
    {
      "epoch": 231.23008849557522,
      "grad_norm": 1.9537120010681974e-07,
      "learning_rate": 0.22656905635001667,
      "loss": 0.0,
      "num_input_tokens_seen": 7489160,
      "step": 13180
    },
    {
      "epoch": 231.31858407079645,
      "grad_norm": 2.7822039783131913e-07,
      "learning_rate": 0.2265183980811337,
      "loss": 0.0,
      "num_input_tokens_seen": 7492088,
      "step": 13185
    },
    {
      "epoch": 231.4070796460177,
      "grad_norm": 2.3056364284457231e-07,
      "learning_rate": 0.22646772801215218,
      "loss": 0.0,
      "num_input_tokens_seen": 7494856,
      "step": 13190
    },
    {
      "epoch": 231.49557522123894,
      "grad_norm": 2.64628283730417e-07,
      "learning_rate": 0.22641704615088598,
      "loss": 0.0,
      "num_input_tokens_seen": 7497224,
      "step": 13195
    },
    {
      "epoch": 231.58407079646017,
      "grad_norm": 9.605379602817266e-08,
      "learning_rate": 0.22636635250515103,
      "loss": 0.0,
      "num_input_tokens_seen": 7500200,
      "step": 13200
    },
    {
      "epoch": 231.58407079646017,
      "eval_loss": 0.5223559737205505,
      "eval_runtime": 0.9387,
      "eval_samples_per_second": 26.632,
      "eval_steps_per_second": 13.849,
      "num_input_tokens_seen": 7500200,
      "step": 13200
    },
    {
      "epoch": 231.67256637168143,
      "grad_norm": 5.35792210598629e-08,
      "learning_rate": 0.2263156470827648,
      "loss": 0.0,
      "num_input_tokens_seen": 7503064,
      "step": 13205
    },
    {
      "epoch": 231.76106194690266,
      "grad_norm": 1.3816706712077576e-07,
      "learning_rate": 0.22626492989154678,
      "loss": 0.0,
      "num_input_tokens_seen": 7506248,
      "step": 13210
    },
    {
      "epoch": 231.8495575221239,
      "grad_norm": 1.2032234053549473e-07,
      "learning_rate": 0.22621420093931813,
      "loss": 0.0,
      "num_input_tokens_seen": 7509816,
      "step": 13215
    },
    {
      "epoch": 231.93805309734512,
      "grad_norm": 2.911886838319333e-07,
      "learning_rate": 0.22616346023390194,
      "loss": 0.0,
      "num_input_tokens_seen": 7512536,
      "step": 13220
    },
    {
      "epoch": 232.01769911504425,
      "grad_norm": 8.917180593925877e-08,
      "learning_rate": 0.22611270778312306,
      "loss": 0.0,
      "num_input_tokens_seen": 7514808,
      "step": 13225
    },
    {
      "epoch": 232.10619469026548,
      "grad_norm": 8.42208436324654e-08,
      "learning_rate": 0.2260619435948081,
      "loss": 0.0,
      "num_input_tokens_seen": 7517448,
      "step": 13230
    },
    {
      "epoch": 232.1946902654867,
      "grad_norm": 2.317940044349598e-07,
      "learning_rate": 0.22601116767678567,
      "loss": 0.0,
      "num_input_tokens_seen": 7520392,
      "step": 13235
    },
    {
      "epoch": 232.28318584070797,
      "grad_norm": 1.4000461590057967e-07,
      "learning_rate": 0.2259603800368859,
      "loss": 0.0,
      "num_input_tokens_seen": 7523544,
      "step": 13240
    },
    {
      "epoch": 232.3716814159292,
      "grad_norm": 2.2937180688131775e-07,
      "learning_rate": 0.22590958068294098,
      "loss": 0.0,
      "num_input_tokens_seen": 7526312,
      "step": 13245
    },
    {
      "epoch": 232.46017699115043,
      "grad_norm": 1.5351567128618626e-08,
      "learning_rate": 0.22585876962278478,
      "loss": 0.0,
      "num_input_tokens_seen": 7529512,
      "step": 13250
    },
    {
      "epoch": 232.5486725663717,
      "grad_norm": 2.931399762928777e-07,
      "learning_rate": 0.22580794686425298,
      "loss": 0.0,
      "num_input_tokens_seen": 7532680,
      "step": 13255
    },
    {
      "epoch": 232.63716814159292,
      "grad_norm": 1.6108195666220126e-07,
      "learning_rate": 0.22575711241518312,
      "loss": 0.0,
      "num_input_tokens_seen": 7535384,
      "step": 13260
    },
    {
      "epoch": 232.72566371681415,
      "grad_norm": 2.2461414062036056e-07,
      "learning_rate": 0.22570626628341453,
      "loss": 0.0,
      "num_input_tokens_seen": 7538376,
      "step": 13265
    },
    {
      "epoch": 232.81415929203538,
      "grad_norm": 2.2685225076202187e-07,
      "learning_rate": 0.22565540847678828,
      "loss": 0.0,
      "num_input_tokens_seen": 7540728,
      "step": 13270
    },
    {
      "epoch": 232.90265486725664,
      "grad_norm": 1.0076487200194606e-07,
      "learning_rate": 0.2256045390031473,
      "loss": 0.0,
      "num_input_tokens_seen": 7543848,
      "step": 13275
    },
    {
      "epoch": 232.99115044247787,
      "grad_norm": 4.863818858780178e-08,
      "learning_rate": 0.22555365787033627,
      "loss": 0.0,
      "num_input_tokens_seen": 7546680,
      "step": 13280
    },
    {
      "epoch": 233.07079646017698,
      "grad_norm": 9.012196500179925e-08,
      "learning_rate": 0.22550276508620173,
      "loss": 0.0,
      "num_input_tokens_seen": 7548792,
      "step": 13285
    },
    {
      "epoch": 233.15929203539824,
      "grad_norm": 8.210285074028434e-08,
      "learning_rate": 0.22545186065859202,
      "loss": 0.0,
      "num_input_tokens_seen": 7552184,
      "step": 13290
    },
    {
      "epoch": 233.24778761061947,
      "grad_norm": 9.337158246580657e-08,
      "learning_rate": 0.2254009445953572,
      "loss": 0.0,
      "num_input_tokens_seen": 7555480,
      "step": 13295
    },
    {
      "epoch": 233.3362831858407,
      "grad_norm": 2.3670577320444863e-07,
      "learning_rate": 0.22535001690434917,
      "loss": 0.0,
      "num_input_tokens_seen": 7558584,
      "step": 13300
    },
    {
      "epoch": 233.42477876106196,
      "grad_norm": 1.2158277229445957e-07,
      "learning_rate": 0.22529907759342163,
      "loss": 0.0,
      "num_input_tokens_seen": 7561160,
      "step": 13305
    },
    {
      "epoch": 233.5132743362832,
      "grad_norm": 5.333944486096698e-08,
      "learning_rate": 0.22524812667043007,
      "loss": 0.0,
      "num_input_tokens_seen": 7563464,
      "step": 13310
    },
    {
      "epoch": 233.60176991150442,
      "grad_norm": 1.5588577184644237e-07,
      "learning_rate": 0.22519716414323177,
      "loss": 0.0,
      "num_input_tokens_seen": 7566280,
      "step": 13315
    },
    {
      "epoch": 233.69026548672565,
      "grad_norm": 9.762036512483974e-08,
      "learning_rate": 0.22514619001968567,
      "loss": 0.0,
      "num_input_tokens_seen": 7569448,
      "step": 13320
    },
    {
      "epoch": 233.7787610619469,
      "grad_norm": 1.3770301166005083e-07,
      "learning_rate": 0.2250952043076528,
      "loss": 0.0,
      "num_input_tokens_seen": 7572344,
      "step": 13325
    },
    {
      "epoch": 233.86725663716814,
      "grad_norm": 3.9409073337992595e-07,
      "learning_rate": 0.2250442070149957,
      "loss": 0.0,
      "num_input_tokens_seen": 7575016,
      "step": 13330
    },
    {
      "epoch": 233.95575221238937,
      "grad_norm": 2.6318477353015624e-07,
      "learning_rate": 0.22499319814957885,
      "loss": 0.0,
      "num_input_tokens_seen": 7577656,
      "step": 13335
    },
    {
      "epoch": 234.0353982300885,
      "grad_norm": 4.177179491193783e-08,
      "learning_rate": 0.2249421777192684,
      "loss": 0.0,
      "num_input_tokens_seen": 7580016,
      "step": 13340
    },
    {
      "epoch": 234.12389380530973,
      "grad_norm": 1.7240311933619523e-07,
      "learning_rate": 0.22489114573193236,
      "loss": 0.0,
      "num_input_tokens_seen": 7583104,
      "step": 13345
    },
    {
      "epoch": 234.21238938053096,
      "grad_norm": 1.5972338474057324e-07,
      "learning_rate": 0.2248401021954405,
      "loss": 0.0,
      "num_input_tokens_seen": 7585952,
      "step": 13350
    },
    {
      "epoch": 234.30088495575222,
      "grad_norm": 1.933827320499404e-07,
      "learning_rate": 0.22478904711766443,
      "loss": 0.0,
      "num_input_tokens_seen": 7588480,
      "step": 13355
    },
    {
      "epoch": 234.38938053097345,
      "grad_norm": 7.857140360556514e-08,
      "learning_rate": 0.22473798050647734,
      "loss": 0.0,
      "num_input_tokens_seen": 7591248,
      "step": 13360
    },
    {
      "epoch": 234.47787610619469,
      "grad_norm": 1.0612455270120336e-07,
      "learning_rate": 0.22468690236975453,
      "loss": 0.0,
      "num_input_tokens_seen": 7594544,
      "step": 13365
    },
    {
      "epoch": 234.56637168141592,
      "grad_norm": 6.72360087605739e-08,
      "learning_rate": 0.22463581271537272,
      "loss": 0.0,
      "num_input_tokens_seen": 7597600,
      "step": 13370
    },
    {
      "epoch": 234.65486725663717,
      "grad_norm": 1.9408464879688836e-07,
      "learning_rate": 0.22458471155121076,
      "loss": 0.0,
      "num_input_tokens_seen": 7600336,
      "step": 13375
    },
    {
      "epoch": 234.7433628318584,
      "grad_norm": 3.3308771207885e-07,
      "learning_rate": 0.2245335988851489,
      "loss": 0.0,
      "num_input_tokens_seen": 7603536,
      "step": 13380
    },
    {
      "epoch": 234.83185840707964,
      "grad_norm": 1.9260458827830007e-07,
      "learning_rate": 0.2244824747250695,
      "loss": 0.0,
      "num_input_tokens_seen": 7606048,
      "step": 13385
    },
    {
      "epoch": 234.9203539823009,
      "grad_norm": 3.122568728031183e-07,
      "learning_rate": 0.22443133907885646,
      "loss": 0.0,
      "num_input_tokens_seen": 7609008,
      "step": 13390
    },
    {
      "epoch": 235.0,
      "grad_norm": 5.298866767589061e-08,
      "learning_rate": 0.22438019195439557,
      "loss": 0.0,
      "num_input_tokens_seen": 7611336,
      "step": 13395
    },
    {
      "epoch": 235.08849557522123,
      "grad_norm": 5.59909842934303e-08,
      "learning_rate": 0.22432903335957435,
      "loss": 0.0,
      "num_input_tokens_seen": 7614696,
      "step": 13400
    },
    {
      "epoch": 235.08849557522123,
      "eval_loss": 0.5160706043243408,
      "eval_runtime": 0.9512,
      "eval_samples_per_second": 26.284,
      "eval_steps_per_second": 13.667,
      "num_input_tokens_seen": 7614696,
      "step": 13400
    },
    {
      "epoch": 235.1769911504425,
      "grad_norm": 8.799434425554864e-08,
      "learning_rate": 0.22427786330228214,
      "loss": 0.0,
      "num_input_tokens_seen": 7617704,
      "step": 13405
    },
    {
      "epoch": 235.26548672566372,
      "grad_norm": 7.064246432264554e-08,
      "learning_rate": 0.22422668179040997,
      "loss": 0.0,
      "num_input_tokens_seen": 7620856,
      "step": 13410
    },
    {
      "epoch": 235.35398230088495,
      "grad_norm": 1.2515444325345015e-07,
      "learning_rate": 0.2241754888318507,
      "loss": 0.0,
      "num_input_tokens_seen": 7623880,
      "step": 13415
    },
    {
      "epoch": 235.44247787610618,
      "grad_norm": 2.0693806845883955e-07,
      "learning_rate": 0.22412428443449886,
      "loss": 0.0,
      "num_input_tokens_seen": 7626696,
      "step": 13420
    },
    {
      "epoch": 235.53097345132744,
      "grad_norm": 1.0405786099454417e-07,
      "learning_rate": 0.22407306860625087,
      "loss": 0.0,
      "num_input_tokens_seen": 7629208,
      "step": 13425
    },
    {
      "epoch": 235.61946902654867,
      "grad_norm": 2.2646235109391455e-08,
      "learning_rate": 0.22402184135500483,
      "loss": 0.0,
      "num_input_tokens_seen": 7631656,
      "step": 13430
    },
    {
      "epoch": 235.7079646017699,
      "grad_norm": 2.432120709272567e-07,
      "learning_rate": 0.22397060268866067,
      "loss": 0.0,
      "num_input_tokens_seen": 7634376,
      "step": 13435
    },
    {
      "epoch": 235.79646017699116,
      "grad_norm": 1.0972663488928447e-07,
      "learning_rate": 0.22391935261511994,
      "loss": 0.0,
      "num_input_tokens_seen": 7637336,
      "step": 13440
    },
    {
      "epoch": 235.8849557522124,
      "grad_norm": 9.786468524453085e-08,
      "learning_rate": 0.22386809114228615,
      "loss": 0.0,
      "num_input_tokens_seen": 7640328,
      "step": 13445
    },
    {
      "epoch": 235.97345132743362,
      "grad_norm": 1.4379567403466353e-07,
      "learning_rate": 0.22381681827806446,
      "loss": 0.0,
      "num_input_tokens_seen": 7643384,
      "step": 13450
    },
    {
      "epoch": 236.05309734513276,
      "grad_norm": 2.1530219385113014e-07,
      "learning_rate": 0.22376553403036173,
      "loss": 0.0,
      "num_input_tokens_seen": 7645784,
      "step": 13455
    },
    {
      "epoch": 236.141592920354,
      "grad_norm": 2.0641188314129977e-07,
      "learning_rate": 0.22371423840708662,
      "loss": 0.0,
      "num_input_tokens_seen": 7649064,
      "step": 13460
    },
    {
      "epoch": 236.23008849557522,
      "grad_norm": 1.4548278670645232e-07,
      "learning_rate": 0.22366293141614962,
      "loss": 0.0,
      "num_input_tokens_seen": 7651896,
      "step": 13465
    },
    {
      "epoch": 236.31858407079645,
      "grad_norm": 1.2186619358089956e-07,
      "learning_rate": 0.22361161306546287,
      "loss": 0.0,
      "num_input_tokens_seen": 7655032,
      "step": 13470
    },
    {
      "epoch": 236.4070796460177,
      "grad_norm": 2.2030299362540973e-07,
      "learning_rate": 0.22356028336294037,
      "loss": 0.0,
      "num_input_tokens_seen": 7657464,
      "step": 13475
    },
    {
      "epoch": 236.49557522123894,
      "grad_norm": 2.1448151699132723e-07,
      "learning_rate": 0.2235089423164977,
      "loss": 0.0,
      "num_input_tokens_seen": 7660840,
      "step": 13480
    },
    {
      "epoch": 236.58407079646017,
      "grad_norm": 1.9818853047581797e-07,
      "learning_rate": 0.22345758993405243,
      "loss": 0.0,
      "num_input_tokens_seen": 7663400,
      "step": 13485
    },
    {
      "epoch": 236.67256637168143,
      "grad_norm": 1.9654122240808647e-07,
      "learning_rate": 0.2234062262235236,
      "loss": 0.0,
      "num_input_tokens_seen": 7666120,
      "step": 13490
    },
    {
      "epoch": 236.76106194690266,
      "grad_norm": 5.6811575888104926e-08,
      "learning_rate": 0.22335485119283222,
      "loss": 0.0,
      "num_input_tokens_seen": 7668920,
      "step": 13495
    },
    {
      "epoch": 236.8495575221239,
      "grad_norm": 1.7074833635888353e-07,
      "learning_rate": 0.22330346484990093,
      "loss": 0.0,
      "num_input_tokens_seen": 7671784,
      "step": 13500
    },
    {
      "epoch": 236.93805309734512,
      "grad_norm": 1.4003654769112472e-07,
      "learning_rate": 0.22325206720265425,
      "loss": 0.0,
      "num_input_tokens_seen": 7674408,
      "step": 13505
    },
    {
      "epoch": 237.01769911504425,
      "grad_norm": 1.3728376302424294e-07,
      "learning_rate": 0.2232006582590182,
      "loss": 0.0,
      "num_input_tokens_seen": 7676832,
      "step": 13510
    },
    {
      "epoch": 237.10619469026548,
      "grad_norm": 2.2931004650672548e-07,
      "learning_rate": 0.22314923802692077,
      "loss": 0.0,
      "num_input_tokens_seen": 7679792,
      "step": 13515
    },
    {
      "epoch": 237.1946902654867,
      "grad_norm": 9.959889268884581e-08,
      "learning_rate": 0.22309780651429156,
      "loss": 0.0,
      "num_input_tokens_seen": 7682752,
      "step": 13520
    },
    {
      "epoch": 237.28318584070797,
      "grad_norm": 3.415758840219496e-07,
      "learning_rate": 0.22304636372906203,
      "loss": 0.0,
      "num_input_tokens_seen": 7685760,
      "step": 13525
    },
    {
      "epoch": 237.3716814159292,
      "grad_norm": 2.431520442769397e-07,
      "learning_rate": 0.22299490967916522,
      "loss": 0.0,
      "num_input_tokens_seen": 7689168,
      "step": 13530
    },
    {
      "epoch": 237.46017699115043,
      "grad_norm": 1.1407337296986952e-07,
      "learning_rate": 0.22294344437253602,
      "loss": 0.0,
      "num_input_tokens_seen": 7691856,
      "step": 13535
    },
    {
      "epoch": 237.5486725663717,
      "grad_norm": 1.829622391369412e-07,
      "learning_rate": 0.22289196781711101,
      "loss": 0.0,
      "num_input_tokens_seen": 7694384,
      "step": 13540
    },
    {
      "epoch": 237.63716814159292,
      "grad_norm": 1.09208642129488e-07,
      "learning_rate": 0.2228404800208286,
      "loss": 0.0,
      "num_input_tokens_seen": 7697440,
      "step": 13545
    },
    {
      "epoch": 237.72566371681415,
      "grad_norm": 1.0060184507665326e-07,
      "learning_rate": 0.22278898099162875,
      "loss": 0.0,
      "num_input_tokens_seen": 7700336,
      "step": 13550
    },
    {
      "epoch": 237.81415929203538,
      "grad_norm": 1.0756939161638002e-07,
      "learning_rate": 0.22273747073745337,
      "loss": 0.0,
      "num_input_tokens_seen": 7703616,
      "step": 13555
    },
    {
      "epoch": 237.90265486725664,
      "grad_norm": 5.3694403590043294e-08,
      "learning_rate": 0.22268594926624588,
      "loss": 0.0,
      "num_input_tokens_seen": 7706144,
      "step": 13560
    },
    {
      "epoch": 237.99115044247787,
      "grad_norm": 1.2821382711081242e-07,
      "learning_rate": 0.22263441658595162,
      "loss": 0.0,
      "num_input_tokens_seen": 7708704,
      "step": 13565
    },
    {
      "epoch": 238.07079646017698,
      "grad_norm": 2.1772461877844762e-07,
      "learning_rate": 0.2225828727045175,
      "loss": 0.0,
      "num_input_tokens_seen": 7710984,
      "step": 13570
    },
    {
      "epoch": 238.15929203539824,
      "grad_norm": 7.493192555330097e-08,
      "learning_rate": 0.22253131762989228,
      "loss": 0.0,
      "num_input_tokens_seen": 7713608,
      "step": 13575
    },
    {
      "epoch": 238.24778761061947,
      "grad_norm": 1.4854119001483923e-07,
      "learning_rate": 0.2224797513700264,
      "loss": 0.0,
      "num_input_tokens_seen": 7716488,
      "step": 13580
    },
    {
      "epoch": 238.3362831858407,
      "grad_norm": 1.404228697765575e-07,
      "learning_rate": 0.22242817393287204,
      "loss": 0.0,
      "num_input_tokens_seen": 7719640,
      "step": 13585
    },
    {
      "epoch": 238.42477876106196,
      "grad_norm": 1.1873232352854757e-07,
      "learning_rate": 0.22237658532638305,
      "loss": 0.0,
      "num_input_tokens_seen": 7722200,
      "step": 13590
    },
    {
      "epoch": 238.5132743362832,
      "grad_norm": 5.755083165581709e-08,
      "learning_rate": 0.22232498555851513,
      "loss": 0.0,
      "num_input_tokens_seen": 7724840,
      "step": 13595
    },
    {
      "epoch": 238.60176991150442,
      "grad_norm": 2.927222908510885e-07,
      "learning_rate": 0.22227337463722546,
      "loss": 0.0,
      "num_input_tokens_seen": 7727608,
      "step": 13600
    },
    {
      "epoch": 238.60176991150442,
      "eval_loss": 0.5257773399353027,
      "eval_runtime": 0.9416,
      "eval_samples_per_second": 26.55,
      "eval_steps_per_second": 13.806,
      "num_input_tokens_seen": 7727608,
      "step": 13600
    },
    {
      "epoch": 238.69026548672565,
      "grad_norm": 1.5359999849806627e-07,
      "learning_rate": 0.2222217525704732,
      "loss": 0.0,
      "num_input_tokens_seen": 7730376,
      "step": 13605
    },
    {
      "epoch": 238.7787610619469,
      "grad_norm": 1.1623088624901357e-07,
      "learning_rate": 0.22217011936621908,
      "loss": 0.0,
      "num_input_tokens_seen": 7733128,
      "step": 13610
    },
    {
      "epoch": 238.86725663716814,
      "grad_norm": 1.3574191370935296e-07,
      "learning_rate": 0.22211847503242566,
      "loss": 0.0,
      "num_input_tokens_seen": 7736056,
      "step": 13615
    },
    {
      "epoch": 238.95575221238937,
      "grad_norm": 4.7832873661945996e-08,
      "learning_rate": 0.22206681957705704,
      "loss": 0.0,
      "num_input_tokens_seen": 7739048,
      "step": 13620
    },
    {
      "epoch": 239.0353982300885,
      "grad_norm": 1.9300378184539113e-08,
      "learning_rate": 0.2220151530080792,
      "loss": 0.0,
      "num_input_tokens_seen": 7741696,
      "step": 13625
    },
    {
      "epoch": 239.12389380530973,
      "grad_norm": 1.7185854517265398e-07,
      "learning_rate": 0.2219634753334598,
      "loss": 0.0,
      "num_input_tokens_seen": 7744384,
      "step": 13630
    },
    {
      "epoch": 239.21238938053096,
      "grad_norm": 1.3765111361863092e-07,
      "learning_rate": 0.22191178656116817,
      "loss": 0.0,
      "num_input_tokens_seen": 7747776,
      "step": 13635
    },
    {
      "epoch": 239.30088495575222,
      "grad_norm": 7.344385011265331e-08,
      "learning_rate": 0.2218600866991753,
      "loss": 0.0,
      "num_input_tokens_seen": 7750400,
      "step": 13640
    },
    {
      "epoch": 239.38938053097345,
      "grad_norm": 1.611604147910839e-07,
      "learning_rate": 0.221808375755454,
      "loss": 0.0,
      "num_input_tokens_seen": 7753376,
      "step": 13645
    },
    {
      "epoch": 239.47787610619469,
      "grad_norm": 2.5138470860497364e-08,
      "learning_rate": 0.22175665373797881,
      "loss": 0.0,
      "num_input_tokens_seen": 7756048,
      "step": 13650
    },
    {
      "epoch": 239.56637168141592,
      "grad_norm": 9.119344923647077e-08,
      "learning_rate": 0.22170492065472583,
      "loss": 0.0,
      "num_input_tokens_seen": 7758720,
      "step": 13655
    },
    {
      "epoch": 239.65486725663717,
      "grad_norm": 3.137144233278377e-08,
      "learning_rate": 0.221653176513673,
      "loss": 0.0,
      "num_input_tokens_seen": 7762096,
      "step": 13660
    },
    {
      "epoch": 239.7433628318584,
      "grad_norm": 1.8929809186829516e-07,
      "learning_rate": 0.2216014213227999,
      "loss": 0.0,
      "num_input_tokens_seen": 7764944,
      "step": 13665
    },
    {
      "epoch": 239.83185840707964,
      "grad_norm": 1.9127828920773027e-07,
      "learning_rate": 0.22154965509008784,
      "loss": 0.0,
      "num_input_tokens_seen": 7767696,
      "step": 13670
    },
    {
      "epoch": 239.9203539823009,
      "grad_norm": 1.0361125646340952e-07,
      "learning_rate": 0.2214978778235198,
      "loss": 0.0,
      "num_input_tokens_seen": 7770096,
      "step": 13675
    },
    {
      "epoch": 240.0,
      "grad_norm": 8.150314556587546e-07,
      "learning_rate": 0.2214460895310805,
      "loss": 0.0,
      "num_input_tokens_seen": 7772536,
      "step": 13680
    },
    {
      "epoch": 240.08849557522123,
      "grad_norm": 1.164557446031722e-07,
      "learning_rate": 0.22139429022075635,
      "loss": 0.0,
      "num_input_tokens_seen": 7775784,
      "step": 13685
    },
    {
      "epoch": 240.1769911504425,
      "grad_norm": 1.4205002685230284e-07,
      "learning_rate": 0.22134247990053546,
      "loss": 0.0,
      "num_input_tokens_seen": 7778312,
      "step": 13690
    },
    {
      "epoch": 240.26548672566372,
      "grad_norm": 1.675143579404903e-07,
      "learning_rate": 0.2212906585784076,
      "loss": 0.0,
      "num_input_tokens_seen": 7781112,
      "step": 13695
    },
    {
      "epoch": 240.35398230088495,
      "grad_norm": 1.510160885231926e-08,
      "learning_rate": 0.22123882626236432,
      "loss": 0.0,
      "num_input_tokens_seen": 7783944,
      "step": 13700
    },
    {
      "epoch": 240.44247787610618,
      "grad_norm": 1.6188612050882512e-07,
      "learning_rate": 0.2211869829603988,
      "loss": 0.0,
      "num_input_tokens_seen": 7787192,
      "step": 13705
    },
    {
      "epoch": 240.53097345132744,
      "grad_norm": 1.7247224093352997e-07,
      "learning_rate": 0.22113512868050592,
      "loss": 0.0,
      "num_input_tokens_seen": 7790280,
      "step": 13710
    },
    {
      "epoch": 240.61946902654867,
      "grad_norm": 1.758469068136037e-07,
      "learning_rate": 0.2210832634306822,
      "loss": 0.0,
      "num_input_tokens_seen": 7793176,
      "step": 13715
    },
    {
      "epoch": 240.7079646017699,
      "grad_norm": 1.0901798930262885e-07,
      "learning_rate": 0.22103138721892598,
      "loss": 0.0,
      "num_input_tokens_seen": 7796072,
      "step": 13720
    },
    {
      "epoch": 240.79646017699116,
      "grad_norm": 1.222950203327855e-07,
      "learning_rate": 0.22097950005323724,
      "loss": 0.0,
      "num_input_tokens_seen": 7798968,
      "step": 13725
    },
    {
      "epoch": 240.8849557522124,
      "grad_norm": 1.4041638962680736e-07,
      "learning_rate": 0.22092760194161762,
      "loss": 0.0,
      "num_input_tokens_seen": 7801640,
      "step": 13730
    },
    {
      "epoch": 240.97345132743362,
      "grad_norm": 2.1586514264981815e-07,
      "learning_rate": 0.2208756928920704,
      "loss": 0.0,
      "num_input_tokens_seen": 7804472,
      "step": 13735
    },
    {
      "epoch": 241.05309734513276,
      "grad_norm": 1.2975438323792332e-07,
      "learning_rate": 0.22082377291260072,
      "loss": 0.0,
      "num_input_tokens_seen": 7806696,
      "step": 13740
    },
    {
      "epoch": 241.141592920354,
      "grad_norm": 1.0007765638420096e-07,
      "learning_rate": 0.2207718420112152,
      "loss": 0.0,
      "num_input_tokens_seen": 7809304,
      "step": 13745
    },
    {
      "epoch": 241.23008849557522,
      "grad_norm": 5.561296845257857e-08,
      "learning_rate": 0.22071990019592228,
      "loss": 0.0,
      "num_input_tokens_seen": 7811912,
      "step": 13750
    },
    {
      "epoch": 241.31858407079645,
      "grad_norm": 1.0997740673701628e-07,
      "learning_rate": 0.22066794747473198,
      "loss": 0.0,
      "num_input_tokens_seen": 7815256,
      "step": 13755
    },
    {
      "epoch": 241.4070796460177,
      "grad_norm": 1.307185186760762e-07,
      "learning_rate": 0.2206159838556562,
      "loss": 0.0,
      "num_input_tokens_seen": 7818152,
      "step": 13760
    },
    {
      "epoch": 241.49557522123894,
      "grad_norm": 1.361951404987849e-07,
      "learning_rate": 0.2205640093467082,
      "loss": 0.0,
      "num_input_tokens_seen": 7821752,
      "step": 13765
    },
    {
      "epoch": 241.58407079646017,
      "grad_norm": 7.977958205174218e-08,
      "learning_rate": 0.22051202395590322,
      "loss": 0.0,
      "num_input_tokens_seen": 7824344,
      "step": 13770
    },
    {
      "epoch": 241.67256637168143,
      "grad_norm": 2.0460380767417519e-07,
      "learning_rate": 0.22046002769125808,
      "loss": 0.0,
      "num_input_tokens_seen": 7827464,
      "step": 13775
    },
    {
      "epoch": 241.76106194690266,
      "grad_norm": 8.143172891550421e-08,
      "learning_rate": 0.2204080205607912,
      "loss": 0.0,
      "num_input_tokens_seen": 7830376,
      "step": 13780
    },
    {
      "epoch": 241.8495575221239,
      "grad_norm": 9.609262008325459e-08,
      "learning_rate": 0.22035600257252272,
      "loss": 0.0,
      "num_input_tokens_seen": 7832856,
      "step": 13785
    },
    {
      "epoch": 241.93805309734512,
      "grad_norm": 6.787737305558039e-08,
      "learning_rate": 0.2203039737344745,
      "loss": 0.0,
      "num_input_tokens_seen": 7835800,
      "step": 13790
    },
    {
      "epoch": 242.01769911504425,
      "grad_norm": 2.448912539421144e-07,
      "learning_rate": 0.22025193405467003,
      "loss": 0.0,
      "num_input_tokens_seen": 7838072,
      "step": 13795
    },
    {
      "epoch": 242.10619469026548,
      "grad_norm": 1.9076000512541214e-07,
      "learning_rate": 0.2201998835411345,
      "loss": 0.0,
      "num_input_tokens_seen": 7840696,
      "step": 13800
    },
    {
      "epoch": 242.10619469026548,
      "eval_loss": 0.5332645773887634,
      "eval_runtime": 0.9356,
      "eval_samples_per_second": 26.721,
      "eval_steps_per_second": 13.895,
      "num_input_tokens_seen": 7840696,
      "step": 13800
    },
    {
      "epoch": 242.1946902654867,
      "grad_norm": 4.217205784584621e-08,
      "learning_rate": 0.22014782220189474,
      "loss": 0.0,
      "num_input_tokens_seen": 7843656,
      "step": 13805
    },
    {
      "epoch": 242.28318584070797,
      "grad_norm": 7.804183610460314e-08,
      "learning_rate": 0.2200957500449793,
      "loss": 0.0,
      "num_input_tokens_seen": 7846568,
      "step": 13810
    },
    {
      "epoch": 242.3716814159292,
      "grad_norm": 1.5613738924002973e-07,
      "learning_rate": 0.22004366707841827,
      "loss": 0.0,
      "num_input_tokens_seen": 7849208,
      "step": 13815
    },
    {
      "epoch": 242.46017699115043,
      "grad_norm": 1.8025227177531633e-07,
      "learning_rate": 0.21999157331024358,
      "loss": 0.0,
      "num_input_tokens_seen": 7851832,
      "step": 13820
    },
    {
      "epoch": 242.5486725663717,
      "grad_norm": 1.671162266347892e-07,
      "learning_rate": 0.21993946874848871,
      "loss": 0.0,
      "num_input_tokens_seen": 7854456,
      "step": 13825
    },
    {
      "epoch": 242.63716814159292,
      "grad_norm": 1.339509623221602e-07,
      "learning_rate": 0.2198873534011888,
      "loss": 0.0,
      "num_input_tokens_seen": 7857464,
      "step": 13830
    },
    {
      "epoch": 242.72566371681415,
      "grad_norm": 2.430681718124106e-07,
      "learning_rate": 0.2198352272763808,
      "loss": 0.0,
      "num_input_tokens_seen": 7860760,
      "step": 13835
    },
    {
      "epoch": 242.81415929203538,
      "grad_norm": 1.359424715019486e-07,
      "learning_rate": 0.2197830903821031,
      "loss": 0.0,
      "num_input_tokens_seen": 7863448,
      "step": 13840
    },
    {
      "epoch": 242.90265486725664,
      "grad_norm": 1.8141631130674796e-07,
      "learning_rate": 0.21973094272639598,
      "loss": 0.0,
      "num_input_tokens_seen": 7866360,
      "step": 13845
    },
    {
      "epoch": 242.99115044247787,
      "grad_norm": 2.608825298011652e-07,
      "learning_rate": 0.21967878431730117,
      "loss": 0.0,
      "num_input_tokens_seen": 7870088,
      "step": 13850
    },
    {
      "epoch": 243.07079646017698,
      "grad_norm": 3.5753551230754965e-08,
      "learning_rate": 0.21962661516286217,
      "loss": 0.0,
      "num_input_tokens_seen": 7872152,
      "step": 13855
    },
    {
      "epoch": 243.15929203539824,
      "grad_norm": 2.0584251103628048e-07,
      "learning_rate": 0.21957443527112414,
      "loss": 0.0,
      "num_input_tokens_seen": 7874824,
      "step": 13860
    },
    {
      "epoch": 243.24778761061947,
      "grad_norm": 9.247877841289665e-08,
      "learning_rate": 0.21952224465013384,
      "loss": 0.0,
      "num_input_tokens_seen": 7877576,
      "step": 13865
    },
    {
      "epoch": 243.3362831858407,
      "grad_norm": 1.1881395067803169e-07,
      "learning_rate": 0.21947004330793976,
      "loss": 0.0,
      "num_input_tokens_seen": 7880216,
      "step": 13870
    },
    {
      "epoch": 243.42477876106196,
      "grad_norm": 1.821852322336781e-07,
      "learning_rate": 0.21941783125259198,
      "loss": 0.0,
      "num_input_tokens_seen": 7883096,
      "step": 13875
    },
    {
      "epoch": 243.5132743362832,
      "grad_norm": 5.999988417215718e-08,
      "learning_rate": 0.21936560849214226,
      "loss": 0.0,
      "num_input_tokens_seen": 7886120,
      "step": 13880
    },
    {
      "epoch": 243.60176991150442,
      "grad_norm": 1.2058698928285594e-07,
      "learning_rate": 0.21931337503464404,
      "loss": 0.0,
      "num_input_tokens_seen": 7889000,
      "step": 13885
    },
    {
      "epoch": 243.69026548672565,
      "grad_norm": 1.394865734027917e-07,
      "learning_rate": 0.21926113088815233,
      "loss": 0.0,
      "num_input_tokens_seen": 7892168,
      "step": 13890
    },
    {
      "epoch": 243.7787610619469,
      "grad_norm": 4.558276600619138e-08,
      "learning_rate": 0.2192088760607238,
      "loss": 0.0,
      "num_input_tokens_seen": 7895192,
      "step": 13895
    },
    {
      "epoch": 243.86725663716814,
      "grad_norm": 8.53177866133592e-08,
      "learning_rate": 0.2191566105604169,
      "loss": 0.0,
      "num_input_tokens_seen": 7898216,
      "step": 13900
    },
    {
      "epoch": 243.95575221238937,
      "grad_norm": 1.7587866807389219e-07,
      "learning_rate": 0.21910433439529153,
      "loss": 0.0,
      "num_input_tokens_seen": 7901288,
      "step": 13905
    },
    {
      "epoch": 244.0353982300885,
      "grad_norm": 5.652385226539991e-08,
      "learning_rate": 0.2190520475734094,
      "loss": 0.0,
      "num_input_tokens_seen": 7903720,
      "step": 13910
    },
    {
      "epoch": 244.12389380530973,
      "grad_norm": 1.0745360157216055e-07,
      "learning_rate": 0.2189997501028338,
      "loss": 0.0,
      "num_input_tokens_seen": 7906184,
      "step": 13915
    },
    {
      "epoch": 244.21238938053096,
      "grad_norm": 2.114705068834155e-07,
      "learning_rate": 0.2189474419916296,
      "loss": 0.0,
      "num_input_tokens_seen": 7908888,
      "step": 13920
    },
    {
      "epoch": 244.30088495575222,
      "grad_norm": 1.0678497375238294e-07,
      "learning_rate": 0.21889512324786342,
      "loss": 0.0,
      "num_input_tokens_seen": 7911736,
      "step": 13925
    },
    {
      "epoch": 244.38938053097345,
      "grad_norm": 2.984749514212126e-08,
      "learning_rate": 0.21884279387960345,
      "loss": 0.0,
      "num_input_tokens_seen": 7914840,
      "step": 13930
    },
    {
      "epoch": 244.47787610619469,
      "grad_norm": 1.0421387486303502e-07,
      "learning_rate": 0.2187904538949195,
      "loss": 0.0,
      "num_input_tokens_seen": 7917976,
      "step": 13935
    },
    {
      "epoch": 244.56637168141592,
      "grad_norm": 2.5855086960291374e-07,
      "learning_rate": 0.2187381033018831,
      "loss": 0.0,
      "num_input_tokens_seen": 7920792,
      "step": 13940
    },
    {
      "epoch": 244.65486725663717,
      "grad_norm": 2.515216834808598e-08,
      "learning_rate": 0.2186857421085673,
      "loss": 0.0,
      "num_input_tokens_seen": 7923960,
      "step": 13945
    },
    {
      "epoch": 244.7433628318584,
      "grad_norm": 1.3160094169961667e-07,
      "learning_rate": 0.21863337032304697,
      "loss": 0.0,
      "num_input_tokens_seen": 7926728,
      "step": 13950
    },
    {
      "epoch": 244.83185840707964,
      "grad_norm": 3.6961058214046716e-08,
      "learning_rate": 0.21858098795339845,
      "loss": 0.0,
      "num_input_tokens_seen": 7929432,
      "step": 13955
    },
    {
      "epoch": 244.9203539823009,
      "grad_norm": 1.7482962277881597e-07,
      "learning_rate": 0.21852859500769975,
      "loss": 0.0,
      "num_input_tokens_seen": 7932200,
      "step": 13960
    },
    {
      "epoch": 245.0,
      "grad_norm": 8.843733922958563e-08,
      "learning_rate": 0.21847619149403044,
      "loss": 0.0,
      "num_input_tokens_seen": 7934424,
      "step": 13965
    },
    {
      "epoch": 245.08849557522123,
      "grad_norm": 9.31732060394097e-08,
      "learning_rate": 0.21842377742047195,
      "loss": 0.0,
      "num_input_tokens_seen": 7937624,
      "step": 13970
    },
    {
      "epoch": 245.1769911504425,
      "grad_norm": 1.5332994962591329e-07,
      "learning_rate": 0.21837135279510705,
      "loss": 0.0,
      "num_input_tokens_seen": 7940504,
      "step": 13975
    },
    {
      "epoch": 245.26548672566372,
      "grad_norm": 1.1206294203702782e-07,
      "learning_rate": 0.21831891762602038,
      "loss": 0.0,
      "num_input_tokens_seen": 7943528,
      "step": 13980
    },
    {
      "epoch": 245.35398230088495,
      "grad_norm": 2.2777787478389655e-07,
      "learning_rate": 0.21826647192129806,
      "loss": 0.0,
      "num_input_tokens_seen": 7946728,
      "step": 13985
    },
    {
      "epoch": 245.44247787610618,
      "grad_norm": 1.203353861001233e-07,
      "learning_rate": 0.21821401568902787,
      "loss": 0.0,
      "num_input_tokens_seen": 7949480,
      "step": 13990
    },
    {
      "epoch": 245.53097345132744,
      "grad_norm": 1.4154991845316545e-07,
      "learning_rate": 0.21816154893729925,
      "loss": 0.0,
      "num_input_tokens_seen": 7952056,
      "step": 13995
    },
    {
      "epoch": 245.61946902654867,
      "grad_norm": 1.3212302008014376e-07,
      "learning_rate": 0.2181090716742032,
      "loss": 0.0,
      "num_input_tokens_seen": 7954632,
      "step": 14000
    },
    {
      "epoch": 245.61946902654867,
      "eval_loss": 0.5419710874557495,
      "eval_runtime": 0.936,
      "eval_samples_per_second": 26.71,
      "eval_steps_per_second": 13.889,
      "num_input_tokens_seen": 7954632,
      "step": 14000
    },
    {
      "epoch": 245.7079646017699,
      "grad_norm": 4.98760002187737e-08,
      "learning_rate": 0.21805658390783236,
      "loss": 0.0,
      "num_input_tokens_seen": 7957656,
      "step": 14005
    },
    {
      "epoch": 245.79646017699116,
      "grad_norm": 5.831762450725364e-08,
      "learning_rate": 0.21800408564628107,
      "loss": 0.0,
      "num_input_tokens_seen": 7960280,
      "step": 14010
    },
    {
      "epoch": 245.8849557522124,
      "grad_norm": 1.240344005282168e-07,
      "learning_rate": 0.21795157689764516,
      "loss": 0.0,
      "num_input_tokens_seen": 7962648,
      "step": 14015
    },
    {
      "epoch": 245.97345132743362,
      "grad_norm": 1.0722148147124244e-07,
      "learning_rate": 0.21789905767002216,
      "loss": 0.0,
      "num_input_tokens_seen": 7965832,
      "step": 14020
    },
    {
      "epoch": 246.05309734513276,
      "grad_norm": 1.968376750483003e-07,
      "learning_rate": 0.2178465279715112,
      "loss": 0.0,
      "num_input_tokens_seen": 7968384,
      "step": 14025
    },
    {
      "epoch": 246.141592920354,
      "grad_norm": 9.992005089998202e-08,
      "learning_rate": 0.21779398781021303,
      "loss": 0.0,
      "num_input_tokens_seen": 7971264,
      "step": 14030
    },
    {
      "epoch": 246.23008849557522,
      "grad_norm": 1.2191790688120818e-07,
      "learning_rate": 0.21774143719422998,
      "loss": 0.0,
      "num_input_tokens_seen": 7974224,
      "step": 14035
    },
    {
      "epoch": 246.31858407079645,
      "grad_norm": 6.926147477770428e-08,
      "learning_rate": 0.21768887613166601,
      "loss": 0.0,
      "num_input_tokens_seen": 7977392,
      "step": 14040
    },
    {
      "epoch": 246.4070796460177,
      "grad_norm": 1.6060921836924535e-07,
      "learning_rate": 0.2176363046306267,
      "loss": 0.0,
      "num_input_tokens_seen": 7980240,
      "step": 14045
    },
    {
      "epoch": 246.49557522123894,
      "grad_norm": 1.259788007246243e-07,
      "learning_rate": 0.21758372269921925,
      "loss": 0.0,
      "num_input_tokens_seen": 7982688,
      "step": 14050
    },
    {
      "epoch": 246.58407079646017,
      "grad_norm": 1.180038111670001e-07,
      "learning_rate": 0.21753113034555244,
      "loss": 0.0,
      "num_input_tokens_seen": 7985552,
      "step": 14055
    },
    {
      "epoch": 246.67256637168143,
      "grad_norm": 1.3127177567184845e-07,
      "learning_rate": 0.2174785275777367,
      "loss": 0.0,
      "num_input_tokens_seen": 7988720,
      "step": 14060
    },
    {
      "epoch": 246.76106194690266,
      "grad_norm": 8.332231260510525e-08,
      "learning_rate": 0.21742591440388404,
      "loss": 0.0,
      "num_input_tokens_seen": 7991920,
      "step": 14065
    },
    {
      "epoch": 246.8495575221239,
      "grad_norm": 8.98556322681543e-08,
      "learning_rate": 0.21737329083210802,
      "loss": 0.0,
      "num_input_tokens_seen": 7994512,
      "step": 14070
    },
    {
      "epoch": 246.93805309734512,
      "grad_norm": 5.770801791982194e-08,
      "learning_rate": 0.2173206568705239,
      "loss": 0.0,
      "num_input_tokens_seen": 7997504,
      "step": 14075
    },
    {
      "epoch": 247.01769911504425,
      "grad_norm": 1.670896807581812e-07,
      "learning_rate": 0.2172680125272485,
      "loss": 0.0,
      "num_input_tokens_seen": 7999840,
      "step": 14080
    },
    {
      "epoch": 247.10619469026548,
      "grad_norm": 1.3387490582772443e-07,
      "learning_rate": 0.2172153578104002,
      "loss": 0.0,
      "num_input_tokens_seen": 8002880,
      "step": 14085
    },
    {
      "epoch": 247.1946902654867,
      "grad_norm": 6.32661922850275e-08,
      "learning_rate": 0.21716269272809902,
      "loss": 0.0,
      "num_input_tokens_seen": 8005616,
      "step": 14090
    },
    {
      "epoch": 247.28318584070797,
      "grad_norm": 2.1322611587493157e-07,
      "learning_rate": 0.21711001728846666,
      "loss": 0.0,
      "num_input_tokens_seen": 8008224,
      "step": 14095
    },
    {
      "epoch": 247.3716814159292,
      "grad_norm": 1.073001527629458e-07,
      "learning_rate": 0.21705733149962628,
      "loss": 0.0,
      "num_input_tokens_seen": 8010832,
      "step": 14100
    },
    {
      "epoch": 247.46017699115043,
      "grad_norm": 1.2538926341676415e-07,
      "learning_rate": 0.21700463536970263,
      "loss": 0.0,
      "num_input_tokens_seen": 8013952,
      "step": 14105
    },
    {
      "epoch": 247.5486725663717,
      "grad_norm": 2.1105446990077326e-07,
      "learning_rate": 0.21695192890682222,
      "loss": 0.0,
      "num_input_tokens_seen": 8016352,
      "step": 14110
    },
    {
      "epoch": 247.63716814159292,
      "grad_norm": 2.702610117921722e-07,
      "learning_rate": 0.21689921211911298,
      "loss": 0.0,
      "num_input_tokens_seen": 8019968,
      "step": 14115
    },
    {
      "epoch": 247.72566371681415,
      "grad_norm": 1.4195252617810183e-07,
      "learning_rate": 0.21684648501470452,
      "loss": 0.0,
      "num_input_tokens_seen": 8022880,
      "step": 14120
    },
    {
      "epoch": 247.81415929203538,
      "grad_norm": 8.601875123304126e-08,
      "learning_rate": 0.216793747601728,
      "loss": 0.0,
      "num_input_tokens_seen": 8025632,
      "step": 14125
    },
    {
      "epoch": 247.90265486725664,
      "grad_norm": 9.098774711446822e-08,
      "learning_rate": 0.21674099988831627,
      "loss": 0.0,
      "num_input_tokens_seen": 8028320,
      "step": 14130
    },
    {
      "epoch": 247.99115044247787,
      "grad_norm": 1.0906324376946941e-07,
      "learning_rate": 0.21668824188260363,
      "loss": 0.0,
      "num_input_tokens_seen": 8031440,
      "step": 14135
    },
    {
      "epoch": 248.07079646017698,
      "grad_norm": 7.774207233524066e-08,
      "learning_rate": 0.21663547359272606,
      "loss": 0.0,
      "num_input_tokens_seen": 8034088,
      "step": 14140
    },
    {
      "epoch": 248.15929203539824,
      "grad_norm": 9.011811386017143e-08,
      "learning_rate": 0.216582695026821,
      "loss": 0.0,
      "num_input_tokens_seen": 8036616,
      "step": 14145
    },
    {
      "epoch": 248.24778761061947,
      "grad_norm": 2.532333098770323e-07,
      "learning_rate": 0.21652990619302767,
      "loss": 0.0,
      "num_input_tokens_seen": 8040040,
      "step": 14150
    },
    {
      "epoch": 248.3362831858407,
      "grad_norm": 2.9666697543007103e-08,
      "learning_rate": 0.21647710709948673,
      "loss": 0.0,
      "num_input_tokens_seen": 8042968,
      "step": 14155
    },
    {
      "epoch": 248.42477876106196,
      "grad_norm": 7.960321823929917e-08,
      "learning_rate": 0.2164242977543405,
      "loss": 0.0,
      "num_input_tokens_seen": 8045512,
      "step": 14160
    },
    {
      "epoch": 248.5132743362832,
      "grad_norm": 7.413921565557757e-08,
      "learning_rate": 0.21637147816573277,
      "loss": 0.0,
      "num_input_tokens_seen": 8048696,
      "step": 14165
    },
    {
      "epoch": 248.60176991150442,
      "grad_norm": 9.485130902930905e-08,
      "learning_rate": 0.21631864834180908,
      "loss": 0.0,
      "num_input_tokens_seen": 8051192,
      "step": 14170
    },
    {
      "epoch": 248.69026548672565,
      "grad_norm": 4.1016342322564014e-08,
      "learning_rate": 0.21626580829071637,
      "loss": 0.0,
      "num_input_tokens_seen": 8054232,
      "step": 14175
    },
    {
      "epoch": 248.7787610619469,
      "grad_norm": 9.267129286172349e-08,
      "learning_rate": 0.21621295802060328,
      "loss": 0.0,
      "num_input_tokens_seen": 8056824,
      "step": 14180
    },
    {
      "epoch": 248.86725663716814,
      "grad_norm": 9.988360005763752e-08,
      "learning_rate": 0.21616009753961996,
      "loss": 0.0,
      "num_input_tokens_seen": 8059928,
      "step": 14185
    },
    {
      "epoch": 248.95575221238937,
      "grad_norm": 1.7210163605341222e-07,
      "learning_rate": 0.2161072268559182,
      "loss": 0.0,
      "num_input_tokens_seen": 8062632,
      "step": 14190
    },
    {
      "epoch": 249.0353982300885,
      "grad_norm": 5.5678292198990675e-08,
      "learning_rate": 0.21605434597765133,
      "loss": 0.0,
      "num_input_tokens_seen": 8065512,
      "step": 14195
    },
    {
      "epoch": 249.12389380530973,
      "grad_norm": 1.870575090379134e-07,
      "learning_rate": 0.21600145491297418,
      "loss": 0.0,
      "num_input_tokens_seen": 8068648,
      "step": 14200
    },
    {
      "epoch": 249.12389380530973,
      "eval_loss": 0.5439008474349976,
      "eval_runtime": 0.9137,
      "eval_samples_per_second": 27.361,
      "eval_steps_per_second": 14.228,
      "num_input_tokens_seen": 8068648,
      "step": 14200
    },
    {
      "epoch": 249.21238938053096,
      "grad_norm": 1.2205622113015124e-07,
      "learning_rate": 0.21594855367004326,
      "loss": 0.0,
      "num_input_tokens_seen": 8071144,
      "step": 14205
    },
    {
      "epoch": 249.30088495575222,
      "grad_norm": 7.311970051659955e-08,
      "learning_rate": 0.21589564225701663,
      "loss": 0.0,
      "num_input_tokens_seen": 8073992,
      "step": 14210
    },
    {
      "epoch": 249.38938053097345,
      "grad_norm": 7.432397097772991e-08,
      "learning_rate": 0.21584272068205385,
      "loss": 0.0,
      "num_input_tokens_seen": 8076584,
      "step": 14215
    },
    {
      "epoch": 249.47787610619469,
      "grad_norm": 1.7361314519348525e-07,
      "learning_rate": 0.2157897889533161,
      "loss": 0.0,
      "num_input_tokens_seen": 8079144,
      "step": 14220
    },
    {
      "epoch": 249.56637168141592,
      "grad_norm": 5.1349960727975485e-08,
      "learning_rate": 0.21573684707896612,
      "loss": 0.0,
      "num_input_tokens_seen": 8082184,
      "step": 14225
    },
    {
      "epoch": 249.65486725663717,
      "grad_norm": 5.518693768635785e-08,
      "learning_rate": 0.21568389506716826,
      "loss": 0.0,
      "num_input_tokens_seen": 8085208,
      "step": 14230
    },
    {
      "epoch": 249.7433628318584,
      "grad_norm": 3.939409509712277e-08,
      "learning_rate": 0.21563093292608831,
      "loss": 0.0,
      "num_input_tokens_seen": 8088440,
      "step": 14235
    },
    {
      "epoch": 249.83185840707964,
      "grad_norm": 1.0374082393127537e-07,
      "learning_rate": 0.21557796066389376,
      "loss": 0.0,
      "num_input_tokens_seen": 8090952,
      "step": 14240
    },
    {
      "epoch": 249.9203539823009,
      "grad_norm": 5.765467747664843e-08,
      "learning_rate": 0.21552497828875353,
      "loss": 0.0,
      "num_input_tokens_seen": 8093816,
      "step": 14245
    },
    {
      "epoch": 250.0,
      "grad_norm": 1.025200973003848e-07,
      "learning_rate": 0.21547198580883828,
      "loss": 0.0,
      "num_input_tokens_seen": 8096192,
      "step": 14250
    },
    {
      "epoch": 250.08849557522123,
      "grad_norm": 1.0608835765424374e-07,
      "learning_rate": 0.21541898323232,
      "loss": 0.0,
      "num_input_tokens_seen": 8098672,
      "step": 14255
    },
    {
      "epoch": 250.1769911504425,
      "grad_norm": 1.2697438478426193e-07,
      "learning_rate": 0.2153659705673724,
      "loss": 0.0,
      "num_input_tokens_seen": 8101504,
      "step": 14260
    },
    {
      "epoch": 250.26548672566372,
      "grad_norm": 9.45214608805145e-08,
      "learning_rate": 0.2153129478221707,
      "loss": 0.0,
      "num_input_tokens_seen": 8104944,
      "step": 14265
    },
    {
      "epoch": 250.35398230088495,
      "grad_norm": 6.674330421674313e-08,
      "learning_rate": 0.21525991500489164,
      "loss": 0.0,
      "num_input_tokens_seen": 8107904,
      "step": 14270
    },
    {
      "epoch": 250.44247787610618,
      "grad_norm": 1.1776819519582205e-07,
      "learning_rate": 0.21520687212371362,
      "loss": 0.0,
      "num_input_tokens_seen": 8110656,
      "step": 14275
    },
    {
      "epoch": 250.53097345132744,
      "grad_norm": 1.417420776306244e-07,
      "learning_rate": 0.21515381918681648,
      "loss": 0.0,
      "num_input_tokens_seen": 8113280,
      "step": 14280
    },
    {
      "epoch": 250.61946902654867,
      "grad_norm": 2.4396086928391014e-07,
      "learning_rate": 0.21510075620238167,
      "loss": 0.0,
      "num_input_tokens_seen": 8115984,
      "step": 14285
    },
    {
      "epoch": 250.7079646017699,
      "grad_norm": 1.107048888115969e-07,
      "learning_rate": 0.21504768317859208,
      "loss": 0.0,
      "num_input_tokens_seen": 8119008,
      "step": 14290
    },
    {
      "epoch": 250.79646017699116,
      "grad_norm": 2.448301756885485e-07,
      "learning_rate": 0.2149946001236323,
      "loss": 0.0,
      "num_input_tokens_seen": 8121792,
      "step": 14295
    },
    {
      "epoch": 250.8849557522124,
      "grad_norm": 3.4540327931154025e-08,
      "learning_rate": 0.21494150704568848,
      "loss": 0.0,
      "num_input_tokens_seen": 8124832,
      "step": 14300
    },
    {
      "epoch": 250.97345132743362,
      "grad_norm": 1.2766338386427378e-07,
      "learning_rate": 0.21488840395294811,
      "loss": 0.0,
      "num_input_tokens_seen": 8127536,
      "step": 14305
    },
    {
      "epoch": 251.05309734513276,
      "grad_norm": 1.3328262582490424e-07,
      "learning_rate": 0.21483529085360042,
      "loss": 0.0,
      "num_input_tokens_seen": 8129992,
      "step": 14310
    },
    {
      "epoch": 251.141592920354,
      "grad_norm": 1.1606729088953216e-07,
      "learning_rate": 0.2147821677558361,
      "loss": 0.0,
      "num_input_tokens_seen": 8133000,
      "step": 14315
    },
    {
      "epoch": 251.23008849557522,
      "grad_norm": 8.224424163927324e-08,
      "learning_rate": 0.2147290346678475,
      "loss": 0.0,
      "num_input_tokens_seen": 8135608,
      "step": 14320
    },
    {
      "epoch": 251.31858407079645,
      "grad_norm": 2.80529235396898e-07,
      "learning_rate": 0.21467589159782827,
      "loss": 0.0,
      "num_input_tokens_seen": 8138792,
      "step": 14325
    },
    {
      "epoch": 251.4070796460177,
      "grad_norm": 6.835380617076225e-08,
      "learning_rate": 0.21462273855397374,
      "loss": 0.0,
      "num_input_tokens_seen": 8141400,
      "step": 14330
    },
    {
      "epoch": 251.49557522123894,
      "grad_norm": 1.2037759233862744e-07,
      "learning_rate": 0.21456957554448083,
      "loss": 0.0,
      "num_input_tokens_seen": 8144280,
      "step": 14335
    },
    {
      "epoch": 251.58407079646017,
      "grad_norm": 8.790140526571122e-08,
      "learning_rate": 0.21451640257754795,
      "loss": 0.0,
      "num_input_tokens_seen": 8147272,
      "step": 14340
    },
    {
      "epoch": 251.67256637168143,
      "grad_norm": 4.162171762800426e-08,
      "learning_rate": 0.21446321966137508,
      "loss": 0.0,
      "num_input_tokens_seen": 8150232,
      "step": 14345
    },
    {
      "epoch": 251.76106194690266,
      "grad_norm": 2.9392221989610334e-08,
      "learning_rate": 0.21441002680416354,
      "loss": 0.0,
      "num_input_tokens_seen": 8153000,
      "step": 14350
    },
    {
      "epoch": 251.8495575221239,
      "grad_norm": 5.572619343752194e-08,
      "learning_rate": 0.21435682401411654,
      "loss": 0.0,
      "num_input_tokens_seen": 8155768,
      "step": 14355
    },
    {
      "epoch": 251.93805309734512,
      "grad_norm": 1.7154786746687023e-07,
      "learning_rate": 0.2143036112994385,
      "loss": 0.0,
      "num_input_tokens_seen": 8158712,
      "step": 14360
    },
    {
      "epoch": 252.01769911504425,
      "grad_norm": 5.814758807787257e-08,
      "learning_rate": 0.21425038866833548,
      "loss": 0.0,
      "num_input_tokens_seen": 8161616,
      "step": 14365
    },
    {
      "epoch": 252.10619469026548,
      "grad_norm": 8.532948925221717e-08,
      "learning_rate": 0.21419715612901508,
      "loss": 0.0,
      "num_input_tokens_seen": 8164576,
      "step": 14370
    },
    {
      "epoch": 252.1946902654867,
      "grad_norm": 1.344986770845935e-07,
      "learning_rate": 0.21414391368968652,
      "loss": 0.0,
      "num_input_tokens_seen": 8167376,
      "step": 14375
    },
    {
      "epoch": 252.28318584070797,
      "grad_norm": 5.007839476434128e-08,
      "learning_rate": 0.21409066135856034,
      "loss": 0.0,
      "num_input_tokens_seen": 8170064,
      "step": 14380
    },
    {
      "epoch": 252.3716814159292,
      "grad_norm": 1.0373684489195512e-07,
      "learning_rate": 0.21403739914384878,
      "loss": 0.0,
      "num_input_tokens_seen": 8172304,
      "step": 14385
    },
    {
      "epoch": 252.46017699115043,
      "grad_norm": 1.4202471732005506e-07,
      "learning_rate": 0.21398412705376554,
      "loss": 0.0,
      "num_input_tokens_seen": 8175664,
      "step": 14390
    },
    {
      "epoch": 252.5486725663717,
      "grad_norm": 1.9772045334320865e-08,
      "learning_rate": 0.2139308450965258,
      "loss": 0.0,
      "num_input_tokens_seen": 8178800,
      "step": 14395
    },
    {
      "epoch": 252.63716814159292,
      "grad_norm": 1.0337406308735808e-07,
      "learning_rate": 0.21387755328034638,
      "loss": 0.0,
      "num_input_tokens_seen": 8181840,
      "step": 14400
    },
    {
      "epoch": 252.63716814159292,
      "eval_loss": 0.5526756048202515,
      "eval_runtime": 0.9216,
      "eval_samples_per_second": 27.127,
      "eval_steps_per_second": 14.106,
      "num_input_tokens_seen": 8181840,
      "step": 14400
    },
    {
      "epoch": 252.72566371681415,
      "grad_norm": 4.2634709984668007e-08,
      "learning_rate": 0.2138242516134455,
      "loss": 0.0,
      "num_input_tokens_seen": 8185568,
      "step": 14405
    },
    {
      "epoch": 252.81415929203538,
      "grad_norm": 1.1562883628357667e-07,
      "learning_rate": 0.2137709401040429,
      "loss": 0.0,
      "num_input_tokens_seen": 8188272,
      "step": 14410
    },
    {
      "epoch": 252.90265486725664,
      "grad_norm": 1.4140744042379083e-07,
      "learning_rate": 0.21371761876036,
      "loss": 0.0,
      "num_input_tokens_seen": 8190960,
      "step": 14415
    },
    {
      "epoch": 252.99115044247787,
      "grad_norm": 7.573427041052128e-08,
      "learning_rate": 0.21366428759061956,
      "loss": 0.0,
      "num_input_tokens_seen": 8193312,
      "step": 14420
    },
    {
      "epoch": 253.07079646017698,
      "grad_norm": 5.2482327816960606e-08,
      "learning_rate": 0.2136109466030459,
      "loss": 0.0,
      "num_input_tokens_seen": 8195488,
      "step": 14425
    },
    {
      "epoch": 253.15929203539824,
      "grad_norm": 6.561331389320912e-08,
      "learning_rate": 0.2135575958058649,
      "loss": 0.0,
      "num_input_tokens_seen": 8198384,
      "step": 14430
    },
    {
      "epoch": 253.24778761061947,
      "grad_norm": 8.939981199773683e-08,
      "learning_rate": 0.2135042352073039,
      "loss": 0.0,
      "num_input_tokens_seen": 8200880,
      "step": 14435
    },
    {
      "epoch": 253.3362831858407,
      "grad_norm": 8.621416469623e-08,
      "learning_rate": 0.2134508648155918,
      "loss": 0.0,
      "num_input_tokens_seen": 8203680,
      "step": 14440
    },
    {
      "epoch": 253.42477876106196,
      "grad_norm": 1.8726748862718523e-07,
      "learning_rate": 0.213397484638959,
      "loss": 0.0,
      "num_input_tokens_seen": 8206480,
      "step": 14445
    },
    {
      "epoch": 253.5132743362832,
      "grad_norm": 3.243851409706622e-08,
      "learning_rate": 0.21334409468563728,
      "loss": 0.0,
      "num_input_tokens_seen": 8209584,
      "step": 14450
    },
    {
      "epoch": 253.60176991150442,
      "grad_norm": 7.378606881047745e-08,
      "learning_rate": 0.2132906949638602,
      "loss": 0.0,
      "num_input_tokens_seen": 8212720,
      "step": 14455
    },
    {
      "epoch": 253.69026548672565,
      "grad_norm": 1.7226251713964302e-07,
      "learning_rate": 0.21323728548186255,
      "loss": 0.0,
      "num_input_tokens_seen": 8215296,
      "step": 14460
    },
    {
      "epoch": 253.7787610619469,
      "grad_norm": 1.5365351657692372e-07,
      "learning_rate": 0.21318386624788088,
      "loss": 0.0,
      "num_input_tokens_seen": 8218864,
      "step": 14465
    },
    {
      "epoch": 253.86725663716814,
      "grad_norm": 1.6260379709365225e-07,
      "learning_rate": 0.21313043727015288,
      "loss": 0.0,
      "num_input_tokens_seen": 8221824,
      "step": 14470
    },
    {
      "epoch": 253.95575221238937,
      "grad_norm": 1.3820306321576936e-07,
      "learning_rate": 0.2130769985569182,
      "loss": 0.0,
      "num_input_tokens_seen": 8224288,
      "step": 14475
    },
    {
      "epoch": 254.0353982300885,
      "grad_norm": 2.1093188706799992e-07,
      "learning_rate": 0.21302355011641766,
      "loss": 0.0,
      "num_input_tokens_seen": 8226832,
      "step": 14480
    },
    {
      "epoch": 254.12389380530973,
      "grad_norm": 4.0449709359791086e-08,
      "learning_rate": 0.21297009195689365,
      "loss": 0.0,
      "num_input_tokens_seen": 8229712,
      "step": 14485
    },
    {
      "epoch": 254.21238938053096,
      "grad_norm": 1.220660834633236e-07,
      "learning_rate": 0.21291662408659015,
      "loss": 0.0,
      "num_input_tokens_seen": 8232016,
      "step": 14490
    },
    {
      "epoch": 254.30088495575222,
      "grad_norm": 1.9121267769151018e-07,
      "learning_rate": 0.21286314651375254,
      "loss": 0.0,
      "num_input_tokens_seen": 8235040,
      "step": 14495
    },
    {
      "epoch": 254.38938053097345,
      "grad_norm": 3.1719356030635026e-08,
      "learning_rate": 0.2128096592466278,
      "loss": 0.0,
      "num_input_tokens_seen": 8238784,
      "step": 14500
    },
    {
      "epoch": 254.47787610619469,
      "grad_norm": 6.530673601901071e-08,
      "learning_rate": 0.21275616229346428,
      "loss": 0.0,
      "num_input_tokens_seen": 8241408,
      "step": 14505
    },
    {
      "epoch": 254.56637168141592,
      "grad_norm": 8.79495729577684e-08,
      "learning_rate": 0.21270265566251184,
      "loss": 0.0,
      "num_input_tokens_seen": 8244320,
      "step": 14510
    },
    {
      "epoch": 254.65486725663717,
      "grad_norm": 6.666812879529971e-08,
      "learning_rate": 0.21264913936202193,
      "loss": 0.0,
      "num_input_tokens_seen": 8246896,
      "step": 14515
    },
    {
      "epoch": 254.7433628318584,
      "grad_norm": 8.969332299102462e-08,
      "learning_rate": 0.2125956134002475,
      "loss": 0.0,
      "num_input_tokens_seen": 8249840,
      "step": 14520
    },
    {
      "epoch": 254.83185840707964,
      "grad_norm": 1.0848515330508235e-07,
      "learning_rate": 0.2125420777854428,
      "loss": 0.0,
      "num_input_tokens_seen": 8252672,
      "step": 14525
    },
    {
      "epoch": 254.9203539823009,
      "grad_norm": 1.9587515964758495e-07,
      "learning_rate": 0.21248853252586372,
      "loss": 0.0,
      "num_input_tokens_seen": 8255904,
      "step": 14530
    },
    {
      "epoch": 255.0,
      "grad_norm": 3.4515304037086025e-07,
      "learning_rate": 0.21243497762976774,
      "loss": 0.0,
      "num_input_tokens_seen": 8258352,
      "step": 14535
    },
    {
      "epoch": 255.08849557522123,
      "grad_norm": 7.929283896146444e-08,
      "learning_rate": 0.21238141310541356,
      "loss": 0.0,
      "num_input_tokens_seen": 8261040,
      "step": 14540
    },
    {
      "epoch": 255.1769911504425,
      "grad_norm": 5.647235212791202e-08,
      "learning_rate": 0.21232783896106153,
      "loss": 0.0,
      "num_input_tokens_seen": 8263712,
      "step": 14545
    },
    {
      "epoch": 255.26548672566372,
      "grad_norm": 1.2173144625649002e-07,
      "learning_rate": 0.21227425520497345,
      "loss": 0.0,
      "num_input_tokens_seen": 8266016,
      "step": 14550
    },
    {
      "epoch": 255.35398230088495,
      "grad_norm": 8.870767942426028e-08,
      "learning_rate": 0.2122206618454127,
      "loss": 0.0,
      "num_input_tokens_seen": 8268928,
      "step": 14555
    },
    {
      "epoch": 255.44247787610618,
      "grad_norm": 6.754464010327865e-08,
      "learning_rate": 0.2121670588906439,
      "loss": 0.0,
      "num_input_tokens_seen": 8272432,
      "step": 14560
    },
    {
      "epoch": 255.53097345132744,
      "grad_norm": 6.132388818969048e-08,
      "learning_rate": 0.21211344634893345,
      "loss": 0.0,
      "num_input_tokens_seen": 8274912,
      "step": 14565
    },
    {
      "epoch": 255.61946902654867,
      "grad_norm": 2.404086956175888e-07,
      "learning_rate": 0.21205982422854897,
      "loss": 0.0,
      "num_input_tokens_seen": 8278032,
      "step": 14570
    },
    {
      "epoch": 255.7079646017699,
      "grad_norm": 7.480979036245117e-08,
      "learning_rate": 0.21200619253775974,
      "loss": 0.0,
      "num_input_tokens_seen": 8280896,
      "step": 14575
    },
    {
      "epoch": 255.79646017699116,
      "grad_norm": 7.515888711395746e-08,
      "learning_rate": 0.21195255128483637,
      "loss": 0.0,
      "num_input_tokens_seen": 8283616,
      "step": 14580
    },
    {
      "epoch": 255.8849557522124,
      "grad_norm": 6.464463808697474e-08,
      "learning_rate": 0.21189890047805102,
      "loss": 0.0,
      "num_input_tokens_seen": 8286352,
      "step": 14585
    },
    {
      "epoch": 255.97345132743362,
      "grad_norm": 6.445137046284799e-08,
      "learning_rate": 0.21184524012567735,
      "loss": 0.0,
      "num_input_tokens_seen": 8289728,
      "step": 14590
    },
    {
      "epoch": 256.05309734513276,
      "grad_norm": 6.145142350533206e-08,
      "learning_rate": 0.2117915702359905,
      "loss": 0.0,
      "num_input_tokens_seen": 8291776,
      "step": 14595
    },
    {
      "epoch": 256.14159292035396,
      "grad_norm": 9.628042363374334e-08,
      "learning_rate": 0.211737890817267,
      "loss": 0.0,
      "num_input_tokens_seen": 8294896,
      "step": 14600
    },
    {
      "epoch": 256.14159292035396,
      "eval_loss": 0.5454229712486267,
      "eval_runtime": 0.9417,
      "eval_samples_per_second": 26.548,
      "eval_steps_per_second": 13.805,
      "num_input_tokens_seen": 8294896,
      "step": 14600
    },
    {
      "epoch": 256.2300884955752,
      "grad_norm": 4.978720369308576e-08,
      "learning_rate": 0.21168420187778483,
      "loss": 0.0,
      "num_input_tokens_seen": 8297696,
      "step": 14605
    },
    {
      "epoch": 256.3185840707965,
      "grad_norm": 3.830402661719745e-08,
      "learning_rate": 0.21163050342582362,
      "loss": 0.0,
      "num_input_tokens_seen": 8300496,
      "step": 14610
    },
    {
      "epoch": 256.4070796460177,
      "grad_norm": 7.984025529594874e-08,
      "learning_rate": 0.21157679546966426,
      "loss": 0.0,
      "num_input_tokens_seen": 8303488,
      "step": 14615
    },
    {
      "epoch": 256.49557522123894,
      "grad_norm": 2.237832319451627e-07,
      "learning_rate": 0.2115230780175892,
      "loss": 0.0,
      "num_input_tokens_seen": 8306624,
      "step": 14620
    },
    {
      "epoch": 256.5840707964602,
      "grad_norm": 1.1179960068830042e-07,
      "learning_rate": 0.21146935107788237,
      "loss": 0.0,
      "num_input_tokens_seen": 8309264,
      "step": 14625
    },
    {
      "epoch": 256.6725663716814,
      "grad_norm": 1.0081719636900743e-07,
      "learning_rate": 0.21141561465882916,
      "loss": 0.0,
      "num_input_tokens_seen": 8312208,
      "step": 14630
    },
    {
      "epoch": 256.76106194690266,
      "grad_norm": 7.300442916857719e-08,
      "learning_rate": 0.21136186876871635,
      "loss": 0.0,
      "num_input_tokens_seen": 8314656,
      "step": 14635
    },
    {
      "epoch": 256.8495575221239,
      "grad_norm": 3.458485053897675e-08,
      "learning_rate": 0.21130811341583225,
      "loss": 0.0,
      "num_input_tokens_seen": 8317728,
      "step": 14640
    },
    {
      "epoch": 256.9380530973451,
      "grad_norm": 2.430682570775389e-07,
      "learning_rate": 0.21125434860846667,
      "loss": 0.0,
      "num_input_tokens_seen": 8321104,
      "step": 14645
    },
    {
      "epoch": 257.01769911504425,
      "grad_norm": 9.837411596436141e-08,
      "learning_rate": 0.2112005743549107,
      "loss": 0.0,
      "num_input_tokens_seen": 8323296,
      "step": 14650
    },
    {
      "epoch": 257.1061946902655,
      "grad_norm": 8.138203355656515e-08,
      "learning_rate": 0.21114679066345707,
      "loss": 0.0,
      "num_input_tokens_seen": 8325968,
      "step": 14655
    },
    {
      "epoch": 257.1946902654867,
      "grad_norm": 7.265523294108789e-08,
      "learning_rate": 0.21109299754239993,
      "loss": 0.0,
      "num_input_tokens_seen": 8328688,
      "step": 14660
    },
    {
      "epoch": 257.283185840708,
      "grad_norm": 3.6527833202626425e-08,
      "learning_rate": 0.21103919500003482,
      "loss": 0.0,
      "num_input_tokens_seen": 8331392,
      "step": 14665
    },
    {
      "epoch": 257.37168141592923,
      "grad_norm": 2.464424575521207e-08,
      "learning_rate": 0.21098538304465872,
      "loss": 0.0,
      "num_input_tokens_seen": 8334592,
      "step": 14670
    },
    {
      "epoch": 257.46017699115043,
      "grad_norm": 1.980818922220351e-07,
      "learning_rate": 0.2109315616845702,
      "loss": 0.0,
      "num_input_tokens_seen": 8337504,
      "step": 14675
    },
    {
      "epoch": 257.5486725663717,
      "grad_norm": 3.516291258165438e-08,
      "learning_rate": 0.21087773092806925,
      "loss": 0.0,
      "num_input_tokens_seen": 8340528,
      "step": 14680
    },
    {
      "epoch": 257.6371681415929,
      "grad_norm": 8.908239834681808e-08,
      "learning_rate": 0.21082389078345704,
      "loss": 0.0,
      "num_input_tokens_seen": 8343168,
      "step": 14685
    },
    {
      "epoch": 257.72566371681415,
      "grad_norm": 1.701280751831291e-07,
      "learning_rate": 0.2107700412590365,
      "loss": 0.0,
      "num_input_tokens_seen": 8346304,
      "step": 14690
    },
    {
      "epoch": 257.8141592920354,
      "grad_norm": 1.2315182118527446e-07,
      "learning_rate": 0.210716182363112,
      "loss": 0.0,
      "num_input_tokens_seen": 8349152,
      "step": 14695
    },
    {
      "epoch": 257.9026548672566,
      "grad_norm": 1.7134304641786002e-08,
      "learning_rate": 0.2106623141039891,
      "loss": 0.0,
      "num_input_tokens_seen": 8351952,
      "step": 14700
    },
    {
      "epoch": 257.9911504424779,
      "grad_norm": 9.37928845701208e-08,
      "learning_rate": 0.21060843648997507,
      "loss": 0.0,
      "num_input_tokens_seen": 8355408,
      "step": 14705
    },
    {
      "epoch": 258.070796460177,
      "grad_norm": 5.810087344571002e-08,
      "learning_rate": 0.21055454952937844,
      "loss": 0.0,
      "num_input_tokens_seen": 8357384,
      "step": 14710
    },
    {
      "epoch": 258.1592920353982,
      "grad_norm": 1.4977747753164294e-07,
      "learning_rate": 0.21050065323050937,
      "loss": 0.0,
      "num_input_tokens_seen": 8360360,
      "step": 14715
    },
    {
      "epoch": 258.24778761061947,
      "grad_norm": 5.9190643497686324e-08,
      "learning_rate": 0.21044674760167928,
      "loss": 0.0,
      "num_input_tokens_seen": 8363064,
      "step": 14720
    },
    {
      "epoch": 258.3362831858407,
      "grad_norm": 8.061144995963332e-08,
      "learning_rate": 0.210392832651201,
      "loss": 0.0,
      "num_input_tokens_seen": 8366088,
      "step": 14725
    },
    {
      "epoch": 258.42477876106193,
      "grad_norm": 2.3237036828049895e-07,
      "learning_rate": 0.210338908387389,
      "loss": 0.0,
      "num_input_tokens_seen": 8368568,
      "step": 14730
    },
    {
      "epoch": 258.5132743362832,
      "grad_norm": 1.0883340451073309e-07,
      "learning_rate": 0.21028497481855912,
      "loss": 0.0,
      "num_input_tokens_seen": 8372072,
      "step": 14735
    },
    {
      "epoch": 258.60176991150445,
      "grad_norm": 1.2009915906219248e-07,
      "learning_rate": 0.21023103195302847,
      "loss": 0.0,
      "num_input_tokens_seen": 8375512,
      "step": 14740
    },
    {
      "epoch": 258.69026548672565,
      "grad_norm": 8.373530846483845e-08,
      "learning_rate": 0.21017707979911582,
      "loss": 0.0,
      "num_input_tokens_seen": 8378376,
      "step": 14745
    },
    {
      "epoch": 258.7787610619469,
      "grad_norm": 8.831920439433816e-08,
      "learning_rate": 0.21012311836514122,
      "loss": 0.0,
      "num_input_tokens_seen": 8380824,
      "step": 14750
    },
    {
      "epoch": 258.86725663716817,
      "grad_norm": 7.827964054740733e-08,
      "learning_rate": 0.21006914765942622,
      "loss": 0.0,
      "num_input_tokens_seen": 8383688,
      "step": 14755
    },
    {
      "epoch": 258.95575221238937,
      "grad_norm": 4.0830013148251965e-08,
      "learning_rate": 0.2100151676902938,
      "loss": 0.0,
      "num_input_tokens_seen": 8386616,
      "step": 14760
    },
    {
      "epoch": 259.0353982300885,
      "grad_norm": 4.3068737909379706e-08,
      "learning_rate": 0.2099611784660683,
      "loss": 0.0,
      "num_input_tokens_seen": 8389072,
      "step": 14765
    },
    {
      "epoch": 259.12389380530976,
      "grad_norm": 1.1273470335027014e-07,
      "learning_rate": 0.20990717999507552,
      "loss": 0.0,
      "num_input_tokens_seen": 8392064,
      "step": 14770
    },
    {
      "epoch": 259.21238938053096,
      "grad_norm": 5.541773262507377e-08,
      "learning_rate": 0.20985317228564276,
      "loss": 0.0,
      "num_input_tokens_seen": 8394864,
      "step": 14775
    },
    {
      "epoch": 259.3008849557522,
      "grad_norm": 1.1699106039486651e-07,
      "learning_rate": 0.20979915534609872,
      "loss": 0.0,
      "num_input_tokens_seen": 8397840,
      "step": 14780
    },
    {
      "epoch": 259.3893805309734,
      "grad_norm": 6.097998550558259e-08,
      "learning_rate": 0.20974512918477342,
      "loss": 0.0,
      "num_input_tokens_seen": 8400288,
      "step": 14785
    },
    {
      "epoch": 259.4778761061947,
      "grad_norm": 1.4602970566102158e-07,
      "learning_rate": 0.2096910938099984,
      "loss": 0.0,
      "num_input_tokens_seen": 8402816,
      "step": 14790
    },
    {
      "epoch": 259.56637168141594,
      "grad_norm": 8.850250310388219e-08,
      "learning_rate": 0.2096370492301066,
      "loss": 0.0,
      "num_input_tokens_seen": 8405616,
      "step": 14795
    },
    {
      "epoch": 259.65486725663715,
      "grad_norm": 9.490360497466099e-08,
      "learning_rate": 0.2095829954534323,
      "loss": 0.0,
      "num_input_tokens_seen": 8408512,
      "step": 14800
    },
    {
      "epoch": 259.65486725663715,
      "eval_loss": 0.5406162142753601,
      "eval_runtime": 0.9371,
      "eval_samples_per_second": 26.679,
      "eval_steps_per_second": 13.873,
      "num_input_tokens_seen": 8408512,
      "step": 14800
    },
    {
      "epoch": 259.7433628318584,
      "grad_norm": 1.3778387142338033e-07,
      "learning_rate": 0.2095289324883114,
      "loss": 0.0,
      "num_input_tokens_seen": 8411120,
      "step": 14805
    },
    {
      "epoch": 259.83185840707966,
      "grad_norm": 1.0958025598029053e-07,
      "learning_rate": 0.20947486034308097,
      "loss": 0.0,
      "num_input_tokens_seen": 8414224,
      "step": 14810
    },
    {
      "epoch": 259.92035398230087,
      "grad_norm": 5.157104610020724e-08,
      "learning_rate": 0.2094207790260797,
      "loss": 0.0,
      "num_input_tokens_seen": 8417392,
      "step": 14815
    },
    {
      "epoch": 260.0,
      "grad_norm": 3.8772668631281704e-08,
      "learning_rate": 0.20936668854564758,
      "loss": 0.0,
      "num_input_tokens_seen": 8420256,
      "step": 14820
    },
    {
      "epoch": 260.08849557522126,
      "grad_norm": 7.08759486656163e-08,
      "learning_rate": 0.20931258891012602,
      "loss": 0.0,
      "num_input_tokens_seen": 8423168,
      "step": 14825
    },
    {
      "epoch": 260.17699115044246,
      "grad_norm": 8.172844445653027e-08,
      "learning_rate": 0.20925848012785792,
      "loss": 0.0,
      "num_input_tokens_seen": 8425760,
      "step": 14830
    },
    {
      "epoch": 260.2654867256637,
      "grad_norm": 9.575229142910757e-08,
      "learning_rate": 0.20920436220718747,
      "loss": 0.0,
      "num_input_tokens_seen": 8428720,
      "step": 14835
    },
    {
      "epoch": 260.353982300885,
      "grad_norm": 3.9533677664849165e-08,
      "learning_rate": 0.20915023515646033,
      "loss": 0.0,
      "num_input_tokens_seen": 8431392,
      "step": 14840
    },
    {
      "epoch": 260.4424778761062,
      "grad_norm": 2.0674222867000935e-07,
      "learning_rate": 0.20909609898402368,
      "loss": 0.0,
      "num_input_tokens_seen": 8434208,
      "step": 14845
    },
    {
      "epoch": 260.53097345132744,
      "grad_norm": 5.632215049899969e-08,
      "learning_rate": 0.2090419536982258,
      "loss": 0.0,
      "num_input_tokens_seen": 8436864,
      "step": 14850
    },
    {
      "epoch": 260.6194690265487,
      "grad_norm": 8.015777552827785e-08,
      "learning_rate": 0.2089877993074168,
      "loss": 0.0,
      "num_input_tokens_seen": 8440288,
      "step": 14855
    },
    {
      "epoch": 260.7079646017699,
      "grad_norm": 1.8263547119090617e-08,
      "learning_rate": 0.20893363581994784,
      "loss": 0.0,
      "num_input_tokens_seen": 8442832,
      "step": 14860
    },
    {
      "epoch": 260.79646017699116,
      "grad_norm": 6.974161692596681e-08,
      "learning_rate": 0.2088794632441716,
      "loss": 0.0,
      "num_input_tokens_seen": 8446208,
      "step": 14865
    },
    {
      "epoch": 260.88495575221236,
      "grad_norm": 1.297104557096418e-08,
      "learning_rate": 0.20882528158844219,
      "loss": 0.0,
      "num_input_tokens_seen": 8448752,
      "step": 14870
    },
    {
      "epoch": 260.9734513274336,
      "grad_norm": 8.777006144100596e-08,
      "learning_rate": 0.20877109086111514,
      "loss": 0.0,
      "num_input_tokens_seen": 8451504,
      "step": 14875
    },
    {
      "epoch": 261.05309734513276,
      "grad_norm": 3.226301004133347e-08,
      "learning_rate": 0.2087168910705473,
      "loss": 0.0,
      "num_input_tokens_seen": 8454168,
      "step": 14880
    },
    {
      "epoch": 261.14159292035396,
      "grad_norm": 4.3140033767485875e-08,
      "learning_rate": 0.208662682225097,
      "loss": 0.0,
      "num_input_tokens_seen": 8457336,
      "step": 14885
    },
    {
      "epoch": 261.2300884955752,
      "grad_norm": 6.398609997404492e-08,
      "learning_rate": 0.2086084643331239,
      "loss": 0.0,
      "num_input_tokens_seen": 8460856,
      "step": 14890
    },
    {
      "epoch": 261.3185840707965,
      "grad_norm": 1.2208509758693253e-07,
      "learning_rate": 0.20855423740298906,
      "loss": 0.0,
      "num_input_tokens_seen": 8464136,
      "step": 14895
    },
    {
      "epoch": 261.4070796460177,
      "grad_norm": 7.810146485098812e-08,
      "learning_rate": 0.208500001443055,
      "loss": 0.0,
      "num_input_tokens_seen": 8466488,
      "step": 14900
    },
    {
      "epoch": 261.49557522123894,
      "grad_norm": 1.784162861895311e-07,
      "learning_rate": 0.20844575646168553,
      "loss": 0.0,
      "num_input_tokens_seen": 8469592,
      "step": 14905
    },
    {
      "epoch": 261.5840707964602,
      "grad_norm": 1.1425645141116547e-07,
      "learning_rate": 0.20839150246724594,
      "loss": 0.0,
      "num_input_tokens_seen": 8472168,
      "step": 14910
    },
    {
      "epoch": 261.6725663716814,
      "grad_norm": 3.754313837589507e-08,
      "learning_rate": 0.20833723946810287,
      "loss": 0.0,
      "num_input_tokens_seen": 8474760,
      "step": 14915
    },
    {
      "epoch": 261.76106194690266,
      "grad_norm": 6.753247561164244e-08,
      "learning_rate": 0.20828296747262437,
      "loss": 0.0,
      "num_input_tokens_seen": 8477768,
      "step": 14920
    },
    {
      "epoch": 261.8495575221239,
      "grad_norm": 2.206452087705202e-08,
      "learning_rate": 0.20822868648917986,
      "loss": 0.0,
      "num_input_tokens_seen": 8480360,
      "step": 14925
    },
    {
      "epoch": 261.9380530973451,
      "grad_norm": 9.415782642463455e-08,
      "learning_rate": 0.20817439652614017,
      "loss": 0.0,
      "num_input_tokens_seen": 8483336,
      "step": 14930
    },
    {
      "epoch": 262.01769911504425,
      "grad_norm": 8.295224773746668e-08,
      "learning_rate": 0.20812009759187744,
      "loss": 0.0,
      "num_input_tokens_seen": 8485600,
      "step": 14935
    },
    {
      "epoch": 262.1061946902655,
      "grad_norm": 1.2563631912598794e-07,
      "learning_rate": 0.2080657896947653,
      "loss": 0.0,
      "num_input_tokens_seen": 8488384,
      "step": 14940
    },
    {
      "epoch": 262.1946902654867,
      "grad_norm": 4.9378890309981216e-08,
      "learning_rate": 0.2080114728431787,
      "loss": 0.0,
      "num_input_tokens_seen": 8491600,
      "step": 14945
    },
    {
      "epoch": 262.283185840708,
      "grad_norm": 9.07646793280037e-08,
      "learning_rate": 0.20795714704549392,
      "loss": 0.0,
      "num_input_tokens_seen": 8494528,
      "step": 14950
    },
    {
      "epoch": 262.37168141592923,
      "grad_norm": 4.9053326733883296e-08,
      "learning_rate": 0.20790281231008875,
      "loss": 0.0,
      "num_input_tokens_seen": 8497312,
      "step": 14955
    },
    {
      "epoch": 262.46017699115043,
      "grad_norm": 1.212995641708403e-07,
      "learning_rate": 0.20784846864534226,
      "loss": 0.0,
      "num_input_tokens_seen": 8500192,
      "step": 14960
    },
    {
      "epoch": 262.5486725663717,
      "grad_norm": 2.826561562585539e-08,
      "learning_rate": 0.20779411605963496,
      "loss": 0.0,
      "num_input_tokens_seen": 8503280,
      "step": 14965
    },
    {
      "epoch": 262.6371681415929,
      "grad_norm": 4.655237617612329e-08,
      "learning_rate": 0.2077397545613487,
      "loss": 0.0,
      "num_input_tokens_seen": 8506448,
      "step": 14970
    },
    {
      "epoch": 262.72566371681415,
      "grad_norm": 7.445149918794414e-08,
      "learning_rate": 0.20768538415886661,
      "loss": 0.0,
      "num_input_tokens_seen": 8508960,
      "step": 14975
    },
    {
      "epoch": 262.8141592920354,
      "grad_norm": 6.373876715315419e-08,
      "learning_rate": 0.20763100486057343,
      "loss": 0.0,
      "num_input_tokens_seen": 8511840,
      "step": 14980
    },
    {
      "epoch": 262.9026548672566,
      "grad_norm": 8.458541600475655e-08,
      "learning_rate": 0.20757661667485502,
      "loss": 0.0,
      "num_input_tokens_seen": 8514992,
      "step": 14985
    },
    {
      "epoch": 262.9911504424779,
      "grad_norm": 9.126129185688114e-08,
      "learning_rate": 0.2075222196100988,
      "loss": 0.0,
      "num_input_tokens_seen": 8517376,
      "step": 14990
    },
    {
      "epoch": 263.070796460177,
      "grad_norm": 6.500360427708074e-08,
      "learning_rate": 0.20746781367469344,
      "loss": 0.0,
      "num_input_tokens_seen": 8520168,
      "step": 14995
    },
    {
      "epoch": 263.1592920353982,
      "grad_norm": 1.350842353531334e-07,
      "learning_rate": 0.207413398877029,
      "loss": 0.0,
      "num_input_tokens_seen": 8522664,
      "step": 15000
    },
    {
      "epoch": 263.1592920353982,
      "eval_loss": 0.5593261122703552,
      "eval_runtime": 0.9358,
      "eval_samples_per_second": 26.714,
      "eval_steps_per_second": 13.891,
      "num_input_tokens_seen": 8522664,
      "step": 15000
    },
    {
      "epoch": 263.24778761061947,
      "grad_norm": 4.053318747310186e-08,
      "learning_rate": 0.20735897522549698,
      "loss": 0.0,
      "num_input_tokens_seen": 8525480,
      "step": 15005
    },
    {
      "epoch": 263.3362831858407,
      "grad_norm": 1.0382632353866939e-07,
      "learning_rate": 0.2073045427284902,
      "loss": 0.0,
      "num_input_tokens_seen": 8528440,
      "step": 15010
    },
    {
      "epoch": 263.42477876106193,
      "grad_norm": 1.2137867599903984e-07,
      "learning_rate": 0.2072501013944027,
      "loss": 0.0,
      "num_input_tokens_seen": 8531224,
      "step": 15015
    },
    {
      "epoch": 263.5132743362832,
      "grad_norm": 7.086700293257309e-08,
      "learning_rate": 0.20719565123163017,
      "loss": 0.0,
      "num_input_tokens_seen": 8534152,
      "step": 15020
    },
    {
      "epoch": 263.60176991150445,
      "grad_norm": 1.0657812055114846e-07,
      "learning_rate": 0.20714119224856944,
      "loss": 0.0,
      "num_input_tokens_seen": 8537240,
      "step": 15025
    },
    {
      "epoch": 263.69026548672565,
      "grad_norm": 5.5095522810688635e-08,
      "learning_rate": 0.2070867244536188,
      "loss": 0.0,
      "num_input_tokens_seen": 8539960,
      "step": 15030
    },
    {
      "epoch": 263.7787610619469,
      "grad_norm": 6.964447862856105e-08,
      "learning_rate": 0.20703224785517785,
      "loss": 0.0,
      "num_input_tokens_seen": 8543400,
      "step": 15035
    },
    {
      "epoch": 263.86725663716817,
      "grad_norm": 1.4728864528024133e-07,
      "learning_rate": 0.20697776246164754,
      "loss": 0.0,
      "num_input_tokens_seen": 8545784,
      "step": 15040
    },
    {
      "epoch": 263.95575221238937,
      "grad_norm": 3.917111257578654e-08,
      "learning_rate": 0.2069232682814303,
      "loss": 0.0,
      "num_input_tokens_seen": 8548536,
      "step": 15045
    },
    {
      "epoch": 264.0353982300885,
      "grad_norm": 1.5698725519541767e-07,
      "learning_rate": 0.20686876532292972,
      "loss": 0.0,
      "num_input_tokens_seen": 8550744,
      "step": 15050
    },
    {
      "epoch": 264.12389380530976,
      "grad_norm": 2.8254461881260795e-08,
      "learning_rate": 0.20681425359455083,
      "loss": 0.0,
      "num_input_tokens_seen": 8553656,
      "step": 15055
    },
    {
      "epoch": 264.21238938053096,
      "grad_norm": 3.659360814367574e-08,
      "learning_rate": 0.20675973310470008,
      "loss": 0.0,
      "num_input_tokens_seen": 8556616,
      "step": 15060
    },
    {
      "epoch": 264.3008849557522,
      "grad_norm": 4.326602365267718e-08,
      "learning_rate": 0.2067052038617852,
      "loss": 0.0,
      "num_input_tokens_seen": 8559176,
      "step": 15065
    },
    {
      "epoch": 264.3893805309734,
      "grad_norm": 7.578302074762178e-08,
      "learning_rate": 0.2066506658742153,
      "loss": 0.0,
      "num_input_tokens_seen": 8562088,
      "step": 15070
    },
    {
      "epoch": 264.4778761061947,
      "grad_norm": 1.4896217237492237e-07,
      "learning_rate": 0.20659611915040077,
      "loss": 0.0,
      "num_input_tokens_seen": 8564936,
      "step": 15075
    },
    {
      "epoch": 264.56637168141594,
      "grad_norm": 1.4910073176110927e-08,
      "learning_rate": 0.20654156369875348,
      "loss": 0.0,
      "num_input_tokens_seen": 8567544,
      "step": 15080
    },
    {
      "epoch": 264.65486725663715,
      "grad_norm": 1.0175042319815475e-07,
      "learning_rate": 0.20648699952768648,
      "loss": 0.0,
      "num_input_tokens_seen": 8570008,
      "step": 15085
    },
    {
      "epoch": 264.7433628318584,
      "grad_norm": 3.7698999477697726e-08,
      "learning_rate": 0.20643242664561437,
      "loss": 0.0,
      "num_input_tokens_seen": 8572808,
      "step": 15090
    },
    {
      "epoch": 264.83185840707966,
      "grad_norm": 7.0585308264981e-08,
      "learning_rate": 0.20637784506095277,
      "loss": 0.0,
      "num_input_tokens_seen": 8575864,
      "step": 15095
    },
    {
      "epoch": 264.92035398230087,
      "grad_norm": 1.223796175509051e-07,
      "learning_rate": 0.20632325478211908,
      "loss": 0.0,
      "num_input_tokens_seen": 8578712,
      "step": 15100
    },
    {
      "epoch": 265.0,
      "grad_norm": 1.1112658881984316e-07,
      "learning_rate": 0.20626865581753165,
      "loss": 0.0,
      "num_input_tokens_seen": 8581640,
      "step": 15105
    },
    {
      "epoch": 265.08849557522126,
      "grad_norm": 8.534235007573443e-08,
      "learning_rate": 0.2062140481756104,
      "loss": 0.0,
      "num_input_tokens_seen": 8584232,
      "step": 15110
    },
    {
      "epoch": 265.17699115044246,
      "grad_norm": 4.245593032692341e-08,
      "learning_rate": 0.20615943186477648,
      "loss": 0.0,
      "num_input_tokens_seen": 8586904,
      "step": 15115
    },
    {
      "epoch": 265.2654867256637,
      "grad_norm": 9.996076499874107e-08,
      "learning_rate": 0.20610480689345242,
      "loss": 0.0,
      "num_input_tokens_seen": 8589480,
      "step": 15120
    },
    {
      "epoch": 265.353982300885,
      "grad_norm": 4.443929313424633e-08,
      "learning_rate": 0.2060501732700621,
      "loss": 0.0,
      "num_input_tokens_seen": 8592392,
      "step": 15125
    },
    {
      "epoch": 265.4424778761062,
      "grad_norm": 7.049123951219372e-08,
      "learning_rate": 0.20599553100303067,
      "loss": 0.0,
      "num_input_tokens_seen": 8595688,
      "step": 15130
    },
    {
      "epoch": 265.53097345132744,
      "grad_norm": 3.611452115137581e-08,
      "learning_rate": 0.20594088010078465,
      "loss": 0.0,
      "num_input_tokens_seen": 8598392,
      "step": 15135
    },
    {
      "epoch": 265.6194690265487,
      "grad_norm": 4.93326730577337e-08,
      "learning_rate": 0.20588622057175196,
      "loss": 0.0,
      "num_input_tokens_seen": 8601064,
      "step": 15140
    },
    {
      "epoch": 265.7079646017699,
      "grad_norm": 1.1810502797970912e-07,
      "learning_rate": 0.20583155242436177,
      "loss": 0.0,
      "num_input_tokens_seen": 8603752,
      "step": 15145
    },
    {
      "epoch": 265.79646017699116,
      "grad_norm": 3.872182574582439e-08,
      "learning_rate": 0.20577687566704453,
      "loss": 0.0,
      "num_input_tokens_seen": 8606504,
      "step": 15150
    },
    {
      "epoch": 265.88495575221236,
      "grad_norm": 9.34594410750833e-08,
      "learning_rate": 0.20572219030823213,
      "loss": 0.0,
      "num_input_tokens_seen": 8609864,
      "step": 15155
    },
    {
      "epoch": 265.9734513274336,
      "grad_norm": 2.8684421948810268e-08,
      "learning_rate": 0.20566749635635775,
      "loss": 0.0,
      "num_input_tokens_seen": 8613224,
      "step": 15160
    },
    {
      "epoch": 266.05309734513276,
      "grad_norm": 9.351485630304524e-08,
      "learning_rate": 0.20561279381985587,
      "loss": 0.0,
      "num_input_tokens_seen": 8615648,
      "step": 15165
    },
    {
      "epoch": 266.14159292035396,
      "grad_norm": 6.399483964969477e-08,
      "learning_rate": 0.2055580827071623,
      "loss": 0.0,
      "num_input_tokens_seen": 8618576,
      "step": 15170
    },
    {
      "epoch": 266.2300884955752,
      "grad_norm": 5.349578913182995e-08,
      "learning_rate": 0.20550336302671418,
      "loss": 0.0,
      "num_input_tokens_seen": 8621280,
      "step": 15175
    },
    {
      "epoch": 266.3185840707965,
      "grad_norm": 5.0094911330234027e-08,
      "learning_rate": 0.20544863478695,
      "loss": 0.0,
      "num_input_tokens_seen": 8624064,
      "step": 15180
    },
    {
      "epoch": 266.4070796460177,
      "grad_norm": 7.176794269980746e-08,
      "learning_rate": 0.20539389799630953,
      "loss": 0.0,
      "num_input_tokens_seen": 8626768,
      "step": 15185
    },
    {
      "epoch": 266.49557522123894,
      "grad_norm": 2.4955779664992406e-08,
      "learning_rate": 0.20533915266323388,
      "loss": 0.0,
      "num_input_tokens_seen": 8630224,
      "step": 15190
    },
    {
      "epoch": 266.5840707964602,
      "grad_norm": 1.7261923446199035e-08,
      "learning_rate": 0.20528439879616542,
      "loss": 0.0,
      "num_input_tokens_seen": 8633296,
      "step": 15195
    },
    {
      "epoch": 266.6725663716814,
      "grad_norm": 4.766123140598211e-08,
      "learning_rate": 0.20522963640354794,
      "loss": 0.0,
      "num_input_tokens_seen": 8636032,
      "step": 15200
    },
    {
      "epoch": 266.6725663716814,
      "eval_loss": 0.5593429803848267,
      "eval_runtime": 0.9141,
      "eval_samples_per_second": 27.35,
      "eval_steps_per_second": 14.222,
      "num_input_tokens_seen": 8636032,
      "step": 15200
    },
    {
      "epoch": 266.76106194690266,
      "grad_norm": 5.91427848917192e-08,
      "learning_rate": 0.20517486549382644,
      "loss": 0.0,
      "num_input_tokens_seen": 8638768,
      "step": 15205
    },
    {
      "epoch": 266.8495575221239,
      "grad_norm": 2.9621311625760427e-08,
      "learning_rate": 0.20512008607544735,
      "loss": 0.0,
      "num_input_tokens_seen": 8641520,
      "step": 15210
    },
    {
      "epoch": 266.9380530973451,
      "grad_norm": 5.509615519372346e-08,
      "learning_rate": 0.20506529815685826,
      "loss": 0.0,
      "num_input_tokens_seen": 8644592,
      "step": 15215
    },
    {
      "epoch": 267.01769911504425,
      "grad_norm": 3.450385221981378e-08,
      "learning_rate": 0.2050105017465082,
      "loss": 0.0,
      "num_input_tokens_seen": 8647352,
      "step": 15220
    },
    {
      "epoch": 267.1061946902655,
      "grad_norm": 1.7605766799988487e-07,
      "learning_rate": 0.20495569685284754,
      "loss": 0.0,
      "num_input_tokens_seen": 8650632,
      "step": 15225
    },
    {
      "epoch": 267.1946902654867,
      "grad_norm": 9.82971286589418e-08,
      "learning_rate": 0.20490088348432778,
      "loss": 0.0,
      "num_input_tokens_seen": 8653288,
      "step": 15230
    },
    {
      "epoch": 267.283185840708,
      "grad_norm": 1.0193586774676078e-07,
      "learning_rate": 0.2048460616494018,
      "loss": 0.0,
      "num_input_tokens_seen": 8656376,
      "step": 15235
    },
    {
      "epoch": 267.37168141592923,
      "grad_norm": 5.908017897127138e-08,
      "learning_rate": 0.2047912313565239,
      "loss": 0.0,
      "num_input_tokens_seen": 8659080,
      "step": 15240
    },
    {
      "epoch": 267.46017699115043,
      "grad_norm": 3.535107850893837e-08,
      "learning_rate": 0.20473639261414958,
      "loss": 0.0,
      "num_input_tokens_seen": 8662184,
      "step": 15245
    },
    {
      "epoch": 267.5486725663717,
      "grad_norm": 5.832183447296302e-08,
      "learning_rate": 0.2046815454307357,
      "loss": 0.0,
      "num_input_tokens_seen": 8664328,
      "step": 15250
    },
    {
      "epoch": 267.6371681415929,
      "grad_norm": 7.983323513371943e-08,
      "learning_rate": 0.20462668981474028,
      "loss": 0.0,
      "num_input_tokens_seen": 8667192,
      "step": 15255
    },
    {
      "epoch": 267.72566371681415,
      "grad_norm": 1.239865525803907e-07,
      "learning_rate": 0.20457182577462288,
      "loss": 0.0,
      "num_input_tokens_seen": 8669912,
      "step": 15260
    },
    {
      "epoch": 267.8141592920354,
      "grad_norm": 1.1480042871880869e-07,
      "learning_rate": 0.2045169533188441,
      "loss": 0.0,
      "num_input_tokens_seen": 8672904,
      "step": 15265
    },
    {
      "epoch": 267.9026548672566,
      "grad_norm": 5.778873557460429e-08,
      "learning_rate": 0.20446207245586603,
      "loss": 0.0,
      "num_input_tokens_seen": 8675608,
      "step": 15270
    },
    {
      "epoch": 267.9911504424779,
      "grad_norm": 7.050958572563104e-08,
      "learning_rate": 0.20440718319415196,
      "loss": 0.0,
      "num_input_tokens_seen": 8678808,
      "step": 15275
    },
    {
      "epoch": 268.070796460177,
      "grad_norm": 2.941243337772903e-08,
      "learning_rate": 0.20435228554216653,
      "loss": 0.0,
      "num_input_tokens_seen": 8681520,
      "step": 15280
    },
    {
      "epoch": 268.1592920353982,
      "grad_norm": 5.803963176731486e-08,
      "learning_rate": 0.20429737950837565,
      "loss": 0.0,
      "num_input_tokens_seen": 8684240,
      "step": 15285
    },
    {
      "epoch": 268.24778761061947,
      "grad_norm": 6.733184676477322e-08,
      "learning_rate": 0.20424246510124647,
      "loss": 0.0,
      "num_input_tokens_seen": 8687232,
      "step": 15290
    },
    {
      "epoch": 268.3362831858407,
      "grad_norm": 6.794768836471121e-08,
      "learning_rate": 0.20418754232924755,
      "loss": 0.0,
      "num_input_tokens_seen": 8690064,
      "step": 15295
    },
    {
      "epoch": 268.42477876106193,
      "grad_norm": 3.334180576075596e-08,
      "learning_rate": 0.20413261120084863,
      "loss": 0.0,
      "num_input_tokens_seen": 8692912,
      "step": 15300
    },
    {
      "epoch": 268.5132743362832,
      "grad_norm": 4.93289356029436e-08,
      "learning_rate": 0.2040776717245208,
      "loss": 0.0,
      "num_input_tokens_seen": 8695648,
      "step": 15305
    },
    {
      "epoch": 268.60176991150445,
      "grad_norm": 1.9532598116711597e-07,
      "learning_rate": 0.2040227239087364,
      "loss": 0.0,
      "num_input_tokens_seen": 8698512,
      "step": 15310
    },
    {
      "epoch": 268.69026548672565,
      "grad_norm": 7.273843039001804e-08,
      "learning_rate": 0.20396776776196904,
      "loss": 0.0,
      "num_input_tokens_seen": 8701232,
      "step": 15315
    },
    {
      "epoch": 268.7787610619469,
      "grad_norm": 5.405801317692749e-08,
      "learning_rate": 0.20391280329269373,
      "loss": 0.0,
      "num_input_tokens_seen": 8704144,
      "step": 15320
    },
    {
      "epoch": 268.86725663716817,
      "grad_norm": 1.1916811537560079e-08,
      "learning_rate": 0.20385783050938663,
      "loss": 0.0,
      "num_input_tokens_seen": 8707104,
      "step": 15325
    },
    {
      "epoch": 268.95575221238937,
      "grad_norm": 1.3493524875229923e-07,
      "learning_rate": 0.20380284942052526,
      "loss": 0.0,
      "num_input_tokens_seen": 8709664,
      "step": 15330
    },
    {
      "epoch": 269.0353982300885,
      "grad_norm": 1.2628824208604783e-07,
      "learning_rate": 0.2037478600345884,
      "loss": 0.0,
      "num_input_tokens_seen": 8711928,
      "step": 15335
    },
    {
      "epoch": 269.12389380530976,
      "grad_norm": 1.1048690851112042e-07,
      "learning_rate": 0.20369286236005604,
      "loss": 0.0,
      "num_input_tokens_seen": 8714520,
      "step": 15340
    },
    {
      "epoch": 269.21238938053096,
      "grad_norm": 6.680797781655201e-08,
      "learning_rate": 0.20363785640540957,
      "loss": 0.0,
      "num_input_tokens_seen": 8717688,
      "step": 15345
    },
    {
      "epoch": 269.3008849557522,
      "grad_norm": 6.453123546634743e-08,
      "learning_rate": 0.2035828421791316,
      "loss": 0.0,
      "num_input_tokens_seen": 8720408,
      "step": 15350
    },
    {
      "epoch": 269.3893805309734,
      "grad_norm": 1.6381722645064656e-08,
      "learning_rate": 0.20352781968970599,
      "loss": 0.0,
      "num_input_tokens_seen": 8723752,
      "step": 15355
    },
    {
      "epoch": 269.4778761061947,
      "grad_norm": 3.405954984714299e-08,
      "learning_rate": 0.2034727889456179,
      "loss": 0.0,
      "num_input_tokens_seen": 8726568,
      "step": 15360
    },
    {
      "epoch": 269.56637168141594,
      "grad_norm": 9.664385913765727e-08,
      "learning_rate": 0.2034177499553538,
      "loss": 0.0,
      "num_input_tokens_seen": 8729064,
      "step": 15365
    },
    {
      "epoch": 269.65486725663715,
      "grad_norm": 1.0605712219557972e-07,
      "learning_rate": 0.2033627027274014,
      "loss": 0.0,
      "num_input_tokens_seen": 8731960,
      "step": 15370
    },
    {
      "epoch": 269.7433628318584,
      "grad_norm": 7.534888624149971e-08,
      "learning_rate": 0.20330764727024955,
      "loss": 0.0,
      "num_input_tokens_seen": 8734936,
      "step": 15375
    },
    {
      "epoch": 269.83185840707966,
      "grad_norm": 6.250710526956027e-08,
      "learning_rate": 0.20325258359238868,
      "loss": 0.0,
      "num_input_tokens_seen": 8738168,
      "step": 15380
    },
    {
      "epoch": 269.92035398230087,
      "grad_norm": 8.147846131123515e-08,
      "learning_rate": 0.20319751170231018,
      "loss": 0.0,
      "num_input_tokens_seen": 8740680,
      "step": 15385
    },
    {
      "epoch": 270.0,
      "grad_norm": 3.005828830282553e-08,
      "learning_rate": 0.2031424316085068,
      "loss": 0.0,
      "num_input_tokens_seen": 8743392,
      "step": 15390
    },
    {
      "epoch": 270.08849557522126,
      "grad_norm": 1.2045488517742342e-07,
      "learning_rate": 0.20308734331947265,
      "loss": 0.0,
      "num_input_tokens_seen": 8746000,
      "step": 15395
    },
    {
      "epoch": 270.17699115044246,
      "grad_norm": 3.535718562375223e-08,
      "learning_rate": 0.20303224684370305,
      "loss": 0.0,
      "num_input_tokens_seen": 8748624,
      "step": 15400
    },
    {
      "epoch": 270.17699115044246,
      "eval_loss": 0.55808424949646,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 26.718,
      "eval_steps_per_second": 13.893,
      "num_input_tokens_seen": 8748624,
      "step": 15400
    },
    {
      "epoch": 270.2654867256637,
      "grad_norm": 9.544882573209179e-08,
      "learning_rate": 0.20297714218969456,
      "loss": 0.0,
      "num_input_tokens_seen": 8751984,
      "step": 15405
    },
    {
      "epoch": 270.353982300885,
      "grad_norm": 7.714422878279947e-08,
      "learning_rate": 0.20292202936594497,
      "loss": 0.0,
      "num_input_tokens_seen": 8754560,
      "step": 15410
    },
    {
      "epoch": 270.4424778761062,
      "grad_norm": 1.3567562007210654e-07,
      "learning_rate": 0.2028669083809534,
      "loss": 0.0,
      "num_input_tokens_seen": 8757600,
      "step": 15415
    },
    {
      "epoch": 270.53097345132744,
      "grad_norm": 2.639728613473835e-08,
      "learning_rate": 0.20281177924322016,
      "loss": 0.0,
      "num_input_tokens_seen": 8760640,
      "step": 15420
    },
    {
      "epoch": 270.6194690265487,
      "grad_norm": 6.162782284491186e-08,
      "learning_rate": 0.2027566419612469,
      "loss": 0.0,
      "num_input_tokens_seen": 8763744,
      "step": 15425
    },
    {
      "epoch": 270.7079646017699,
      "grad_norm": 1.0672608397044314e-07,
      "learning_rate": 0.20270149654353647,
      "loss": 0.0,
      "num_input_tokens_seen": 8766160,
      "step": 15430
    },
    {
      "epoch": 270.79646017699116,
      "grad_norm": 6.111816475140586e-08,
      "learning_rate": 0.202646342998593,
      "loss": 0.0,
      "num_input_tokens_seen": 8768832,
      "step": 15435
    },
    {
      "epoch": 270.88495575221236,
      "grad_norm": 9.849783566551196e-08,
      "learning_rate": 0.20259118133492185,
      "loss": 0.0,
      "num_input_tokens_seen": 8771968,
      "step": 15440
    },
    {
      "epoch": 270.9734513274336,
      "grad_norm": 4.267842612648565e-08,
      "learning_rate": 0.20253601156102966,
      "loss": 0.0,
      "num_input_tokens_seen": 8775184,
      "step": 15445
    },
    {
      "epoch": 271.05309734513276,
      "grad_norm": 3.863820907668014e-08,
      "learning_rate": 0.20248083368542422,
      "loss": 0.0,
      "num_input_tokens_seen": 8777984,
      "step": 15450
    },
    {
      "epoch": 271.14159292035396,
      "grad_norm": 2.3924330960767293e-08,
      "learning_rate": 0.2024256477166147,
      "loss": 0.0,
      "num_input_tokens_seen": 8780960,
      "step": 15455
    },
    {
      "epoch": 271.2300884955752,
      "grad_norm": 6.320313872265615e-08,
      "learning_rate": 0.2023704536631115,
      "loss": 0.0,
      "num_input_tokens_seen": 8784320,
      "step": 15460
    },
    {
      "epoch": 271.3185840707965,
      "grad_norm": 6.806480001841919e-08,
      "learning_rate": 0.20231525153342625,
      "loss": 0.0,
      "num_input_tokens_seen": 8787312,
      "step": 15465
    },
    {
      "epoch": 271.4070796460177,
      "grad_norm": 1.2528064985417586e-07,
      "learning_rate": 0.20226004133607173,
      "loss": 0.0,
      "num_input_tokens_seen": 8790288,
      "step": 15470
    },
    {
      "epoch": 271.49557522123894,
      "grad_norm": 4.099985773109438e-08,
      "learning_rate": 0.20220482307956214,
      "loss": 0.0,
      "num_input_tokens_seen": 8792736,
      "step": 15475
    },
    {
      "epoch": 271.5840707964602,
      "grad_norm": 1.6849686090836258e-08,
      "learning_rate": 0.20214959677241276,
      "loss": 0.0,
      "num_input_tokens_seen": 8795216,
      "step": 15480
    },
    {
      "epoch": 271.6725663716814,
      "grad_norm": 3.920475677432478e-08,
      "learning_rate": 0.20209436242314022,
      "loss": 0.0,
      "num_input_tokens_seen": 8798064,
      "step": 15485
    },
    {
      "epoch": 271.76106194690266,
      "grad_norm": 1.9954629237872723e-07,
      "learning_rate": 0.2020391200402623,
      "loss": 0.0,
      "num_input_tokens_seen": 8800928,
      "step": 15490
    },
    {
      "epoch": 271.8495575221239,
      "grad_norm": 6.408408381730624e-08,
      "learning_rate": 0.2019838696322981,
      "loss": 0.0,
      "num_input_tokens_seen": 8803616,
      "step": 15495
    },
    {
      "epoch": 271.9380530973451,
      "grad_norm": 8.908627791015533e-08,
      "learning_rate": 0.20192861120776798,
      "loss": 0.0,
      "num_input_tokens_seen": 8806624,
      "step": 15500
    },
    {
      "epoch": 272.01769911504425,
      "grad_norm": 1.2983086605800054e-07,
      "learning_rate": 0.20187334477519345,
      "loss": 0.0,
      "num_input_tokens_seen": 8808880,
      "step": 15505
    },
    {
      "epoch": 272.1061946902655,
      "grad_norm": 1.769193502809685e-08,
      "learning_rate": 0.20181807034309726,
      "loss": 0.0,
      "num_input_tokens_seen": 8812192,
      "step": 15510
    },
    {
      "epoch": 272.1946902654867,
      "grad_norm": 1.1228719642986107e-07,
      "learning_rate": 0.2017627879200034,
      "loss": 0.0,
      "num_input_tokens_seen": 8814656,
      "step": 15515
    },
    {
      "epoch": 272.283185840708,
      "grad_norm": 2.186459724384804e-08,
      "learning_rate": 0.2017074975144372,
      "loss": 0.0,
      "num_input_tokens_seen": 8817760,
      "step": 15520
    },
    {
      "epoch": 272.37168141592923,
      "grad_norm": 8.692725828041148e-08,
      "learning_rate": 0.20165219913492508,
      "loss": 0.0,
      "num_input_tokens_seen": 8820352,
      "step": 15525
    },
    {
      "epoch": 272.46017699115043,
      "grad_norm": 6.793039375452281e-08,
      "learning_rate": 0.20159689278999468,
      "loss": 0.0,
      "num_input_tokens_seen": 8823968,
      "step": 15530
    },
    {
      "epoch": 272.5486725663717,
      "grad_norm": 2.9531625145295948e-08,
      "learning_rate": 0.20154157848817508,
      "loss": 0.0,
      "num_input_tokens_seen": 8826720,
      "step": 15535
    },
    {
      "epoch": 272.6371681415929,
      "grad_norm": 3.627345535051063e-08,
      "learning_rate": 0.20148625623799632,
      "loss": 0.0,
      "num_input_tokens_seen": 8829792,
      "step": 15540
    },
    {
      "epoch": 272.72566371681415,
      "grad_norm": 1.3397683851223974e-08,
      "learning_rate": 0.20143092604798984,
      "loss": 0.0,
      "num_input_tokens_seen": 8832256,
      "step": 15545
    },
    {
      "epoch": 272.8141592920354,
      "grad_norm": 4.097962147397993e-08,
      "learning_rate": 0.2013755879266883,
      "loss": 0.0,
      "num_input_tokens_seen": 8835088,
      "step": 15550
    },
    {
      "epoch": 272.9026548672566,
      "grad_norm": 3.8378686895157443e-08,
      "learning_rate": 0.20132024188262543,
      "loss": 0.0,
      "num_input_tokens_seen": 8837888,
      "step": 15555
    },
    {
      "epoch": 272.9911504424779,
      "grad_norm": 3.2294874330318635e-08,
      "learning_rate": 0.2012648879243363,
      "loss": 0.0,
      "num_input_tokens_seen": 8840656,
      "step": 15560
    },
    {
      "epoch": 273.070796460177,
      "grad_norm": 7.933802947945878e-08,
      "learning_rate": 0.20120952606035725,
      "loss": 0.0,
      "num_input_tokens_seen": 8842656,
      "step": 15565
    },
    {
      "epoch": 273.1592920353982,
      "grad_norm": 5.204551101201105e-08,
      "learning_rate": 0.20115415629922576,
      "loss": 0.0,
      "num_input_tokens_seen": 8845168,
      "step": 15570
    },
    {
      "epoch": 273.24778761061947,
      "grad_norm": 8.269056905874095e-08,
      "learning_rate": 0.20109877864948048,
      "loss": 0.0,
      "num_input_tokens_seen": 8848032,
      "step": 15575
    },
    {
      "epoch": 273.3362831858407,
      "grad_norm": 1.4811506332534918e-07,
      "learning_rate": 0.20104339311966138,
      "loss": 0.0,
      "num_input_tokens_seen": 8851328,
      "step": 15580
    },
    {
      "epoch": 273.42477876106193,
      "grad_norm": 5.473173203540682e-08,
      "learning_rate": 0.2009879997183097,
      "loss": 0.0,
      "num_input_tokens_seen": 8854064,
      "step": 15585
    },
    {
      "epoch": 273.5132743362832,
      "grad_norm": 8.103808113446576e-08,
      "learning_rate": 0.20093259845396763,
      "loss": 0.0,
      "num_input_tokens_seen": 8856848,
      "step": 15590
    },
    {
      "epoch": 273.60176991150445,
      "grad_norm": 6.404443553265082e-08,
      "learning_rate": 0.20087718933517884,
      "loss": 0.0,
      "num_input_tokens_seen": 8860368,
      "step": 15595
    },
    {
      "epoch": 273.69026548672565,
      "grad_norm": 7.48431503438951e-08,
      "learning_rate": 0.20082177237048807,
      "loss": 0.0,
      "num_input_tokens_seen": 8863248,
      "step": 15600
    },
    {
      "epoch": 273.69026548672565,
      "eval_loss": 0.5557603240013123,
      "eval_runtime": 0.9213,
      "eval_samples_per_second": 27.137,
      "eval_steps_per_second": 14.111,
      "num_input_tokens_seen": 8863248,
      "step": 15600
    },
    {
      "epoch": 273.7787610619469,
      "grad_norm": 7.866319862159799e-08,
      "learning_rate": 0.20076634756844133,
      "loss": 0.0,
      "num_input_tokens_seen": 8865936,
      "step": 15605
    },
    {
      "epoch": 273.86725663716817,
      "grad_norm": 1.6804195368536057e-08,
      "learning_rate": 0.20071091493758586,
      "loss": 0.0,
      "num_input_tokens_seen": 8868976,
      "step": 15610
    },
    {
      "epoch": 273.95575221238937,
      "grad_norm": 4.504391171167299e-08,
      "learning_rate": 0.20065547448647003,
      "loss": 0.0,
      "num_input_tokens_seen": 8871712,
      "step": 15615
    },
    {
      "epoch": 274.0353982300885,
      "grad_norm": 6.991735546080236e-08,
      "learning_rate": 0.20060002622364348,
      "loss": 0.0,
      "num_input_tokens_seen": 8874184,
      "step": 15620
    },
    {
      "epoch": 274.12389380530976,
      "grad_norm": 7.701447657382232e-08,
      "learning_rate": 0.20054457015765695,
      "loss": 0.0,
      "num_input_tokens_seen": 8877032,
      "step": 15625
    },
    {
      "epoch": 274.21238938053096,
      "grad_norm": 1.316049065280822e-07,
      "learning_rate": 0.20048910629706254,
      "loss": 0.0,
      "num_input_tokens_seen": 8879656,
      "step": 15630
    },
    {
      "epoch": 274.3008849557522,
      "grad_norm": 6.343760361460227e-08,
      "learning_rate": 0.20043363465041347,
      "loss": 0.0,
      "num_input_tokens_seen": 8882664,
      "step": 15635
    },
    {
      "epoch": 274.3893805309734,
      "grad_norm": 2.9718540517365e-08,
      "learning_rate": 0.2003781552262641,
      "loss": 0.0,
      "num_input_tokens_seen": 8885672,
      "step": 15640
    },
    {
      "epoch": 274.4778761061947,
      "grad_norm": 5.785954115822278e-08,
      "learning_rate": 0.20032266803317014,
      "loss": 0.0,
      "num_input_tokens_seen": 8888840,
      "step": 15645
    },
    {
      "epoch": 274.56637168141594,
      "grad_norm": 3.259883385453577e-08,
      "learning_rate": 0.2002671730796884,
      "loss": 0.0,
      "num_input_tokens_seen": 8891448,
      "step": 15650
    },
    {
      "epoch": 274.65486725663715,
      "grad_norm": 7.861389406116359e-08,
      "learning_rate": 0.20021167037437684,
      "loss": 0.0,
      "num_input_tokens_seen": 8894760,
      "step": 15655
    },
    {
      "epoch": 274.7433628318584,
      "grad_norm": 1.1214579132001745e-07,
      "learning_rate": 0.20015615992579472,
      "loss": 0.0,
      "num_input_tokens_seen": 8897688,
      "step": 15660
    },
    {
      "epoch": 274.83185840707966,
      "grad_norm": 7.662499967864278e-08,
      "learning_rate": 0.20010064174250244,
      "loss": 0.0,
      "num_input_tokens_seen": 8900424,
      "step": 15665
    },
    {
      "epoch": 274.92035398230087,
      "grad_norm": 8.410760443666732e-08,
      "learning_rate": 0.2000451158330616,
      "loss": 0.0,
      "num_input_tokens_seen": 8903096,
      "step": 15670
    },
    {
      "epoch": 275.0,
      "grad_norm": 4.934129549383215e-08,
      "learning_rate": 0.199989582206035,
      "loss": 0.0,
      "num_input_tokens_seen": 8905392,
      "step": 15675
    },
    {
      "epoch": 275.08849557522126,
      "grad_norm": 1.7556903486593e-08,
      "learning_rate": 0.1999340408699866,
      "loss": 0.0,
      "num_input_tokens_seen": 8907856,
      "step": 15680
    },
    {
      "epoch": 275.17699115044246,
      "grad_norm": 2.980762303650408e-08,
      "learning_rate": 0.19987849183348155,
      "loss": 0.0,
      "num_input_tokens_seen": 8910768,
      "step": 15685
    },
    {
      "epoch": 275.2654867256637,
      "grad_norm": 2.8074124358568042e-08,
      "learning_rate": 0.19982293510508628,
      "loss": 0.0,
      "num_input_tokens_seen": 8913536,
      "step": 15690
    },
    {
      "epoch": 275.353982300885,
      "grad_norm": 5.7987250556834624e-08,
      "learning_rate": 0.19976737069336833,
      "loss": 0.0,
      "num_input_tokens_seen": 8916816,
      "step": 15695
    },
    {
      "epoch": 275.4424778761062,
      "grad_norm": 8.304599674602287e-08,
      "learning_rate": 0.1997117986068964,
      "loss": 0.0,
      "num_input_tokens_seen": 8919168,
      "step": 15700
    },
    {
      "epoch": 275.53097345132744,
      "grad_norm": 4.163000966173058e-08,
      "learning_rate": 0.19965621885424037,
      "loss": 0.0,
      "num_input_tokens_seen": 8922000,
      "step": 15705
    },
    {
      "epoch": 275.6194690265487,
      "grad_norm": 1.0364058944389853e-08,
      "learning_rate": 0.19960063144397142,
      "loss": 0.0,
      "num_input_tokens_seen": 8924960,
      "step": 15710
    },
    {
      "epoch": 275.7079646017699,
      "grad_norm": 8.263473461056492e-08,
      "learning_rate": 0.19954503638466176,
      "loss": 0.0,
      "num_input_tokens_seen": 8928016,
      "step": 15715
    },
    {
      "epoch": 275.79646017699116,
      "grad_norm": 4.879960968651176e-08,
      "learning_rate": 0.1994894336848848,
      "loss": 0.0,
      "num_input_tokens_seen": 8931008,
      "step": 15720
    },
    {
      "epoch": 275.88495575221236,
      "grad_norm": 5.52937144959742e-08,
      "learning_rate": 0.1994338233532153,
      "loss": 0.0,
      "num_input_tokens_seen": 8933776,
      "step": 15725
    },
    {
      "epoch": 275.9734513274336,
      "grad_norm": 6.170581201558889e-08,
      "learning_rate": 0.19937820539822904,
      "loss": 0.0,
      "num_input_tokens_seen": 8936688,
      "step": 15730
    },
    {
      "epoch": 276.05309734513276,
      "grad_norm": 7.043544769658183e-08,
      "learning_rate": 0.199322579828503,
      "loss": 0.0,
      "num_input_tokens_seen": 8938744,
      "step": 15735
    },
    {
      "epoch": 276.14159292035396,
      "grad_norm": 3.5735393311142616e-08,
      "learning_rate": 0.19926694665261527,
      "loss": 0.0,
      "num_input_tokens_seen": 8941256,
      "step": 15740
    },
    {
      "epoch": 276.2300884955752,
      "grad_norm": 1.9407226048429038e-08,
      "learning_rate": 0.19921130587914526,
      "loss": 0.0,
      "num_input_tokens_seen": 8943992,
      "step": 15745
    },
    {
      "epoch": 276.3185840707965,
      "grad_norm": 4.5293635508869556e-08,
      "learning_rate": 0.19915565751667344,
      "loss": 0.0,
      "num_input_tokens_seen": 8946424,
      "step": 15750
    },
    {
      "epoch": 276.4070796460177,
      "grad_norm": 3.445737206675403e-08,
      "learning_rate": 0.19910000157378152,
      "loss": 0.0,
      "num_input_tokens_seen": 8949752,
      "step": 15755
    },
    {
      "epoch": 276.49557522123894,
      "grad_norm": 5.6084775934550635e-08,
      "learning_rate": 0.1990443380590523,
      "loss": 0.0,
      "num_input_tokens_seen": 8953256,
      "step": 15760
    },
    {
      "epoch": 276.5840707964602,
      "grad_norm": 5.4348820555105704e-08,
      "learning_rate": 0.19898866698106984,
      "loss": 0.0,
      "num_input_tokens_seen": 8956360,
      "step": 15765
    },
    {
      "epoch": 276.6725663716814,
      "grad_norm": 4.5476980403691414e-08,
      "learning_rate": 0.19893298834841933,
      "loss": 0.0,
      "num_input_tokens_seen": 8959352,
      "step": 15770
    },
    {
      "epoch": 276.76106194690266,
      "grad_norm": 5.5120846553791125e-08,
      "learning_rate": 0.19887730216968705,
      "loss": 0.0,
      "num_input_tokens_seen": 8962104,
      "step": 15775
    },
    {
      "epoch": 276.8495575221239,
      "grad_norm": 1.0280210460678063e-07,
      "learning_rate": 0.19882160845346053,
      "loss": 0.0,
      "num_input_tokens_seen": 8964984,
      "step": 15780
    },
    {
      "epoch": 276.9380530973451,
      "grad_norm": 2.0166586978120904e-07,
      "learning_rate": 0.1987659072083285,
      "loss": 0.0,
      "num_input_tokens_seen": 8968008,
      "step": 15785
    },
    {
      "epoch": 277.01769911504425,
      "grad_norm": 3.310387342025933e-08,
      "learning_rate": 0.1987101984428807,
      "loss": 0.0,
      "num_input_tokens_seen": 8970792,
      "step": 15790
    },
    {
      "epoch": 277.1061946902655,
      "grad_norm": 3.0108864734756935e-08,
      "learning_rate": 0.19865448216570822,
      "loss": 0.0,
      "num_input_tokens_seen": 8973832,
      "step": 15795
    },
    {
      "epoch": 277.1946902654867,
      "grad_norm": 2.2687256517883725e-08,
      "learning_rate": 0.19859875838540317,
      "loss": 0.0,
      "num_input_tokens_seen": 8976424,
      "step": 15800
    },
    {
      "epoch": 277.1946902654867,
      "eval_loss": 0.5764188170433044,
      "eval_runtime": 0.917,
      "eval_samples_per_second": 27.263,
      "eval_steps_per_second": 14.177,
      "num_input_tokens_seen": 8976424,
      "step": 15800
    },
    {
      "epoch": 277.283185840708,
      "grad_norm": 4.697780298101861e-08,
      "learning_rate": 0.1985430271105588,
      "loss": 0.0,
      "num_input_tokens_seen": 8978760,
      "step": 15805
    },
    {
      "epoch": 277.37168141592923,
      "grad_norm": 1.1015589329588238e-08,
      "learning_rate": 0.19848728834976961,
      "loss": 0.0,
      "num_input_tokens_seen": 8981528,
      "step": 15810
    },
    {
      "epoch": 277.46017699115043,
      "grad_norm": 8.907077386766105e-08,
      "learning_rate": 0.19843154211163128,
      "loss": 0.0,
      "num_input_tokens_seen": 8984568,
      "step": 15815
    },
    {
      "epoch": 277.5486725663717,
      "grad_norm": 1.0746138912054448e-07,
      "learning_rate": 0.1983757884047405,
      "loss": 0.0,
      "num_input_tokens_seen": 8987656,
      "step": 15820
    },
    {
      "epoch": 277.6371681415929,
      "grad_norm": 4.0184335858839404e-08,
      "learning_rate": 0.1983200272376952,
      "loss": 0.0,
      "num_input_tokens_seen": 8990296,
      "step": 15825
    },
    {
      "epoch": 277.72566371681415,
      "grad_norm": 4.0687016422680244e-08,
      "learning_rate": 0.1982642586190945,
      "loss": 0.0,
      "num_input_tokens_seen": 8993656,
      "step": 15830
    },
    {
      "epoch": 277.8141592920354,
      "grad_norm": 1.8733274487203744e-08,
      "learning_rate": 0.1982084825575386,
      "loss": 0.0,
      "num_input_tokens_seen": 8996840,
      "step": 15835
    },
    {
      "epoch": 277.9026548672566,
      "grad_norm": 5.365781063915165e-08,
      "learning_rate": 0.19815269906162883,
      "loss": 0.0,
      "num_input_tokens_seen": 8999512,
      "step": 15840
    },
    {
      "epoch": 277.9911504424779,
      "grad_norm": 1.0786343551671962e-07,
      "learning_rate": 0.19809690813996775,
      "loss": 0.0,
      "num_input_tokens_seen": 9001832,
      "step": 15845
    },
    {
      "epoch": 278.070796460177,
      "grad_norm": 3.340190701806023e-08,
      "learning_rate": 0.19804110980115905,
      "loss": 0.0,
      "num_input_tokens_seen": 9004264,
      "step": 15850
    },
    {
      "epoch": 278.1592920353982,
      "grad_norm": 3.488537458906649e-08,
      "learning_rate": 0.19798530405380746,
      "loss": 0.0,
      "num_input_tokens_seen": 9006920,
      "step": 15855
    },
    {
      "epoch": 278.24778761061947,
      "grad_norm": 2.9722139416321625e-08,
      "learning_rate": 0.19792949090651893,
      "loss": 0.0,
      "num_input_tokens_seen": 9009544,
      "step": 15860
    },
    {
      "epoch": 278.3362831858407,
      "grad_norm": 1.0514196446820279e-07,
      "learning_rate": 0.19787367036790066,
      "loss": 0.0,
      "num_input_tokens_seen": 9012472,
      "step": 15865
    },
    {
      "epoch": 278.42477876106193,
      "grad_norm": 9.628074337797443e-08,
      "learning_rate": 0.19781784244656075,
      "loss": 0.0,
      "num_input_tokens_seen": 9015304,
      "step": 15870
    },
    {
      "epoch": 278.5132743362832,
      "grad_norm": 4.822276267191228e-08,
      "learning_rate": 0.19776200715110864,
      "loss": 0.0,
      "num_input_tokens_seen": 9017928,
      "step": 15875
    },
    {
      "epoch": 278.60176991150445,
      "grad_norm": 9.086424768156576e-08,
      "learning_rate": 0.1977061644901548,
      "loss": 0.0,
      "num_input_tokens_seen": 9021272,
      "step": 15880
    },
    {
      "epoch": 278.69026548672565,
      "grad_norm": 4.902606676182586e-08,
      "learning_rate": 0.1976503144723109,
      "loss": 0.0,
      "num_input_tokens_seen": 9024472,
      "step": 15885
    },
    {
      "epoch": 278.7787610619469,
      "grad_norm": 5.230311472814719e-08,
      "learning_rate": 0.19759445710618967,
      "loss": 0.0,
      "num_input_tokens_seen": 9027288,
      "step": 15890
    },
    {
      "epoch": 278.86725663716817,
      "grad_norm": 9.063401762432477e-08,
      "learning_rate": 0.19753859240040508,
      "loss": 0.0,
      "num_input_tokens_seen": 9030024,
      "step": 15895
    },
    {
      "epoch": 278.95575221238937,
      "grad_norm": 5.581898676609853e-08,
      "learning_rate": 0.1974827203635721,
      "loss": 0.0,
      "num_input_tokens_seen": 9032984,
      "step": 15900
    },
    {
      "epoch": 279.0353982300885,
      "grad_norm": 6.188042789290193e-08,
      "learning_rate": 0.19742684100430694,
      "loss": 0.0,
      "num_input_tokens_seen": 9035232,
      "step": 15905
    },
    {
      "epoch": 279.12389380530976,
      "grad_norm": 4.417390897515361e-08,
      "learning_rate": 0.19737095433122692,
      "loss": 0.0,
      "num_input_tokens_seen": 9037920,
      "step": 15910
    },
    {
      "epoch": 279.21238938053096,
      "grad_norm": 3.180096541655075e-08,
      "learning_rate": 0.19731506035295046,
      "loss": 0.0,
      "num_input_tokens_seen": 9040560,
      "step": 15915
    },
    {
      "epoch": 279.3008849557522,
      "grad_norm": 5.5686051325665176e-08,
      "learning_rate": 0.19725915907809702,
      "loss": 0.0,
      "num_input_tokens_seen": 9043664,
      "step": 15920
    },
    {
      "epoch": 279.3893805309734,
      "grad_norm": 2.550174649229575e-08,
      "learning_rate": 0.1972032505152874,
      "loss": 0.0,
      "num_input_tokens_seen": 9046544,
      "step": 15925
    },
    {
      "epoch": 279.4778761061947,
      "grad_norm": 7.3959760982234e-08,
      "learning_rate": 0.19714733467314338,
      "loss": 0.0,
      "num_input_tokens_seen": 9049456,
      "step": 15930
    },
    {
      "epoch": 279.56637168141594,
      "grad_norm": 4.703490930069165e-08,
      "learning_rate": 0.19709141156028784,
      "loss": 0.0,
      "num_input_tokens_seen": 9052560,
      "step": 15935
    },
    {
      "epoch": 279.65486725663715,
      "grad_norm": 7.445943595030258e-08,
      "learning_rate": 0.1970354811853448,
      "loss": 0.0,
      "num_input_tokens_seen": 9055312,
      "step": 15940
    },
    {
      "epoch": 279.7433628318584,
      "grad_norm": 3.6241516454538214e-08,
      "learning_rate": 0.19697954355693953,
      "loss": 0.0,
      "num_input_tokens_seen": 9058080,
      "step": 15945
    },
    {
      "epoch": 279.83185840707966,
      "grad_norm": 6.715493583442367e-08,
      "learning_rate": 0.19692359868369827,
      "loss": 0.0,
      "num_input_tokens_seen": 9061280,
      "step": 15950
    },
    {
      "epoch": 279.92035398230087,
      "grad_norm": 5.176018547103922e-08,
      "learning_rate": 0.1968676465742484,
      "loss": 0.0,
      "num_input_tokens_seen": 9064128,
      "step": 15955
    },
    {
      "epoch": 280.0,
      "grad_norm": 1.3714140045806289e-08,
      "learning_rate": 0.19681168723721845,
      "loss": 0.0,
      "num_input_tokens_seen": 9066520,
      "step": 15960
    },
    {
      "epoch": 280.08849557522126,
      "grad_norm": 1.2863401011031783e-08,
      "learning_rate": 0.19675572068123803,
      "loss": 0.0,
      "num_input_tokens_seen": 9069192,
      "step": 15965
    },
    {
      "epoch": 280.17699115044246,
      "grad_norm": 7.397404289122278e-08,
      "learning_rate": 0.19669974691493794,
      "loss": 0.0,
      "num_input_tokens_seen": 9072264,
      "step": 15970
    },
    {
      "epoch": 280.2654867256637,
      "grad_norm": 4.458270197460479e-08,
      "learning_rate": 0.19664376594695002,
      "loss": 0.0,
      "num_input_tokens_seen": 9074808,
      "step": 15975
    },
    {
      "epoch": 280.353982300885,
      "grad_norm": 9.256390853806806e-08,
      "learning_rate": 0.19658777778590722,
      "loss": 0.0,
      "num_input_tokens_seen": 9077592,
      "step": 15980
    },
    {
      "epoch": 280.4424778761062,
      "grad_norm": 2.4110882179684268e-08,
      "learning_rate": 0.19653178244044364,
      "loss": 0.0,
      "num_input_tokens_seen": 9080632,
      "step": 15985
    },
    {
      "epoch": 280.53097345132744,
      "grad_norm": 6.495034199360816e-08,
      "learning_rate": 0.19647577991919443,
      "loss": 0.0,
      "num_input_tokens_seen": 9083816,
      "step": 15990
    },
    {
      "epoch": 280.6194690265487,
      "grad_norm": 8.03336988042247e-08,
      "learning_rate": 0.1964197702307959,
      "loss": 0.0,
      "num_input_tokens_seen": 9086312,
      "step": 15995
    },
    {
      "epoch": 280.7079646017699,
      "grad_norm": 5.919824275224528e-08,
      "learning_rate": 0.19636375338388545,
      "loss": 0.0,
      "num_input_tokens_seen": 9088984,
      "step": 16000
    },
    {
      "epoch": 280.7079646017699,
      "eval_loss": 0.5791025161743164,
      "eval_runtime": 0.9334,
      "eval_samples_per_second": 26.783,
      "eval_steps_per_second": 13.927,
      "num_input_tokens_seen": 9088984,
      "step": 16000
    },
    {
      "epoch": 280.79646017699116,
      "grad_norm": 7.141792934817204e-08,
      "learning_rate": 0.1963077293871016,
      "loss": 0.0,
      "num_input_tokens_seen": 9091704,
      "step": 16005
    },
    {
      "epoch": 280.88495575221236,
      "grad_norm": 4.188337499044792e-08,
      "learning_rate": 0.19625169824908395,
      "loss": 0.0,
      "num_input_tokens_seen": 9095176,
      "step": 16010
    },
    {
      "epoch": 280.9734513274336,
      "grad_norm": 7.238379851060017e-08,
      "learning_rate": 0.19619565997847319,
      "loss": 0.0,
      "num_input_tokens_seen": 9098424,
      "step": 16015
    },
    {
      "epoch": 281.05309734513276,
      "grad_norm": 7.557277115211036e-08,
      "learning_rate": 0.19613961458391113,
      "loss": 0.0,
      "num_input_tokens_seen": 9100800,
      "step": 16020
    },
    {
      "epoch": 281.14159292035396,
      "grad_norm": 8.267761586466804e-08,
      "learning_rate": 0.19608356207404065,
      "loss": 0.0,
      "num_input_tokens_seen": 9103552,
      "step": 16025
    },
    {
      "epoch": 281.2300884955752,
      "grad_norm": 2.0887615193032616e-08,
      "learning_rate": 0.1960275024575058,
      "loss": 0.0,
      "num_input_tokens_seen": 9106368,
      "step": 16030
    },
    {
      "epoch": 281.3185840707965,
      "grad_norm": 8.878649282451079e-08,
      "learning_rate": 0.19597143574295164,
      "loss": 0.0,
      "num_input_tokens_seen": 9109008,
      "step": 16035
    },
    {
      "epoch": 281.4070796460177,
      "grad_norm": 4.3165886864926506e-08,
      "learning_rate": 0.1959153619390244,
      "loss": 0.0,
      "num_input_tokens_seen": 9112544,
      "step": 16040
    },
    {
      "epoch": 281.49557522123894,
      "grad_norm": 2.933246356917607e-08,
      "learning_rate": 0.1958592810543713,
      "loss": 0.0,
      "num_input_tokens_seen": 9115280,
      "step": 16045
    },
    {
      "epoch": 281.5840707964602,
      "grad_norm": 7.311957261890711e-08,
      "learning_rate": 0.19580319309764077,
      "loss": 0.0,
      "num_input_tokens_seen": 9117760,
      "step": 16050
    },
    {
      "epoch": 281.6725663716814,
      "grad_norm": 6.512665606805967e-08,
      "learning_rate": 0.1957470980774823,
      "loss": 0.0,
      "num_input_tokens_seen": 9120688,
      "step": 16055
    },
    {
      "epoch": 281.76106194690266,
      "grad_norm": 1.1208196326606412e-07,
      "learning_rate": 0.19569099600254639,
      "loss": 0.0,
      "num_input_tokens_seen": 9123408,
      "step": 16060
    },
    {
      "epoch": 281.8495575221239,
      "grad_norm": 3.310541529799593e-08,
      "learning_rate": 0.1956348868814847,
      "loss": 0.0,
      "num_input_tokens_seen": 9126368,
      "step": 16065
    },
    {
      "epoch": 281.9380530973451,
      "grad_norm": 6.28610052899603e-08,
      "learning_rate": 0.19557877072295,
      "loss": 0.0,
      "num_input_tokens_seen": 9129552,
      "step": 16070
    },
    {
      "epoch": 282.01769911504425,
      "grad_norm": 3.8616189357298936e-08,
      "learning_rate": 0.19552264753559603,
      "loss": 0.0,
      "num_input_tokens_seen": 9132240,
      "step": 16075
    },
    {
      "epoch": 282.1061946902655,
      "grad_norm": 5.18208551625321e-08,
      "learning_rate": 0.19546651732807774,
      "loss": 0.0,
      "num_input_tokens_seen": 9134768,
      "step": 16080
    },
    {
      "epoch": 282.1946902654867,
      "grad_norm": 1.1138522992837352e-08,
      "learning_rate": 0.19541038010905112,
      "loss": 0.0,
      "num_input_tokens_seen": 9137856,
      "step": 16085
    },
    {
      "epoch": 282.283185840708,
      "grad_norm": 5.721481599607614e-08,
      "learning_rate": 0.19535423588717324,
      "loss": 0.0,
      "num_input_tokens_seen": 9140448,
      "step": 16090
    },
    {
      "epoch": 282.37168141592923,
      "grad_norm": 4.434350486803851e-08,
      "learning_rate": 0.19529808467110224,
      "loss": 0.0,
      "num_input_tokens_seen": 9143200,
      "step": 16095
    },
    {
      "epoch": 282.46017699115043,
      "grad_norm": 3.773866907863521e-08,
      "learning_rate": 0.19524192646949734,
      "loss": 0.0,
      "num_input_tokens_seen": 9146016,
      "step": 16100
    },
    {
      "epoch": 282.5486725663717,
      "grad_norm": 4.351641180733168e-08,
      "learning_rate": 0.19518576129101878,
      "loss": 0.0,
      "num_input_tokens_seen": 9148576,
      "step": 16105
    },
    {
      "epoch": 282.6371681415929,
      "grad_norm": 3.5214704041663936e-08,
      "learning_rate": 0.19512958914432804,
      "loss": 0.0,
      "num_input_tokens_seen": 9151776,
      "step": 16110
    },
    {
      "epoch": 282.72566371681415,
      "grad_norm": 4.7451951701305006e-08,
      "learning_rate": 0.1950734100380875,
      "loss": 0.0,
      "num_input_tokens_seen": 9154896,
      "step": 16115
    },
    {
      "epoch": 282.8141592920354,
      "grad_norm": 7.380545241630898e-08,
      "learning_rate": 0.19501722398096066,
      "loss": 0.0,
      "num_input_tokens_seen": 9157744,
      "step": 16120
    },
    {
      "epoch": 282.9026548672566,
      "grad_norm": 6.765228732774631e-08,
      "learning_rate": 0.1949610309816122,
      "loss": 0.0,
      "num_input_tokens_seen": 9160704,
      "step": 16125
    },
    {
      "epoch": 282.9911504424779,
      "grad_norm": 1.9251709559853225e-08,
      "learning_rate": 0.1949048310487078,
      "loss": 0.0,
      "num_input_tokens_seen": 9164096,
      "step": 16130
    },
    {
      "epoch": 283.070796460177,
      "grad_norm": 6.415272224558066e-08,
      "learning_rate": 0.19484862419091406,
      "loss": 0.0,
      "num_input_tokens_seen": 9166512,
      "step": 16135
    },
    {
      "epoch": 283.1592920353982,
      "grad_norm": 4.660168428927136e-08,
      "learning_rate": 0.19479241041689893,
      "loss": 0.0,
      "num_input_tokens_seen": 9169456,
      "step": 16140
    },
    {
      "epoch": 283.24778761061947,
      "grad_norm": 4.0298804293570356e-08,
      "learning_rate": 0.19473618973533116,
      "loss": 0.0,
      "num_input_tokens_seen": 9172336,
      "step": 16145
    },
    {
      "epoch": 283.3362831858407,
      "grad_norm": 7.046748606853726e-08,
      "learning_rate": 0.19467996215488076,
      "loss": 0.0,
      "num_input_tokens_seen": 9174784,
      "step": 16150
    },
    {
      "epoch": 283.42477876106193,
      "grad_norm": 3.330206865825858e-08,
      "learning_rate": 0.1946237276842187,
      "loss": 0.0,
      "num_input_tokens_seen": 9177456,
      "step": 16155
    },
    {
      "epoch": 283.5132743362832,
      "grad_norm": 9.651753174466648e-08,
      "learning_rate": 0.19456748633201712,
      "loss": 0.0,
      "num_input_tokens_seen": 9179952,
      "step": 16160
    },
    {
      "epoch": 283.60176991150445,
      "grad_norm": 3.0168543219133426e-08,
      "learning_rate": 0.194511238106949,
      "loss": 0.0,
      "num_input_tokens_seen": 9182992,
      "step": 16165
    },
    {
      "epoch": 283.69026548672565,
      "grad_norm": 1.4637682177465194e-08,
      "learning_rate": 0.19445498301768863,
      "loss": 0.0,
      "num_input_tokens_seen": 9186336,
      "step": 16170
    },
    {
      "epoch": 283.7787610619469,
      "grad_norm": 4.6742393067233934e-08,
      "learning_rate": 0.19439872107291126,
      "loss": 0.0,
      "num_input_tokens_seen": 9189264,
      "step": 16175
    },
    {
      "epoch": 283.86725663716817,
      "grad_norm": 4.3774665670071045e-08,
      "learning_rate": 0.1943424522812931,
      "loss": 0.0,
      "num_input_tokens_seen": 9192480,
      "step": 16180
    },
    {
      "epoch": 283.95575221238937,
      "grad_norm": 5.2659473226412956e-08,
      "learning_rate": 0.19428617665151157,
      "loss": 0.0,
      "num_input_tokens_seen": 9195264,
      "step": 16185
    },
    {
      "epoch": 284.0353982300885,
      "grad_norm": 3.986876606631995e-08,
      "learning_rate": 0.19422989419224507,
      "loss": 0.0,
      "num_input_tokens_seen": 9197744,
      "step": 16190
    },
    {
      "epoch": 284.12389380530976,
      "grad_norm": 4.466793157575921e-08,
      "learning_rate": 0.19417360491217303,
      "loss": 0.0,
      "num_input_tokens_seen": 9200960,
      "step": 16195
    },
    {
      "epoch": 284.21238938053096,
      "grad_norm": 1.771601887412544e-08,
      "learning_rate": 0.19411730881997605,
      "loss": 0.0,
      "num_input_tokens_seen": 9204128,
      "step": 16200
    },
    {
      "epoch": 284.21238938053096,
      "eval_loss": 0.5731778740882874,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 26.703,
      "eval_steps_per_second": 13.886,
      "num_input_tokens_seen": 9204128,
      "step": 16200
    },
    {
      "epoch": 284.3008849557522,
      "grad_norm": 5.0735177836713774e-08,
      "learning_rate": 0.1940610059243356,
      "loss": 0.0,
      "num_input_tokens_seen": 9207184,
      "step": 16205
    },
    {
      "epoch": 284.3893805309734,
      "grad_norm": 3.561446959565728e-08,
      "learning_rate": 0.19400469623393435,
      "loss": 0.0,
      "num_input_tokens_seen": 9210048,
      "step": 16210
    },
    {
      "epoch": 284.4778761061947,
      "grad_norm": 4.1122635963120047e-08,
      "learning_rate": 0.1939483797574559,
      "loss": 0.0,
      "num_input_tokens_seen": 9212576,
      "step": 16215
    },
    {
      "epoch": 284.56637168141594,
      "grad_norm": 4.0147543955981746e-08,
      "learning_rate": 0.19389205650358504,
      "loss": 0.0,
      "num_input_tokens_seen": 9215312,
      "step": 16220
    },
    {
      "epoch": 284.65486725663715,
      "grad_norm": 8.266813011914564e-08,
      "learning_rate": 0.19383572648100747,
      "loss": 0.0,
      "num_input_tokens_seen": 9217968,
      "step": 16225
    },
    {
      "epoch": 284.7433628318584,
      "grad_norm": 2.298250656451728e-08,
      "learning_rate": 0.19377938969841,
      "loss": 0.0,
      "num_input_tokens_seen": 9220736,
      "step": 16230
    },
    {
      "epoch": 284.83185840707966,
      "grad_norm": 5.238841538357519e-08,
      "learning_rate": 0.1937230461644805,
      "loss": 0.0,
      "num_input_tokens_seen": 9223760,
      "step": 16235
    },
    {
      "epoch": 284.92035398230087,
      "grad_norm": 5.51482841615325e-08,
      "learning_rate": 0.19366669588790777,
      "loss": 0.0,
      "num_input_tokens_seen": 9226512,
      "step": 16240
    },
    {
      "epoch": 285.0,
      "grad_norm": 2.00334717703754e-08,
      "learning_rate": 0.19361033887738185,
      "loss": 0.0,
      "num_input_tokens_seen": 9229264,
      "step": 16245
    },
    {
      "epoch": 285.08849557522126,
      "grad_norm": 7.492477038795187e-08,
      "learning_rate": 0.19355397514159361,
      "loss": 0.0,
      "num_input_tokens_seen": 9232288,
      "step": 16250
    },
    {
      "epoch": 285.17699115044246,
      "grad_norm": 9.965557268287739e-08,
      "learning_rate": 0.19349760468923508,
      "loss": 0.0,
      "num_input_tokens_seen": 9235472,
      "step": 16255
    },
    {
      "epoch": 285.2654867256637,
      "grad_norm": 9.543219192664765e-08,
      "learning_rate": 0.19344122752899925,
      "loss": 0.0,
      "num_input_tokens_seen": 9238064,
      "step": 16260
    },
    {
      "epoch": 285.353982300885,
      "grad_norm": 5.619471110662744e-08,
      "learning_rate": 0.1933848436695802,
      "loss": 0.0,
      "num_input_tokens_seen": 9241216,
      "step": 16265
    },
    {
      "epoch": 285.4424778761062,
      "grad_norm": 4.9771390564501417e-08,
      "learning_rate": 0.1933284531196731,
      "loss": 0.0,
      "num_input_tokens_seen": 9243888,
      "step": 16270
    },
    {
      "epoch": 285.53097345132744,
      "grad_norm": 1.0878910217115845e-08,
      "learning_rate": 0.19327205588797403,
      "loss": 0.0,
      "num_input_tokens_seen": 9247008,
      "step": 16275
    },
    {
      "epoch": 285.6194690265487,
      "grad_norm": 3.8934945933988274e-08,
      "learning_rate": 0.19321565198318014,
      "loss": 0.0,
      "num_input_tokens_seen": 9250032,
      "step": 16280
    },
    {
      "epoch": 285.7079646017699,
      "grad_norm": 1.8918038691140282e-08,
      "learning_rate": 0.1931592414139896,
      "loss": 0.0,
      "num_input_tokens_seen": 9252544,
      "step": 16285
    },
    {
      "epoch": 285.79646017699116,
      "grad_norm": 1.8219845188127692e-08,
      "learning_rate": 0.19310282418910169,
      "loss": 0.0,
      "num_input_tokens_seen": 9255056,
      "step": 16290
    },
    {
      "epoch": 285.88495575221236,
      "grad_norm": 4.0929524658395167e-08,
      "learning_rate": 0.1930464003172166,
      "loss": 0.0,
      "num_input_tokens_seen": 9257888,
      "step": 16295
    },
    {
      "epoch": 285.9734513274336,
      "grad_norm": 3.653700986205877e-08,
      "learning_rate": 0.19298996980703567,
      "loss": 0.0,
      "num_input_tokens_seen": 9260832,
      "step": 16300
    },
    {
      "epoch": 286.05309734513276,
      "grad_norm": 3.16213402129506e-08,
      "learning_rate": 0.19293353266726113,
      "loss": 0.0,
      "num_input_tokens_seen": 9263472,
      "step": 16305
    },
    {
      "epoch": 286.14159292035396,
      "grad_norm": 6.699293919609772e-08,
      "learning_rate": 0.19287708890659633,
      "loss": 0.0,
      "num_input_tokens_seen": 9266176,
      "step": 16310
    },
    {
      "epoch": 286.2300884955752,
      "grad_norm": 8.353413960549005e-08,
      "learning_rate": 0.19282063853374556,
      "loss": 0.0,
      "num_input_tokens_seen": 9269056,
      "step": 16315
    },
    {
      "epoch": 286.3185840707965,
      "grad_norm": 3.03939486911986e-08,
      "learning_rate": 0.19276418155741423,
      "loss": 0.0,
      "num_input_tokens_seen": 9271840,
      "step": 16320
    },
    {
      "epoch": 286.4070796460177,
      "grad_norm": 8.881924173920197e-08,
      "learning_rate": 0.19270771798630867,
      "loss": 0.0,
      "num_input_tokens_seen": 9274848,
      "step": 16325
    },
    {
      "epoch": 286.49557522123894,
      "grad_norm": 2.9505274667940284e-08,
      "learning_rate": 0.1926512478291363,
      "loss": 0.0,
      "num_input_tokens_seen": 9277568,
      "step": 16330
    },
    {
      "epoch": 286.5840707964602,
      "grad_norm": 4.3635957069909637e-08,
      "learning_rate": 0.19259477109460557,
      "loss": 0.0,
      "num_input_tokens_seen": 9280544,
      "step": 16335
    },
    {
      "epoch": 286.6725663716814,
      "grad_norm": 4.4941984356228204e-08,
      "learning_rate": 0.19253828779142584,
      "loss": 0.0,
      "num_input_tokens_seen": 9283040,
      "step": 16340
    },
    {
      "epoch": 286.76106194690266,
      "grad_norm": 9.56378265470903e-09,
      "learning_rate": 0.19248179792830755,
      "loss": 0.0,
      "num_input_tokens_seen": 9286048,
      "step": 16345
    },
    {
      "epoch": 286.8495575221239,
      "grad_norm": 3.0640425308092745e-08,
      "learning_rate": 0.19242530151396217,
      "loss": 0.0,
      "num_input_tokens_seen": 9289392,
      "step": 16350
    },
    {
      "epoch": 286.9380530973451,
      "grad_norm": 7.545553160070995e-08,
      "learning_rate": 0.19236879855710215,
      "loss": 0.0,
      "num_input_tokens_seen": 9291952,
      "step": 16355
    },
    {
      "epoch": 287.01769911504425,
      "grad_norm": 6.334239088801041e-08,
      "learning_rate": 0.19231228906644096,
      "loss": 0.0,
      "num_input_tokens_seen": 9294184,
      "step": 16360
    },
    {
      "epoch": 287.1061946902655,
      "grad_norm": 7.876735708123306e-08,
      "learning_rate": 0.19225577305069302,
      "loss": 0.0,
      "num_input_tokens_seen": 9297176,
      "step": 16365
    },
    {
      "epoch": 287.1946902654867,
      "grad_norm": 4.1219756496957416e-08,
      "learning_rate": 0.1921992505185739,
      "loss": 0.0,
      "num_input_tokens_seen": 9299880,
      "step": 16370
    },
    {
      "epoch": 287.283185840708,
      "grad_norm": 5.33434718619219e-09,
      "learning_rate": 0.19214272147880004,
      "loss": 0.0,
      "num_input_tokens_seen": 9302536,
      "step": 16375
    },
    {
      "epoch": 287.37168141592923,
      "grad_norm": 4.432377309626645e-08,
      "learning_rate": 0.19208618594008892,
      "loss": 0.0,
      "num_input_tokens_seen": 9305528,
      "step": 16380
    },
    {
      "epoch": 287.46017699115043,
      "grad_norm": 3.71861439418808e-08,
      "learning_rate": 0.19202964391115904,
      "loss": 0.0,
      "num_input_tokens_seen": 9308488,
      "step": 16385
    },
    {
      "epoch": 287.5486725663717,
      "grad_norm": 6.090736803798791e-08,
      "learning_rate": 0.1919730954007299,
      "loss": 0.0,
      "num_input_tokens_seen": 9311608,
      "step": 16390
    },
    {
      "epoch": 287.6371681415929,
      "grad_norm": 2.255715791932289e-08,
      "learning_rate": 0.19191654041752199,
      "loss": 0.0,
      "num_input_tokens_seen": 9314376,
      "step": 16395
    },
    {
      "epoch": 287.72566371681415,
      "grad_norm": 2.6625951221603827e-08,
      "learning_rate": 0.19185997897025678,
      "loss": 0.0,
      "num_input_tokens_seen": 9317208,
      "step": 16400
    },
    {
      "epoch": 287.72566371681415,
      "eval_loss": 0.5829854607582092,
      "eval_runtime": 0.9206,
      "eval_samples_per_second": 27.157,
      "eval_steps_per_second": 14.122,
      "num_input_tokens_seen": 9317208,
      "step": 16400
    },
    {
      "epoch": 287.8141592920354,
      "grad_norm": 4.3530043569717236e-08,
      "learning_rate": 0.19180341106765672,
      "loss": 0.0,
      "num_input_tokens_seen": 9320456,
      "step": 16405
    },
    {
      "epoch": 287.9026548672566,
      "grad_norm": 1.5380811291265672e-08,
      "learning_rate": 0.19174683671844536,
      "loss": 0.0,
      "num_input_tokens_seen": 9323400,
      "step": 16410
    },
    {
      "epoch": 287.9911504424779,
      "grad_norm": 2.4110809349053852e-08,
      "learning_rate": 0.19169025593134717,
      "loss": 0.0,
      "num_input_tokens_seen": 9325992,
      "step": 16415
    },
    {
      "epoch": 288.070796460177,
      "grad_norm": 5.3318331083573867e-08,
      "learning_rate": 0.19163366871508764,
      "loss": 0.0,
      "num_input_tokens_seen": 9328424,
      "step": 16420
    },
    {
      "epoch": 288.1592920353982,
      "grad_norm": 2.6718598888919587e-08,
      "learning_rate": 0.19157707507839317,
      "loss": 0.0,
      "num_input_tokens_seen": 9331688,
      "step": 16425
    },
    {
      "epoch": 288.24778761061947,
      "grad_norm": 4.200950698418637e-08,
      "learning_rate": 0.19152047502999123,
      "loss": 0.0,
      "num_input_tokens_seen": 9334360,
      "step": 16430
    },
    {
      "epoch": 288.3362831858407,
      "grad_norm": 4.48682975218162e-08,
      "learning_rate": 0.19146386857861025,
      "loss": 0.0,
      "num_input_tokens_seen": 9337096,
      "step": 16435
    },
    {
      "epoch": 288.42477876106193,
      "grad_norm": 1.650240832873351e-08,
      "learning_rate": 0.19140725573297968,
      "loss": 0.0,
      "num_input_tokens_seen": 9340024,
      "step": 16440
    },
    {
      "epoch": 288.5132743362832,
      "grad_norm": 3.880736443306887e-08,
      "learning_rate": 0.19135063650182987,
      "loss": 0.0,
      "num_input_tokens_seen": 9343224,
      "step": 16445
    },
    {
      "epoch": 288.60176991150445,
      "grad_norm": 3.528749914494256e-08,
      "learning_rate": 0.19129401089389234,
      "loss": 0.0,
      "num_input_tokens_seen": 9345896,
      "step": 16450
    },
    {
      "epoch": 288.69026548672565,
      "grad_norm": 2.3178204244800327e-08,
      "learning_rate": 0.19123737891789938,
      "loss": 0.0,
      "num_input_tokens_seen": 9349160,
      "step": 16455
    },
    {
      "epoch": 288.7787610619469,
      "grad_norm": 4.84892694885275e-08,
      "learning_rate": 0.19118074058258439,
      "loss": 0.0,
      "num_input_tokens_seen": 9351720,
      "step": 16460
    },
    {
      "epoch": 288.86725663716817,
      "grad_norm": 6.294360588299241e-08,
      "learning_rate": 0.1911240958966816,
      "loss": 0.0,
      "num_input_tokens_seen": 9353976,
      "step": 16465
    },
    {
      "epoch": 288.95575221238937,
      "grad_norm": 1.9192977873672135e-08,
      "learning_rate": 0.19106744486892652,
      "loss": 0.0,
      "num_input_tokens_seen": 9356632,
      "step": 16470
    },
    {
      "epoch": 289.0353982300885,
      "grad_norm": 7.14319625672033e-08,
      "learning_rate": 0.1910107875080553,
      "loss": 0.0,
      "num_input_tokens_seen": 9359328,
      "step": 16475
    },
    {
      "epoch": 289.12389380530976,
      "grad_norm": 8.615845814574641e-08,
      "learning_rate": 0.19095412382280533,
      "loss": 0.0,
      "num_input_tokens_seen": 9362272,
      "step": 16480
    },
    {
      "epoch": 289.21238938053096,
      "grad_norm": 1.8070384300017395e-08,
      "learning_rate": 0.19089745382191473,
      "loss": 0.0,
      "num_input_tokens_seen": 9364960,
      "step": 16485
    },
    {
      "epoch": 289.3008849557522,
      "grad_norm": 4.7998472751942245e-08,
      "learning_rate": 0.19084077751412284,
      "loss": 0.0,
      "num_input_tokens_seen": 9368096,
      "step": 16490
    },
    {
      "epoch": 289.3893805309734,
      "grad_norm": 5.000238090246967e-08,
      "learning_rate": 0.19078409490816986,
      "loss": 0.0,
      "num_input_tokens_seen": 9371136,
      "step": 16495
    },
    {
      "epoch": 289.4778761061947,
      "grad_norm": 4.792290297928048e-08,
      "learning_rate": 0.19072740601279686,
      "loss": 0.0,
      "num_input_tokens_seen": 9373904,
      "step": 16500
    },
    {
      "epoch": 289.56637168141594,
      "grad_norm": 3.454715979955836e-08,
      "learning_rate": 0.19067071083674605,
      "loss": 0.0,
      "num_input_tokens_seen": 9376608,
      "step": 16505
    },
    {
      "epoch": 289.65486725663715,
      "grad_norm": 4.6097582639959e-08,
      "learning_rate": 0.19061400938876052,
      "loss": 0.0,
      "num_input_tokens_seen": 9379040,
      "step": 16510
    },
    {
      "epoch": 289.7433628318584,
      "grad_norm": 9.730552363862444e-08,
      "learning_rate": 0.1905573016775844,
      "loss": 0.0,
      "num_input_tokens_seen": 9381904,
      "step": 16515
    },
    {
      "epoch": 289.83185840707966,
      "grad_norm": 3.659082281615156e-08,
      "learning_rate": 0.19050058771196263,
      "loss": 0.0,
      "num_input_tokens_seen": 9384944,
      "step": 16520
    },
    {
      "epoch": 289.92035398230087,
      "grad_norm": 4.0630474984482134e-08,
      "learning_rate": 0.19044386750064132,
      "loss": 0.0,
      "num_input_tokens_seen": 9388400,
      "step": 16525
    },
    {
      "epoch": 290.0,
      "grad_norm": 4.197710978814939e-08,
      "learning_rate": 0.19038714105236737,
      "loss": 0.0,
      "num_input_tokens_seen": 9390648,
      "step": 16530
    },
    {
      "epoch": 290.08849557522126,
      "grad_norm": 6.408247088529606e-08,
      "learning_rate": 0.19033040837588874,
      "loss": 0.0,
      "num_input_tokens_seen": 9393784,
      "step": 16535
    },
    {
      "epoch": 290.17699115044246,
      "grad_norm": 3.0754986113379346e-08,
      "learning_rate": 0.1902736694799543,
      "loss": 0.0,
      "num_input_tokens_seen": 9396680,
      "step": 16540
    },
    {
      "epoch": 290.2654867256637,
      "grad_norm": 3.7041477440880044e-08,
      "learning_rate": 0.19021692437331392,
      "loss": 0.0,
      "num_input_tokens_seen": 9399784,
      "step": 16545
    },
    {
      "epoch": 290.353982300885,
      "grad_norm": 6.338833458130466e-08,
      "learning_rate": 0.1901601730647184,
      "loss": 0.0,
      "num_input_tokens_seen": 9402632,
      "step": 16550
    },
    {
      "epoch": 290.4424778761062,
      "grad_norm": 4.488640925615073e-08,
      "learning_rate": 0.19010341556291954,
      "loss": 0.0,
      "num_input_tokens_seen": 9405368,
      "step": 16555
    },
    {
      "epoch": 290.53097345132744,
      "grad_norm": 3.6203488207320333e-08,
      "learning_rate": 0.19004665187667,
      "loss": 0.0,
      "num_input_tokens_seen": 9408456,
      "step": 16560
    },
    {
      "epoch": 290.6194690265487,
      "grad_norm": 4.4784055575064485e-08,
      "learning_rate": 0.1899898820147235,
      "loss": 0.0,
      "num_input_tokens_seen": 9410920,
      "step": 16565
    },
    {
      "epoch": 290.7079646017699,
      "grad_norm": 1.9964266684269205e-08,
      "learning_rate": 0.18993310598583465,
      "loss": 0.0,
      "num_input_tokens_seen": 9414024,
      "step": 16570
    },
    {
      "epoch": 290.79646017699116,
      "grad_norm": 4.2441460124109653e-08,
      "learning_rate": 0.18987632379875904,
      "loss": 0.0,
      "num_input_tokens_seen": 9417144,
      "step": 16575
    },
    {
      "epoch": 290.88495575221236,
      "grad_norm": 7.196140927590022e-08,
      "learning_rate": 0.18981953546225314,
      "loss": 0.0,
      "num_input_tokens_seen": 9420024,
      "step": 16580
    },
    {
      "epoch": 290.9734513274336,
      "grad_norm": 3.026540440487224e-08,
      "learning_rate": 0.18976274098507445,
      "loss": 0.0,
      "num_input_tokens_seen": 9422488,
      "step": 16585
    },
    {
      "epoch": 291.05309734513276,
      "grad_norm": 3.0443668919133415e-08,
      "learning_rate": 0.18970594037598146,
      "loss": 0.0,
      "num_input_tokens_seen": 9424856,
      "step": 16590
    },
    {
      "epoch": 291.14159292035396,
      "grad_norm": 2.210692606752218e-08,
      "learning_rate": 0.1896491336437335,
      "loss": 0.0,
      "num_input_tokens_seen": 9427880,
      "step": 16595
    },
    {
      "epoch": 291.2300884955752,
      "grad_norm": 1.7064269997035808e-08,
      "learning_rate": 0.18959232079709085,
      "loss": 0.0,
      "num_input_tokens_seen": 9431208,
      "step": 16600
    },
    {
      "epoch": 291.2300884955752,
      "eval_loss": 0.589658260345459,
      "eval_runtime": 0.9454,
      "eval_samples_per_second": 26.444,
      "eval_steps_per_second": 13.751,
      "num_input_tokens_seen": 9431208,
      "step": 16600
    },
    {
      "epoch": 291.3185840707965,
      "grad_norm": 5.584525908375326e-08,
      "learning_rate": 0.18953550184481477,
      "loss": 0.0,
      "num_input_tokens_seen": 9433976,
      "step": 16605
    },
    {
      "epoch": 291.4070796460177,
      "grad_norm": 2.7258566959176278e-08,
      "learning_rate": 0.18947867679566752,
      "loss": 0.0,
      "num_input_tokens_seen": 9436232,
      "step": 16610
    },
    {
      "epoch": 291.49557522123894,
      "grad_norm": 4.647998963491773e-08,
      "learning_rate": 0.18942184565841216,
      "loss": 0.0,
      "num_input_tokens_seen": 9439112,
      "step": 16615
    },
    {
      "epoch": 291.5840707964602,
      "grad_norm": 4.213920945517202e-08,
      "learning_rate": 0.18936500844181278,
      "loss": 0.0,
      "num_input_tokens_seen": 9441736,
      "step": 16620
    },
    {
      "epoch": 291.6725663716814,
      "grad_norm": 4.8096300275801696e-08,
      "learning_rate": 0.18930816515463436,
      "loss": 0.0,
      "num_input_tokens_seen": 9444504,
      "step": 16625
    },
    {
      "epoch": 291.76106194690266,
      "grad_norm": 3.193757081021431e-08,
      "learning_rate": 0.18925131580564297,
      "loss": 0.0,
      "num_input_tokens_seen": 9447496,
      "step": 16630
    },
    {
      "epoch": 291.8495575221239,
      "grad_norm": 2.579761115839574e-08,
      "learning_rate": 0.1891944604036054,
      "loss": 0.0,
      "num_input_tokens_seen": 9450648,
      "step": 16635
    },
    {
      "epoch": 291.9380530973451,
      "grad_norm": 5.6573117745983836e-08,
      "learning_rate": 0.1891375989572895,
      "loss": 0.0,
      "num_input_tokens_seen": 9453880,
      "step": 16640
    },
    {
      "epoch": 292.01769911504425,
      "grad_norm": 3.78931339639621e-08,
      "learning_rate": 0.18908073147546398,
      "loss": 0.0,
      "num_input_tokens_seen": 9455992,
      "step": 16645
    },
    {
      "epoch": 292.1061946902655,
      "grad_norm": 2.329855952609705e-08,
      "learning_rate": 0.18902385796689858,
      "loss": 0.0,
      "num_input_tokens_seen": 9458872,
      "step": 16650
    },
    {
      "epoch": 292.1946902654867,
      "grad_norm": 2.74485287832249e-08,
      "learning_rate": 0.18896697844036384,
      "loss": 0.0,
      "num_input_tokens_seen": 9461512,
      "step": 16655
    },
    {
      "epoch": 292.283185840708,
      "grad_norm": 6.326374091258913e-08,
      "learning_rate": 0.18891009290463137,
      "loss": 0.0,
      "num_input_tokens_seen": 9464216,
      "step": 16660
    },
    {
      "epoch": 292.37168141592923,
      "grad_norm": 2.018120959235148e-08,
      "learning_rate": 0.18885320136847353,
      "loss": 0.0,
      "num_input_tokens_seen": 9467160,
      "step": 16665
    },
    {
      "epoch": 292.46017699115043,
      "grad_norm": 6.190057177946073e-08,
      "learning_rate": 0.1887963038406639,
      "loss": 0.0,
      "num_input_tokens_seen": 9469880,
      "step": 16670
    },
    {
      "epoch": 292.5486725663717,
      "grad_norm": 5.467841646122906e-08,
      "learning_rate": 0.18873940032997658,
      "loss": 0.0,
      "num_input_tokens_seen": 9472392,
      "step": 16675
    },
    {
      "epoch": 292.6371681415929,
      "grad_norm": 3.2032961172490104e-08,
      "learning_rate": 0.18868249084518693,
      "loss": 0.0,
      "num_input_tokens_seen": 9475096,
      "step": 16680
    },
    {
      "epoch": 292.72566371681415,
      "grad_norm": 2.7805063140817765e-08,
      "learning_rate": 0.18862557539507102,
      "loss": 0.0,
      "num_input_tokens_seen": 9478024,
      "step": 16685
    },
    {
      "epoch": 292.8141592920354,
      "grad_norm": 4.291906208209184e-08,
      "learning_rate": 0.18856865398840605,
      "loss": 0.0,
      "num_input_tokens_seen": 9481416,
      "step": 16690
    },
    {
      "epoch": 292.9026548672566,
      "grad_norm": 2.9684994018452926e-08,
      "learning_rate": 0.18851172663396995,
      "loss": 0.0,
      "num_input_tokens_seen": 9484360,
      "step": 16695
    },
    {
      "epoch": 292.9911504424779,
      "grad_norm": 2.3631660184264547e-08,
      "learning_rate": 0.1884547933405416,
      "loss": 0.0,
      "num_input_tokens_seen": 9487320,
      "step": 16700
    },
    {
      "epoch": 293.070796460177,
      "grad_norm": 4.2839293001861734e-08,
      "learning_rate": 0.1883978541169009,
      "loss": 0.0,
      "num_input_tokens_seen": 9489576,
      "step": 16705
    },
    {
      "epoch": 293.1592920353982,
      "grad_norm": 6.895344029089756e-08,
      "learning_rate": 0.18834090897182854,
      "loss": 0.0,
      "num_input_tokens_seen": 9492840,
      "step": 16710
    },
    {
      "epoch": 293.24778761061947,
      "grad_norm": 6.088077952881576e-08,
      "learning_rate": 0.1882839579141062,
      "loss": 0.0,
      "num_input_tokens_seen": 9495496,
      "step": 16715
    },
    {
      "epoch": 293.3362831858407,
      "grad_norm": 5.0252399574901574e-08,
      "learning_rate": 0.18822700095251646,
      "loss": 0.0,
      "num_input_tokens_seen": 9498504,
      "step": 16720
    },
    {
      "epoch": 293.42477876106193,
      "grad_norm": 3.512106871994547e-08,
      "learning_rate": 0.18817003809584273,
      "loss": 0.0,
      "num_input_tokens_seen": 9501544,
      "step": 16725
    },
    {
      "epoch": 293.5132743362832,
      "grad_norm": 4.2326153248950504e-08,
      "learning_rate": 0.1881130693528695,
      "loss": 0.0,
      "num_input_tokens_seen": 9504392,
      "step": 16730
    },
    {
      "epoch": 293.60176991150445,
      "grad_norm": 5.897478771998976e-08,
      "learning_rate": 0.18805609473238197,
      "loss": 0.0,
      "num_input_tokens_seen": 9507256,
      "step": 16735
    },
    {
      "epoch": 293.69026548672565,
      "grad_norm": 1.4322976582548108e-08,
      "learning_rate": 0.18799911424316643,
      "loss": 0.0,
      "num_input_tokens_seen": 9510008,
      "step": 16740
    },
    {
      "epoch": 293.7787610619469,
      "grad_norm": 4.178708579161139e-08,
      "learning_rate": 0.18794212789400994,
      "loss": 0.0,
      "num_input_tokens_seen": 9513064,
      "step": 16745
    },
    {
      "epoch": 293.86725663716817,
      "grad_norm": 3.618806587724066e-08,
      "learning_rate": 0.18788513569370052,
      "loss": 0.0,
      "num_input_tokens_seen": 9516232,
      "step": 16750
    },
    {
      "epoch": 293.95575221238937,
      "grad_norm": 3.3374025321109e-08,
      "learning_rate": 0.1878281376510271,
      "loss": 0.0,
      "num_input_tokens_seen": 9518824,
      "step": 16755
    },
    {
      "epoch": 294.0353982300885,
      "grad_norm": 3.581173757538636e-08,
      "learning_rate": 0.18777113377477941,
      "loss": 0.0,
      "num_input_tokens_seen": 9521096,
      "step": 16760
    },
    {
      "epoch": 294.12389380530976,
      "grad_norm": 2.4314308788575545e-08,
      "learning_rate": 0.1877141240737483,
      "loss": 0.0,
      "num_input_tokens_seen": 9524568,
      "step": 16765
    },
    {
      "epoch": 294.21238938053096,
      "grad_norm": 7.729803996880946e-08,
      "learning_rate": 0.18765710855672527,
      "loss": 0.0,
      "num_input_tokens_seen": 9527400,
      "step": 16770
    },
    {
      "epoch": 294.3008849557522,
      "grad_norm": 4.1474628176274564e-08,
      "learning_rate": 0.18760008723250288,
      "loss": 0.0,
      "num_input_tokens_seen": 9530360,
      "step": 16775
    },
    {
      "epoch": 294.3893805309734,
      "grad_norm": 2.6699408905983546e-08,
      "learning_rate": 0.18754306010987457,
      "loss": 0.0,
      "num_input_tokens_seen": 9533032,
      "step": 16780
    },
    {
      "epoch": 294.4778761061947,
      "grad_norm": 3.722354335877753e-08,
      "learning_rate": 0.18748602719763457,
      "loss": 0.0,
      "num_input_tokens_seen": 9535704,
      "step": 16785
    },
    {
      "epoch": 294.56637168141594,
      "grad_norm": 1.1901504137767915e-07,
      "learning_rate": 0.18742898850457804,
      "loss": 0.0,
      "num_input_tokens_seen": 9538488,
      "step": 16790
    },
    {
      "epoch": 294.65486725663715,
      "grad_norm": 2.572314627968808e-08,
      "learning_rate": 0.1873719440395012,
      "loss": 0.0,
      "num_input_tokens_seen": 9541448,
      "step": 16795
    },
    {
      "epoch": 294.7433628318584,
      "grad_norm": 3.163159334462762e-08,
      "learning_rate": 0.1873148938112009,
      "loss": 0.0,
      "num_input_tokens_seen": 9544328,
      "step": 16800
    },
    {
      "epoch": 294.7433628318584,
      "eval_loss": 0.5792176723480225,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 26.719,
      "eval_steps_per_second": 13.894,
      "num_input_tokens_seen": 9544328,
      "step": 16800
    },
    {
      "epoch": 294.83185840707966,
      "grad_norm": 2.961683165381146e-08,
      "learning_rate": 0.18725783782847508,
      "loss": 0.0,
      "num_input_tokens_seen": 9547144,
      "step": 16805
    },
    {
      "epoch": 294.92035398230087,
      "grad_norm": 3.014040217408365e-08,
      "learning_rate": 0.1872007761001224,
      "loss": 0.0,
      "num_input_tokens_seen": 9549912,
      "step": 16810
    },
    {
      "epoch": 295.0,
      "grad_norm": 1.2850875918957172e-08,
      "learning_rate": 0.1871437086349426,
      "loss": 0.0,
      "num_input_tokens_seen": 9552440,
      "step": 16815
    },
    {
      "epoch": 295.08849557522126,
      "grad_norm": 5.006890191339153e-08,
      "learning_rate": 0.18708663544173615,
      "loss": 0.0,
      "num_input_tokens_seen": 9554984,
      "step": 16820
    },
    {
      "epoch": 295.17699115044246,
      "grad_norm": 3.834453821127681e-08,
      "learning_rate": 0.18702955652930442,
      "loss": 0.0,
      "num_input_tokens_seen": 9557992,
      "step": 16825
    },
    {
      "epoch": 295.2654867256637,
      "grad_norm": 2.581457714256885e-08,
      "learning_rate": 0.18697247190644972,
      "loss": 0.0,
      "num_input_tokens_seen": 9560760,
      "step": 16830
    },
    {
      "epoch": 295.353982300885,
      "grad_norm": 1.1647167674766479e-08,
      "learning_rate": 0.18691538158197527,
      "loss": 0.0,
      "num_input_tokens_seen": 9563624,
      "step": 16835
    },
    {
      "epoch": 295.4424778761062,
      "grad_norm": 6.591552903500997e-08,
      "learning_rate": 0.1868582855646851,
      "loss": 0.0,
      "num_input_tokens_seen": 9566248,
      "step": 16840
    },
    {
      "epoch": 295.53097345132744,
      "grad_norm": 5.6345562882142985e-08,
      "learning_rate": 0.18680118386338404,
      "loss": 0.0,
      "num_input_tokens_seen": 9568904,
      "step": 16845
    },
    {
      "epoch": 295.6194690265487,
      "grad_norm": 1.3678167931630014e-08,
      "learning_rate": 0.18674407648687794,
      "loss": 0.0,
      "num_input_tokens_seen": 9571656,
      "step": 16850
    },
    {
      "epoch": 295.7079646017699,
      "grad_norm": 4.091856098398239e-08,
      "learning_rate": 0.1866869634439736,
      "loss": 0.0,
      "num_input_tokens_seen": 9574584,
      "step": 16855
    },
    {
      "epoch": 295.79646017699116,
      "grad_norm": 3.691071626121811e-08,
      "learning_rate": 0.18662984474347838,
      "loss": 0.0,
      "num_input_tokens_seen": 9577592,
      "step": 16860
    },
    {
      "epoch": 295.88495575221236,
      "grad_norm": 3.5944648146823965e-08,
      "learning_rate": 0.1865727203942008,
      "loss": 0.0,
      "num_input_tokens_seen": 9580920,
      "step": 16865
    },
    {
      "epoch": 295.9734513274336,
      "grad_norm": 2.9852337490865466e-08,
      "learning_rate": 0.1865155904049501,
      "loss": 0.0,
      "num_input_tokens_seen": 9583784,
      "step": 16870
    },
    {
      "epoch": 296.05309734513276,
      "grad_norm": 4.145040932712618e-08,
      "learning_rate": 0.1864584547845365,
      "loss": 0.0,
      "num_input_tokens_seen": 9586176,
      "step": 16875
    },
    {
      "epoch": 296.14159292035396,
      "grad_norm": 3.149389371515099e-08,
      "learning_rate": 0.186401313541771,
      "loss": 0.0,
      "num_input_tokens_seen": 9588896,
      "step": 16880
    },
    {
      "epoch": 296.2300884955752,
      "grad_norm": 2.6964842803067768e-08,
      "learning_rate": 0.18634416668546552,
      "loss": 0.0,
      "num_input_tokens_seen": 9591328,
      "step": 16885
    },
    {
      "epoch": 296.3185840707965,
      "grad_norm": 2.046340874528596e-08,
      "learning_rate": 0.1862870142244328,
      "loss": 0.0,
      "num_input_tokens_seen": 9593808,
      "step": 16890
    },
    {
      "epoch": 296.4070796460177,
      "grad_norm": 5.55764465559605e-08,
      "learning_rate": 0.1862298561674865,
      "loss": 0.0,
      "num_input_tokens_seen": 9596912,
      "step": 16895
    },
    {
      "epoch": 296.49557522123894,
      "grad_norm": 3.77847406696219e-08,
      "learning_rate": 0.18617269252344104,
      "loss": 0.0,
      "num_input_tokens_seen": 9600096,
      "step": 16900
    },
    {
      "epoch": 296.5840707964602,
      "grad_norm": 5.608920972122178e-08,
      "learning_rate": 0.18611552330111186,
      "loss": 0.0,
      "num_input_tokens_seen": 9602528,
      "step": 16905
    },
    {
      "epoch": 296.6725663716814,
      "grad_norm": 1.719443787351338e-08,
      "learning_rate": 0.18605834850931507,
      "loss": 0.0,
      "num_input_tokens_seen": 9605296,
      "step": 16910
    },
    {
      "epoch": 296.76106194690266,
      "grad_norm": 2.9844613891327754e-08,
      "learning_rate": 0.18600116815686787,
      "loss": 0.0,
      "num_input_tokens_seen": 9608256,
      "step": 16915
    },
    {
      "epoch": 296.8495575221239,
      "grad_norm": 5.350101162093779e-08,
      "learning_rate": 0.1859439822525881,
      "loss": 0.0,
      "num_input_tokens_seen": 9611536,
      "step": 16920
    },
    {
      "epoch": 296.9380530973451,
      "grad_norm": 8.16066716424757e-08,
      "learning_rate": 0.18588679080529455,
      "loss": 0.0,
      "num_input_tokens_seen": 9614832,
      "step": 16925
    },
    {
      "epoch": 297.01769911504425,
      "grad_norm": 4.64931666499524e-08,
      "learning_rate": 0.1858295938238069,
      "loss": 0.0,
      "num_input_tokens_seen": 9617344,
      "step": 16930
    },
    {
      "epoch": 297.1061946902655,
      "grad_norm": 6.300354726818114e-08,
      "learning_rate": 0.18577239131694562,
      "loss": 0.0,
      "num_input_tokens_seen": 9620544,
      "step": 16935
    },
    {
      "epoch": 297.1946902654867,
      "grad_norm": 4.9674369506647054e-08,
      "learning_rate": 0.18571518329353204,
      "loss": 0.0,
      "num_input_tokens_seen": 9623280,
      "step": 16940
    },
    {
      "epoch": 297.283185840708,
      "grad_norm": 4.554261678890725e-08,
      "learning_rate": 0.18565796976238838,
      "loss": 0.0,
      "num_input_tokens_seen": 9626464,
      "step": 16945
    },
    {
      "epoch": 297.37168141592923,
      "grad_norm": 5.000198655125132e-08,
      "learning_rate": 0.18560075073233764,
      "loss": 0.0,
      "num_input_tokens_seen": 9629072,
      "step": 16950
    },
    {
      "epoch": 297.46017699115043,
      "grad_norm": 7.067796303772411e-08,
      "learning_rate": 0.18554352621220377,
      "loss": 0.0,
      "num_input_tokens_seen": 9631872,
      "step": 16955
    },
    {
      "epoch": 297.5486725663717,
      "grad_norm": 3.767374323615513e-08,
      "learning_rate": 0.18548629621081153,
      "loss": 0.0,
      "num_input_tokens_seen": 9634640,
      "step": 16960
    },
    {
      "epoch": 297.6371681415929,
      "grad_norm": 1.493044443634517e-08,
      "learning_rate": 0.18542906073698645,
      "loss": 0.0,
      "num_input_tokens_seen": 9637360,
      "step": 16965
    },
    {
      "epoch": 297.72566371681415,
      "grad_norm": 8.849911381503262e-08,
      "learning_rate": 0.18537181979955494,
      "loss": 0.0,
      "num_input_tokens_seen": 9640224,
      "step": 16970
    },
    {
      "epoch": 297.8141592920354,
      "grad_norm": 2.0982536597102808e-08,
      "learning_rate": 0.18531457340734434,
      "loss": 0.0,
      "num_input_tokens_seen": 9642800,
      "step": 16975
    },
    {
      "epoch": 297.9026548672566,
      "grad_norm": 3.7881065395595215e-08,
      "learning_rate": 0.1852573215691827,
      "loss": 0.0,
      "num_input_tokens_seen": 9645600,
      "step": 16980
    },
    {
      "epoch": 297.9911504424779,
      "grad_norm": 1.8675574864346345e-08,
      "learning_rate": 0.18520006429389904,
      "loss": 0.0,
      "num_input_tokens_seen": 9648912,
      "step": 16985
    },
    {
      "epoch": 298.070796460177,
      "grad_norm": 4.1193338518041855e-08,
      "learning_rate": 0.1851428015903231,
      "loss": 0.0,
      "num_input_tokens_seen": 9651656,
      "step": 16990
    },
    {
      "epoch": 298.1592920353982,
      "grad_norm": 1.529139659339762e-08,
      "learning_rate": 0.1850855334672855,
      "loss": 0.0,
      "num_input_tokens_seen": 9654552,
      "step": 16995
    },
    {
      "epoch": 298.24778761061947,
      "grad_norm": 2.4684826627208167e-08,
      "learning_rate": 0.1850282599336178,
      "loss": 0.0,
      "num_input_tokens_seen": 9657432,
      "step": 17000
    },
    {
      "epoch": 298.24778761061947,
      "eval_loss": 0.5745419859886169,
      "eval_runtime": 0.9411,
      "eval_samples_per_second": 26.565,
      "eval_steps_per_second": 13.814,
      "num_input_tokens_seen": 9657432,
      "step": 17000
    },
    {
      "epoch": 298.3362831858407,
      "grad_norm": 2.0465824590587545e-08,
      "learning_rate": 0.18497098099815215,
      "loss": 0.0,
      "num_input_tokens_seen": 9660344,
      "step": 17005
    },
    {
      "epoch": 298.42477876106193,
      "grad_norm": 3.763595302075373e-08,
      "learning_rate": 0.18491369666972174,
      "loss": 0.0,
      "num_input_tokens_seen": 9662776,
      "step": 17010
    },
    {
      "epoch": 298.5132743362832,
      "grad_norm": 5.158343441280522e-08,
      "learning_rate": 0.1848564069571606,
      "loss": 0.0,
      "num_input_tokens_seen": 9665656,
      "step": 17015
    },
    {
      "epoch": 298.60176991150445,
      "grad_norm": 2.339233340364899e-08,
      "learning_rate": 0.18479911186930348,
      "loss": 0.0,
      "num_input_tokens_seen": 9668296,
      "step": 17020
    },
    {
      "epoch": 298.69026548672565,
      "grad_norm": 2.935281351312824e-08,
      "learning_rate": 0.18474181141498597,
      "loss": 0.0,
      "num_input_tokens_seen": 9671640,
      "step": 17025
    },
    {
      "epoch": 298.7787610619469,
      "grad_norm": 5.3919677611702355e-08,
      "learning_rate": 0.18468450560304453,
      "loss": 0.0,
      "num_input_tokens_seen": 9674872,
      "step": 17030
    },
    {
      "epoch": 298.86725663716817,
      "grad_norm": 4.663074903987763e-08,
      "learning_rate": 0.1846271944423165,
      "loss": 0.0,
      "num_input_tokens_seen": 9677672,
      "step": 17035
    },
    {
      "epoch": 298.95575221238937,
      "grad_norm": 4.397637098918494e-08,
      "learning_rate": 0.18456987794163993,
      "loss": 0.0,
      "num_input_tokens_seen": 9680264,
      "step": 17040
    },
    {
      "epoch": 299.0353982300885,
      "grad_norm": 1.6805094205096793e-08,
      "learning_rate": 0.18451255610985373,
      "loss": 0.0,
      "num_input_tokens_seen": 9682552,
      "step": 17045
    },
    {
      "epoch": 299.12389380530976,
      "grad_norm": 5.1185374161377695e-08,
      "learning_rate": 0.18445522895579766,
      "loss": 0.0,
      "num_input_tokens_seen": 9685464,
      "step": 17050
    },
    {
      "epoch": 299.21238938053096,
      "grad_norm": 3.78581255233712e-08,
      "learning_rate": 0.1843978964883123,
      "loss": 0.0,
      "num_input_tokens_seen": 9688632,
      "step": 17055
    },
    {
      "epoch": 299.3008849557522,
      "grad_norm": 3.35241985283119e-08,
      "learning_rate": 0.18434055871623906,
      "loss": 0.0,
      "num_input_tokens_seen": 9691288,
      "step": 17060
    },
    {
      "epoch": 299.3893805309734,
      "grad_norm": 6.846974542895623e-08,
      "learning_rate": 0.18428321564842007,
      "loss": 0.0,
      "num_input_tokens_seen": 9694168,
      "step": 17065
    },
    {
      "epoch": 299.4778761061947,
      "grad_norm": 3.927053171537409e-08,
      "learning_rate": 0.18422586729369841,
      "loss": 0.0,
      "num_input_tokens_seen": 9697080,
      "step": 17070
    },
    {
      "epoch": 299.56637168141594,
      "grad_norm": 4.05815789861208e-08,
      "learning_rate": 0.1841685136609179,
      "loss": 0.0,
      "num_input_tokens_seen": 9699832,
      "step": 17075
    },
    {
      "epoch": 299.65486725663715,
      "grad_norm": 1.8075443364296007e-08,
      "learning_rate": 0.18411115475892326,
      "loss": 0.0,
      "num_input_tokens_seen": 9702712,
      "step": 17080
    },
    {
      "epoch": 299.7433628318584,
      "grad_norm": 8.423398867307696e-08,
      "learning_rate": 0.18405379059655982,
      "loss": 0.0,
      "num_input_tokens_seen": 9705416,
      "step": 17085
    },
    {
      "epoch": 299.83185840707966,
      "grad_norm": 2.2795122234242626e-08,
      "learning_rate": 0.1839964211826739,
      "loss": 0.0,
      "num_input_tokens_seen": 9708824,
      "step": 17090
    },
    {
      "epoch": 299.92035398230087,
      "grad_norm": 3.6615862342159744e-08,
      "learning_rate": 0.18393904652611265,
      "loss": 0.0,
      "num_input_tokens_seen": 9711688,
      "step": 17095
    },
    {
      "epoch": 300.0,
      "grad_norm": 1.481422273741373e-08,
      "learning_rate": 0.18388166663572392,
      "loss": 0.0,
      "num_input_tokens_seen": 9713848,
      "step": 17100
    },
    {
      "epoch": 300.08849557522126,
      "grad_norm": 2.0962501068311212e-08,
      "learning_rate": 0.18382428152035643,
      "loss": 0.0,
      "num_input_tokens_seen": 9716568,
      "step": 17105
    },
    {
      "epoch": 300.17699115044246,
      "grad_norm": 4.435221256926525e-08,
      "learning_rate": 0.1837668911888596,
      "loss": 0.0,
      "num_input_tokens_seen": 9719288,
      "step": 17110
    },
    {
      "epoch": 300.2654867256637,
      "grad_norm": 3.290381300757872e-08,
      "learning_rate": 0.18370949565008388,
      "loss": 0.0,
      "num_input_tokens_seen": 9722344,
      "step": 17115
    },
    {
      "epoch": 300.353982300885,
      "grad_norm": 1.6830760785069288e-08,
      "learning_rate": 0.1836520949128803,
      "loss": 0.0,
      "num_input_tokens_seen": 9725496,
      "step": 17120
    },
    {
      "epoch": 300.4424778761062,
      "grad_norm": 2.8398792650818905e-08,
      "learning_rate": 0.18359468898610076,
      "loss": 0.0,
      "num_input_tokens_seen": 9728456,
      "step": 17125
    },
    {
      "epoch": 300.53097345132744,
      "grad_norm": 3.014494609487883e-08,
      "learning_rate": 0.18353727787859797,
      "loss": 0.0,
      "num_input_tokens_seen": 9731016,
      "step": 17130
    },
    {
      "epoch": 300.6194690265487,
      "grad_norm": 1.9530506989440255e-08,
      "learning_rate": 0.18347986159922552,
      "loss": 0.0,
      "num_input_tokens_seen": 9734408,
      "step": 17135
    },
    {
      "epoch": 300.7079646017699,
      "grad_norm": 4.811677456473262e-08,
      "learning_rate": 0.1834224401568377,
      "loss": 0.0,
      "num_input_tokens_seen": 9737176,
      "step": 17140
    },
    {
      "epoch": 300.79646017699116,
      "grad_norm": 5.697454596997886e-08,
      "learning_rate": 0.1833650135602896,
      "loss": 0.0,
      "num_input_tokens_seen": 9739896,
      "step": 17145
    },
    {
      "epoch": 300.88495575221236,
      "grad_norm": 2.2096170226859613e-08,
      "learning_rate": 0.18330758181843707,
      "loss": 0.0,
      "num_input_tokens_seen": 9742952,
      "step": 17150
    },
    {
      "epoch": 300.9734513274336,
      "grad_norm": 1.1858986681545503e-07,
      "learning_rate": 0.18325014494013686,
      "loss": 0.0,
      "num_input_tokens_seen": 9745896,
      "step": 17155
    },
    {
      "epoch": 301.05309734513276,
      "grad_norm": 2.6347242609858768e-08,
      "learning_rate": 0.18319270293424647,
      "loss": 0.0,
      "num_input_tokens_seen": 9748024,
      "step": 17160
    },
    {
      "epoch": 301.14159292035396,
      "grad_norm": 3.522952951584557e-08,
      "learning_rate": 0.18313525580962417,
      "loss": 0.0,
      "num_input_tokens_seen": 9750664,
      "step": 17165
    },
    {
      "epoch": 301.2300884955752,
      "grad_norm": 3.230042011637124e-08,
      "learning_rate": 0.18307780357512896,
      "loss": 0.0,
      "num_input_tokens_seen": 9753512,
      "step": 17170
    },
    {
      "epoch": 301.3185840707965,
      "grad_norm": 3.341066800999215e-08,
      "learning_rate": 0.1830203462396208,
      "loss": 0.0,
      "num_input_tokens_seen": 9756664,
      "step": 17175
    },
    {
      "epoch": 301.4070796460177,
      "grad_norm": 2.9485537567097708e-08,
      "learning_rate": 0.18296288381196033,
      "loss": 0.0,
      "num_input_tokens_seen": 9759736,
      "step": 17180
    },
    {
      "epoch": 301.49557522123894,
      "grad_norm": 2.7112134759477158e-08,
      "learning_rate": 0.1829054163010089,
      "loss": 0.0,
      "num_input_tokens_seen": 9762728,
      "step": 17185
    },
    {
      "epoch": 301.5840707964602,
      "grad_norm": 3.825106276167389e-08,
      "learning_rate": 0.18284794371562874,
      "loss": 0.0,
      "num_input_tokens_seen": 9765288,
      "step": 17190
    },
    {
      "epoch": 301.6725663716814,
      "grad_norm": 1.2838841101370235e-08,
      "learning_rate": 0.18279046606468288,
      "loss": 0.0,
      "num_input_tokens_seen": 9768312,
      "step": 17195
    },
    {
      "epoch": 301.76106194690266,
      "grad_norm": 2.753484551476504e-08,
      "learning_rate": 0.1827329833570351,
      "loss": 0.0,
      "num_input_tokens_seen": 9770824,
      "step": 17200
    },
    {
      "epoch": 301.76106194690266,
      "eval_loss": 0.5907140374183655,
      "eval_runtime": 0.9299,
      "eval_samples_per_second": 26.884,
      "eval_steps_per_second": 13.98,
      "num_input_tokens_seen": 9770824,
      "step": 17200
    },
    {
      "epoch": 301.8495575221239,
      "grad_norm": 8.460879996619042e-08,
      "learning_rate": 0.18267549560154991,
      "loss": 0.0,
      "num_input_tokens_seen": 9773640,
      "step": 17205
    },
    {
      "epoch": 301.9380530973451,
      "grad_norm": 3.8869423235610157e-08,
      "learning_rate": 0.18261800280709267,
      "loss": 0.0,
      "num_input_tokens_seen": 9777448,
      "step": 17210
    },
    {
      "epoch": 302.01769911504425,
      "grad_norm": 2.1254525250924416e-08,
      "learning_rate": 0.18256050498252957,
      "loss": 0.0,
      "num_input_tokens_seen": 9779976,
      "step": 17215
    },
    {
      "epoch": 302.1061946902655,
      "grad_norm": 3.7917409656529344e-08,
      "learning_rate": 0.18250300213672735,
      "loss": 0.0,
      "num_input_tokens_seen": 9782440,
      "step": 17220
    },
    {
      "epoch": 302.1946902654867,
      "grad_norm": 2.702996582115702e-08,
      "learning_rate": 0.18244549427855378,
      "loss": 0.0,
      "num_input_tokens_seen": 9785128,
      "step": 17225
    },
    {
      "epoch": 302.283185840708,
      "grad_norm": 4.234401984604119e-08,
      "learning_rate": 0.1823879814168772,
      "loss": 0.0,
      "num_input_tokens_seen": 9788056,
      "step": 17230
    },
    {
      "epoch": 302.37168141592923,
      "grad_norm": 2.25520366825549e-08,
      "learning_rate": 0.18233046356056692,
      "loss": 0.0,
      "num_input_tokens_seen": 9791032,
      "step": 17235
    },
    {
      "epoch": 302.46017699115043,
      "grad_norm": 3.6480916065784186e-08,
      "learning_rate": 0.18227294071849284,
      "loss": 0.0,
      "num_input_tokens_seen": 9793784,
      "step": 17240
    },
    {
      "epoch": 302.5486725663717,
      "grad_norm": 6.592506451852387e-08,
      "learning_rate": 0.18221541289952578,
      "loss": 0.0,
      "num_input_tokens_seen": 9796808,
      "step": 17245
    },
    {
      "epoch": 302.6371681415929,
      "grad_norm": 2.178947688946664e-08,
      "learning_rate": 0.18215788011253717,
      "loss": 0.0,
      "num_input_tokens_seen": 9799528,
      "step": 17250
    },
    {
      "epoch": 302.72566371681415,
      "grad_norm": 2.1903499458630904e-08,
      "learning_rate": 0.18210034236639935,
      "loss": 0.0,
      "num_input_tokens_seen": 9802040,
      "step": 17255
    },
    {
      "epoch": 302.8141592920354,
      "grad_norm": 5.4742944399777116e-08,
      "learning_rate": 0.1820427996699853,
      "loss": 0.0,
      "num_input_tokens_seen": 9804648,
      "step": 17260
    },
    {
      "epoch": 302.9026548672566,
      "grad_norm": 3.224980815730305e-08,
      "learning_rate": 0.1819852520321689,
      "loss": 0.0,
      "num_input_tokens_seen": 9807784,
      "step": 17265
    },
    {
      "epoch": 302.9911504424779,
      "grad_norm": 2.3666672177569126e-08,
      "learning_rate": 0.18192769946182466,
      "loss": 0.0,
      "num_input_tokens_seen": 9811080,
      "step": 17270
    },
    {
      "epoch": 303.070796460177,
      "grad_norm": 2.979492919052973e-08,
      "learning_rate": 0.18187014196782794,
      "loss": 0.0,
      "num_input_tokens_seen": 9813280,
      "step": 17275
    },
    {
      "epoch": 303.1592920353982,
      "grad_norm": 1.67226161806866e-08,
      "learning_rate": 0.18181257955905486,
      "loss": 0.0,
      "num_input_tokens_seen": 9816496,
      "step": 17280
    },
    {
      "epoch": 303.24778761061947,
      "grad_norm": 2.1843913344810062e-08,
      "learning_rate": 0.18175501224438217,
      "loss": 0.0,
      "num_input_tokens_seen": 9819424,
      "step": 17285
    },
    {
      "epoch": 303.3362831858407,
      "grad_norm": 3.538334425456924e-08,
      "learning_rate": 0.18169744003268756,
      "loss": 0.0,
      "num_input_tokens_seen": 9822464,
      "step": 17290
    },
    {
      "epoch": 303.42477876106193,
      "grad_norm": 3.8928011036887256e-08,
      "learning_rate": 0.18163986293284937,
      "loss": 0.0,
      "num_input_tokens_seen": 9825504,
      "step": 17295
    },
    {
      "epoch": 303.5132743362832,
      "grad_norm": 3.043370710997806e-08,
      "learning_rate": 0.18158228095374673,
      "loss": 0.0,
      "num_input_tokens_seen": 9828240,
      "step": 17300
    },
    {
      "epoch": 303.60176991150445,
      "grad_norm": 4.5588105734850615e-08,
      "learning_rate": 0.18152469410425945,
      "loss": 0.0,
      "num_input_tokens_seen": 9831120,
      "step": 17305
    },
    {
      "epoch": 303.69026548672565,
      "grad_norm": 3.518137248192943e-08,
      "learning_rate": 0.18146710239326813,
      "loss": 0.0,
      "num_input_tokens_seen": 9834368,
      "step": 17310
    },
    {
      "epoch": 303.7787610619469,
      "grad_norm": 5.079579068478779e-08,
      "learning_rate": 0.18140950582965423,
      "loss": 0.0,
      "num_input_tokens_seen": 9836976,
      "step": 17315
    },
    {
      "epoch": 303.86725663716817,
      "grad_norm": 5.085959386974537e-08,
      "learning_rate": 0.1813519044222998,
      "loss": 0.0,
      "num_input_tokens_seen": 9839680,
      "step": 17320
    },
    {
      "epoch": 303.95575221238937,
      "grad_norm": 9.627172659065764e-08,
      "learning_rate": 0.18129429818008772,
      "loss": 0.0,
      "num_input_tokens_seen": 9842240,
      "step": 17325
    },
    {
      "epoch": 304.0353982300885,
      "grad_norm": 3.881357457657941e-08,
      "learning_rate": 0.18123668711190163,
      "loss": 0.0,
      "num_input_tokens_seen": 9844824,
      "step": 17330
    },
    {
      "epoch": 304.12389380530976,
      "grad_norm": 2.155466738429368e-08,
      "learning_rate": 0.18117907122662583,
      "loss": 0.0,
      "num_input_tokens_seen": 9847880,
      "step": 17335
    },
    {
      "epoch": 304.21238938053096,
      "grad_norm": 2.201389648348595e-08,
      "learning_rate": 0.1811214505331454,
      "loss": 0.0,
      "num_input_tokens_seen": 9850632,
      "step": 17340
    },
    {
      "epoch": 304.3008849557522,
      "grad_norm": 2.2083453288246346e-08,
      "learning_rate": 0.1810638250403462,
      "loss": 0.0,
      "num_input_tokens_seen": 9853176,
      "step": 17345
    },
    {
      "epoch": 304.3893805309734,
      "grad_norm": 2.8793548878525144e-08,
      "learning_rate": 0.1810061947571148,
      "loss": 0.0,
      "num_input_tokens_seen": 9855736,
      "step": 17350
    },
    {
      "epoch": 304.4778761061947,
      "grad_norm": 4.892988769711337e-08,
      "learning_rate": 0.1809485596923385,
      "loss": 0.0,
      "num_input_tokens_seen": 9858520,
      "step": 17355
    },
    {
      "epoch": 304.56637168141594,
      "grad_norm": 3.909457646500414e-08,
      "learning_rate": 0.18089091985490546,
      "loss": 0.0,
      "num_input_tokens_seen": 9861288,
      "step": 17360
    },
    {
      "epoch": 304.65486725663715,
      "grad_norm": 3.358822198151756e-08,
      "learning_rate": 0.18083327525370432,
      "loss": 0.0,
      "num_input_tokens_seen": 9863976,
      "step": 17365
    },
    {
      "epoch": 304.7433628318584,
      "grad_norm": 4.301829292785442e-08,
      "learning_rate": 0.18077562589762464,
      "loss": 0.0,
      "num_input_tokens_seen": 9867192,
      "step": 17370
    },
    {
      "epoch": 304.83185840707966,
      "grad_norm": 9.283578350505195e-08,
      "learning_rate": 0.1807179717955567,
      "loss": 0.0,
      "num_input_tokens_seen": 9870552,
      "step": 17375
    },
    {
      "epoch": 304.92035398230087,
      "grad_norm": 8.00100252718039e-09,
      "learning_rate": 0.1806603129563915,
      "loss": 0.0,
      "num_input_tokens_seen": 9873416,
      "step": 17380
    },
    {
      "epoch": 305.0,
      "grad_norm": 2.9843127435924544e-07,
      "learning_rate": 0.1806026493890208,
      "loss": 0.0,
      "num_input_tokens_seen": 9875816,
      "step": 17385
    },
    {
      "epoch": 305.08849557522126,
      "grad_norm": 2.2709354396965864e-08,
      "learning_rate": 0.18054498110233688,
      "loss": 0.0,
      "num_input_tokens_seen": 9879064,
      "step": 17390
    },
    {
      "epoch": 305.17699115044246,
      "grad_norm": 2.366227747074845e-08,
      "learning_rate": 0.1804873081052331,
      "loss": 0.0,
      "num_input_tokens_seen": 9881928,
      "step": 17395
    },
    {
      "epoch": 305.2654867256637,
      "grad_norm": 1.9995521682858453e-08,
      "learning_rate": 0.18042963040660326,
      "loss": 0.0,
      "num_input_tokens_seen": 9884648,
      "step": 17400
    },
    {
      "epoch": 305.2654867256637,
      "eval_loss": 0.5804447531700134,
      "eval_runtime": 0.9527,
      "eval_samples_per_second": 26.24,
      "eval_steps_per_second": 13.645,
      "num_input_tokens_seen": 9884648,
      "step": 17400
    },
    {
      "epoch": 305.353982300885,
      "grad_norm": 4.283429788642934e-08,
      "learning_rate": 0.180371948015342,
      "loss": 0.0,
      "num_input_tokens_seen": 9887384,
      "step": 17405
    },
    {
      "epoch": 305.4424778761062,
      "grad_norm": 3.3788996489647616e-08,
      "learning_rate": 0.18031426094034472,
      "loss": 0.0,
      "num_input_tokens_seen": 9890344,
      "step": 17410
    },
    {
      "epoch": 305.53097345132744,
      "grad_norm": 5.2805820160983785e-08,
      "learning_rate": 0.18025656919050737,
      "loss": 0.0,
      "num_input_tokens_seen": 9893144,
      "step": 17415
    },
    {
      "epoch": 305.6194690265487,
      "grad_norm": 2.5231379652268515e-08,
      "learning_rate": 0.18019887277472688,
      "loss": 0.0,
      "num_input_tokens_seen": 9895848,
      "step": 17420
    },
    {
      "epoch": 305.7079646017699,
      "grad_norm": 2.0794196586848557e-08,
      "learning_rate": 0.18014117170190067,
      "loss": 0.0,
      "num_input_tokens_seen": 9899160,
      "step": 17425
    },
    {
      "epoch": 305.79646017699116,
      "grad_norm": 4.026223976438814e-08,
      "learning_rate": 0.18008346598092703,
      "loss": 0.0,
      "num_input_tokens_seen": 9902024,
      "step": 17430
    },
    {
      "epoch": 305.88495575221236,
      "grad_norm": 3.763779687915303e-08,
      "learning_rate": 0.18002575562070489,
      "loss": 0.0,
      "num_input_tokens_seen": 9904728,
      "step": 17435
    },
    {
      "epoch": 305.9734513274336,
      "grad_norm": 2.472095239625105e-08,
      "learning_rate": 0.1799680406301339,
      "loss": 0.0,
      "num_input_tokens_seen": 9907576,
      "step": 17440
    },
    {
      "epoch": 306.05309734513276,
      "grad_norm": 1.954628103817413e-08,
      "learning_rate": 0.17991032101811447,
      "loss": 0.0,
      "num_input_tokens_seen": 9909608,
      "step": 17445
    },
    {
      "epoch": 306.14159292035396,
      "grad_norm": 3.244623059117657e-08,
      "learning_rate": 0.1798525967935476,
      "loss": 0.0,
      "num_input_tokens_seen": 9912584,
      "step": 17450
    },
    {
      "epoch": 306.2300884955752,
      "grad_norm": 2.853486869014432e-08,
      "learning_rate": 0.17979486796533517,
      "loss": 0.0,
      "num_input_tokens_seen": 9915208,
      "step": 17455
    },
    {
      "epoch": 306.3185840707965,
      "grad_norm": 2.4397111886287348e-08,
      "learning_rate": 0.1797371345423797,
      "loss": 0.0,
      "num_input_tokens_seen": 9918184,
      "step": 17460
    },
    {
      "epoch": 306.4070796460177,
      "grad_norm": 4.546855691955898e-08,
      "learning_rate": 0.17967939653358436,
      "loss": 0.0,
      "num_input_tokens_seen": 9920776,
      "step": 17465
    },
    {
      "epoch": 306.49557522123894,
      "grad_norm": 2.1557857721177243e-08,
      "learning_rate": 0.17962165394785315,
      "loss": 0.0,
      "num_input_tokens_seen": 9923400,
      "step": 17470
    },
    {
      "epoch": 306.5840707964602,
      "grad_norm": 3.0174735599075575e-08,
      "learning_rate": 0.17956390679409057,
      "loss": 0.0,
      "num_input_tokens_seen": 9927256,
      "step": 17475
    },
    {
      "epoch": 306.6725663716814,
      "grad_norm": 1.5208980741476807e-08,
      "learning_rate": 0.1795061550812021,
      "loss": 0.0,
      "num_input_tokens_seen": 9929704,
      "step": 17480
    },
    {
      "epoch": 306.76106194690266,
      "grad_norm": 4.5653084868035876e-08,
      "learning_rate": 0.1794483988180937,
      "loss": 0.0,
      "num_input_tokens_seen": 9932872,
      "step": 17485
    },
    {
      "epoch": 306.8495575221239,
      "grad_norm": 6.080500725147431e-08,
      "learning_rate": 0.17939063801367214,
      "loss": 0.0,
      "num_input_tokens_seen": 9935608,
      "step": 17490
    },
    {
      "epoch": 306.9380530973451,
      "grad_norm": 2.7800538759947813e-08,
      "learning_rate": 0.17933287267684483,
      "loss": 0.0,
      "num_input_tokens_seen": 9938632,
      "step": 17495
    },
    {
      "epoch": 307.01769911504425,
      "grad_norm": 3.716049334911986e-08,
      "learning_rate": 0.17927510281651995,
      "loss": 0.0,
      "num_input_tokens_seen": 9940824,
      "step": 17500
    },
    {
      "epoch": 307.1061946902655,
      "grad_norm": 3.454014319004273e-08,
      "learning_rate": 0.17921732844160634,
      "loss": 0.0,
      "num_input_tokens_seen": 9943704,
      "step": 17505
    },
    {
      "epoch": 307.1946902654867,
      "grad_norm": 2.6029793431803228e-08,
      "learning_rate": 0.17915954956101351,
      "loss": 0.0,
      "num_input_tokens_seen": 9946184,
      "step": 17510
    },
    {
      "epoch": 307.283185840708,
      "grad_norm": 2.8860005940600786e-08,
      "learning_rate": 0.17910176618365165,
      "loss": 0.0,
      "num_input_tokens_seen": 9949000,
      "step": 17515
    },
    {
      "epoch": 307.37168141592923,
      "grad_norm": 4.0711775284307805e-08,
      "learning_rate": 0.17904397831843177,
      "loss": 0.0,
      "num_input_tokens_seen": 9952072,
      "step": 17520
    },
    {
      "epoch": 307.46017699115043,
      "grad_norm": 2.943823318446448e-08,
      "learning_rate": 0.17898618597426547,
      "loss": 0.0,
      "num_input_tokens_seen": 9954440,
      "step": 17525
    },
    {
      "epoch": 307.5486725663717,
      "grad_norm": 2.0201213146719965e-08,
      "learning_rate": 0.17892838916006495,
      "loss": 0.0,
      "num_input_tokens_seen": 9957560,
      "step": 17530
    },
    {
      "epoch": 307.6371681415929,
      "grad_norm": 3.506389845142621e-08,
      "learning_rate": 0.17887058788474333,
      "loss": 0.0,
      "num_input_tokens_seen": 9960216,
      "step": 17535
    },
    {
      "epoch": 307.72566371681415,
      "grad_norm": 2.2905688012997416e-08,
      "learning_rate": 0.17881278215721427,
      "loss": 0.0,
      "num_input_tokens_seen": 9963288,
      "step": 17540
    },
    {
      "epoch": 307.8141592920354,
      "grad_norm": 2.3333022625138256e-08,
      "learning_rate": 0.1787549719863921,
      "loss": 0.0,
      "num_input_tokens_seen": 9965960,
      "step": 17545
    },
    {
      "epoch": 307.9026548672566,
      "grad_norm": 2.6960890409100102e-08,
      "learning_rate": 0.17869715738119188,
      "loss": 0.0,
      "num_input_tokens_seen": 9969000,
      "step": 17550
    },
    {
      "epoch": 307.9911504424779,
      "grad_norm": 2.1280564865833185e-08,
      "learning_rate": 0.17863933835052936,
      "loss": 0.0,
      "num_input_tokens_seen": 9971928,
      "step": 17555
    },
    {
      "epoch": 308.070796460177,
      "grad_norm": 1.5566447686410356e-08,
      "learning_rate": 0.17858151490332097,
      "loss": 0.0,
      "num_input_tokens_seen": 9974680,
      "step": 17560
    },
    {
      "epoch": 308.1592920353982,
      "grad_norm": 3.775135937189589e-08,
      "learning_rate": 0.17852368704848381,
      "loss": 0.0,
      "num_input_tokens_seen": 9977352,
      "step": 17565
    },
    {
      "epoch": 308.24778761061947,
      "grad_norm": 1.871382160345547e-08,
      "learning_rate": 0.17846585479493565,
      "loss": 0.0,
      "num_input_tokens_seen": 9980584,
      "step": 17570
    },
    {
      "epoch": 308.3362831858407,
      "grad_norm": 5.1029221737053376e-08,
      "learning_rate": 0.178408018151595,
      "loss": 0.0,
      "num_input_tokens_seen": 9983064,
      "step": 17575
    },
    {
      "epoch": 308.42477876106193,
      "grad_norm": 1.6652190737431738e-08,
      "learning_rate": 0.17835017712738085,
      "loss": 0.0,
      "num_input_tokens_seen": 9985768,
      "step": 17580
    },
    {
      "epoch": 308.5132743362832,
      "grad_norm": 1.4837170603243521e-08,
      "learning_rate": 0.17829233173121323,
      "loss": 0.0,
      "num_input_tokens_seen": 9988856,
      "step": 17585
    },
    {
      "epoch": 308.60176991150445,
      "grad_norm": 3.310266905032222e-08,
      "learning_rate": 0.17823448197201244,
      "loss": 0.0,
      "num_input_tokens_seen": 9991352,
      "step": 17590
    },
    {
      "epoch": 308.69026548672565,
      "grad_norm": 3.6944566517149724e-08,
      "learning_rate": 0.1781766278586997,
      "loss": 0.0,
      "num_input_tokens_seen": 9994584,
      "step": 17595
    },
    {
      "epoch": 308.7787610619469,
      "grad_norm": 1.957417516962323e-08,
      "learning_rate": 0.1781187694001969,
      "loss": 0.0,
      "num_input_tokens_seen": 9997288,
      "step": 17600
    },
    {
      "epoch": 308.7787610619469,
      "eval_loss": 0.6051491498947144,
      "eval_runtime": 0.9375,
      "eval_samples_per_second": 26.666,
      "eval_steps_per_second": 13.866,
      "num_input_tokens_seen": 9997288,
      "step": 17600
    },
    {
      "epoch": 308.86725663716817,
      "grad_norm": 1.4702422390655556e-08,
      "learning_rate": 0.1780609066054265,
      "loss": 0.0,
      "num_input_tokens_seen": 9999976,
      "step": 17605
    },
    {
      "epoch": 308.95575221238937,
      "grad_norm": 3.413185112322026e-08,
      "learning_rate": 0.17800303948331164,
      "loss": 0.0,
      "num_input_tokens_seen": 10003320,
      "step": 17610
    },
    {
      "epoch": 309.0353982300885,
      "grad_norm": 3.325484598803996e-08,
      "learning_rate": 0.1779451680427762,
      "loss": 0.0,
      "num_input_tokens_seen": 10005664,
      "step": 17615
    },
    {
      "epoch": 309.12389380530976,
      "grad_norm": 3.8090522735956256e-08,
      "learning_rate": 0.17788729229274464,
      "loss": 0.0,
      "num_input_tokens_seen": 10008288,
      "step": 17620
    },
    {
      "epoch": 309.21238938053096,
      "grad_norm": 3.1150396040402484e-08,
      "learning_rate": 0.17782941224214222,
      "loss": 0.0,
      "num_input_tokens_seen": 10011440,
      "step": 17625
    },
    {
      "epoch": 309.3008849557522,
      "grad_norm": 3.800900927330986e-08,
      "learning_rate": 0.17777152789989464,
      "loss": 0.0,
      "num_input_tokens_seen": 10014720,
      "step": 17630
    },
    {
      "epoch": 309.3893805309734,
      "grad_norm": 3.4252494174324966e-08,
      "learning_rate": 0.17771363927492845,
      "loss": 0.0,
      "num_input_tokens_seen": 10017408,
      "step": 17635
    },
    {
      "epoch": 309.4778761061947,
      "grad_norm": 1.5553482057839574e-08,
      "learning_rate": 0.17765574637617085,
      "loss": 0.0,
      "num_input_tokens_seen": 10020096,
      "step": 17640
    },
    {
      "epoch": 309.56637168141594,
      "grad_norm": 3.755023669782531e-08,
      "learning_rate": 0.17759784921254962,
      "loss": 0.0,
      "num_input_tokens_seen": 10022944,
      "step": 17645
    },
    {
      "epoch": 309.65486725663715,
      "grad_norm": 3.344444365893651e-08,
      "learning_rate": 0.1775399477929932,
      "loss": 0.0,
      "num_input_tokens_seen": 10025344,
      "step": 17650
    },
    {
      "epoch": 309.7433628318584,
      "grad_norm": 3.3115984621190364e-08,
      "learning_rate": 0.17748204212643076,
      "loss": 0.0,
      "num_input_tokens_seen": 10028656,
      "step": 17655
    },
    {
      "epoch": 309.83185840707966,
      "grad_norm": 2.402643772825286e-08,
      "learning_rate": 0.17742413222179204,
      "loss": 0.0,
      "num_input_tokens_seen": 10031616,
      "step": 17660
    },
    {
      "epoch": 309.92035398230087,
      "grad_norm": 2.2982078462518984e-08,
      "learning_rate": 0.17736621808800754,
      "loss": 0.0,
      "num_input_tokens_seen": 10034416,
      "step": 17665
    },
    {
      "epoch": 310.0,
      "grad_norm": 1.1491189333412422e-08,
      "learning_rate": 0.17730829973400827,
      "loss": 0.0,
      "num_input_tokens_seen": 10036912,
      "step": 17670
    },
    {
      "epoch": 310.08849557522126,
      "grad_norm": 7.406769952922332e-08,
      "learning_rate": 0.17725037716872602,
      "loss": 0.0,
      "num_input_tokens_seen": 10039408,
      "step": 17675
    },
    {
      "epoch": 310.17699115044246,
      "grad_norm": 4.5581735719224525e-08,
      "learning_rate": 0.17719245040109313,
      "loss": 0.0,
      "num_input_tokens_seen": 10042304,
      "step": 17680
    },
    {
      "epoch": 310.2654867256637,
      "grad_norm": 3.112058166720999e-08,
      "learning_rate": 0.17713451944004271,
      "loss": 0.0,
      "num_input_tokens_seen": 10045232,
      "step": 17685
    },
    {
      "epoch": 310.353982300885,
      "grad_norm": 2.4843648915862104e-08,
      "learning_rate": 0.17707658429450843,
      "loss": 0.0,
      "num_input_tokens_seen": 10048608,
      "step": 17690
    },
    {
      "epoch": 310.4424778761062,
      "grad_norm": 7.288542036576473e-08,
      "learning_rate": 0.1770186449734245,
      "loss": 0.0,
      "num_input_tokens_seen": 10051568,
      "step": 17695
    },
    {
      "epoch": 310.53097345132744,
      "grad_norm": 2.948867283691925e-08,
      "learning_rate": 0.17696070148572599,
      "loss": 0.0,
      "num_input_tokens_seen": 10054544,
      "step": 17700
    },
    {
      "epoch": 310.6194690265487,
      "grad_norm": 3.889022792691321e-08,
      "learning_rate": 0.17690275384034856,
      "loss": 0.0,
      "num_input_tokens_seen": 10057088,
      "step": 17705
    },
    {
      "epoch": 310.7079646017699,
      "grad_norm": 1.5657384722089773e-08,
      "learning_rate": 0.17684480204622835,
      "loss": 0.0,
      "num_input_tokens_seen": 10059776,
      "step": 17710
    },
    {
      "epoch": 310.79646017699116,
      "grad_norm": 4.207302595204965e-08,
      "learning_rate": 0.1767868461123023,
      "loss": 0.0,
      "num_input_tokens_seen": 10062576,
      "step": 17715
    },
    {
      "epoch": 310.88495575221236,
      "grad_norm": 2.8579403732464925e-08,
      "learning_rate": 0.176728886047508,
      "loss": 0.0,
      "num_input_tokens_seen": 10065632,
      "step": 17720
    },
    {
      "epoch": 310.9734513274336,
      "grad_norm": 1.3303099066774848e-08,
      "learning_rate": 0.17667092186078362,
      "loss": 0.0,
      "num_input_tokens_seen": 10068256,
      "step": 17725
    },
    {
      "epoch": 311.05309734513276,
      "grad_norm": 1.2292377782330277e-08,
      "learning_rate": 0.17661295356106785,
      "loss": 0.0,
      "num_input_tokens_seen": 10070480,
      "step": 17730
    },
    {
      "epoch": 311.14159292035396,
      "grad_norm": 2.2120381970580638e-08,
      "learning_rate": 0.1765549811573002,
      "loss": 0.0,
      "num_input_tokens_seen": 10073136,
      "step": 17735
    },
    {
      "epoch": 311.2300884955752,
      "grad_norm": 2.4424201328088202e-08,
      "learning_rate": 0.17649700465842078,
      "loss": 0.0,
      "num_input_tokens_seen": 10076160,
      "step": 17740
    },
    {
      "epoch": 311.3185840707965,
      "grad_norm": 2.0671638623070976e-08,
      "learning_rate": 0.17643902407337023,
      "loss": 0.0,
      "num_input_tokens_seen": 10079312,
      "step": 17745
    },
    {
      "epoch": 311.4070796460177,
      "grad_norm": 4.1152375729325286e-08,
      "learning_rate": 0.17638103941108993,
      "loss": 0.0,
      "num_input_tokens_seen": 10082208,
      "step": 17750
    },
    {
      "epoch": 311.49557522123894,
      "grad_norm": 2.4225277783784804e-08,
      "learning_rate": 0.1763230506805218,
      "loss": 0.0,
      "num_input_tokens_seen": 10085312,
      "step": 17755
    },
    {
      "epoch": 311.5840707964602,
      "grad_norm": 2.5195948438749838e-08,
      "learning_rate": 0.1762650578906085,
      "loss": 0.0,
      "num_input_tokens_seen": 10088352,
      "step": 17760
    },
    {
      "epoch": 311.6725663716814,
      "grad_norm": 1.4683208426902183e-08,
      "learning_rate": 0.1762070610502932,
      "loss": 0.0,
      "num_input_tokens_seen": 10091216,
      "step": 17765
    },
    {
      "epoch": 311.76106194690266,
      "grad_norm": 2.5868082786928426e-08,
      "learning_rate": 0.17614906016851975,
      "loss": 0.0,
      "num_input_tokens_seen": 10093760,
      "step": 17770
    },
    {
      "epoch": 311.8495575221239,
      "grad_norm": 3.312248963993625e-08,
      "learning_rate": 0.17609105525423258,
      "loss": 0.0,
      "num_input_tokens_seen": 10097136,
      "step": 17775
    },
    {
      "epoch": 311.9380530973451,
      "grad_norm": 4.0875747231439163e-08,
      "learning_rate": 0.1760330463163768,
      "loss": 0.0,
      "num_input_tokens_seen": 10099984,
      "step": 17780
    },
    {
      "epoch": 312.01769911504425,
      "grad_norm": 1.0893040247594854e-08,
      "learning_rate": 0.17597503336389816,
      "loss": 0.0,
      "num_input_tokens_seen": 10102272,
      "step": 17785
    },
    {
      "epoch": 312.1061946902655,
      "grad_norm": 4.0469526396691435e-08,
      "learning_rate": 0.17591701640574298,
      "loss": 0.0,
      "num_input_tokens_seen": 10104768,
      "step": 17790
    },
    {
      "epoch": 312.1946902654867,
      "grad_norm": 2.211688077125018e-08,
      "learning_rate": 0.17585899545085815,
      "loss": 0.0,
      "num_input_tokens_seen": 10108304,
      "step": 17795
    },
    {
      "epoch": 312.283185840708,
      "grad_norm": 3.5944978549196094e-08,
      "learning_rate": 0.17580097050819124,
      "loss": 0.0,
      "num_input_tokens_seen": 10111472,
      "step": 17800
    },
    {
      "epoch": 312.283185840708,
      "eval_loss": 0.5829068422317505,
      "eval_runtime": 0.9161,
      "eval_samples_per_second": 27.29,
      "eval_steps_per_second": 14.191,
      "num_input_tokens_seen": 10111472,
      "step": 17800
    },
    {
      "epoch": 312.37168141592923,
      "grad_norm": 4.848090284781392e-08,
      "learning_rate": 0.17574294158669046,
      "loss": 0.0,
      "num_input_tokens_seen": 10113968,
      "step": 17805
    },
    {
      "epoch": 312.46017699115043,
      "grad_norm": 2.23792397946454e-08,
      "learning_rate": 0.17568490869530456,
      "loss": 0.0,
      "num_input_tokens_seen": 10116528,
      "step": 17810
    },
    {
      "epoch": 312.5486725663717,
      "grad_norm": 4.8029367150093094e-08,
      "learning_rate": 0.17562687184298295,
      "loss": 0.0,
      "num_input_tokens_seen": 10119520,
      "step": 17815
    },
    {
      "epoch": 312.6371681415929,
      "grad_norm": 2.6664874752668766e-08,
      "learning_rate": 0.1755688310386757,
      "loss": 0.0,
      "num_input_tokens_seen": 10122592,
      "step": 17820
    },
    {
      "epoch": 312.72566371681415,
      "grad_norm": 2.75298663865442e-08,
      "learning_rate": 0.17551078629133335,
      "loss": 0.0,
      "num_input_tokens_seen": 10125376,
      "step": 17825
    },
    {
      "epoch": 312.8141592920354,
      "grad_norm": 2.757562178601347e-08,
      "learning_rate": 0.17545273760990718,
      "loss": 0.0,
      "num_input_tokens_seen": 10127920,
      "step": 17830
    },
    {
      "epoch": 312.9026548672566,
      "grad_norm": 2.9937258005929834e-08,
      "learning_rate": 0.17539468500334904,
      "loss": 0.0,
      "num_input_tokens_seen": 10131136,
      "step": 17835
    },
    {
      "epoch": 312.9911504424779,
      "grad_norm": 2.5270885828376777e-08,
      "learning_rate": 0.17533662848061132,
      "loss": 0.0,
      "num_input_tokens_seen": 10133664,
      "step": 17840
    },
    {
      "epoch": 313.070796460177,
      "grad_norm": 6.40913455640657e-08,
      "learning_rate": 0.1752785680506471,
      "loss": 0.0,
      "num_input_tokens_seen": 10136024,
      "step": 17845
    },
    {
      "epoch": 313.1592920353982,
      "grad_norm": 2.8228601678392806e-08,
      "learning_rate": 0.17522050372241,
      "loss": 0.0,
      "num_input_tokens_seen": 10139064,
      "step": 17850
    },
    {
      "epoch": 313.24778761061947,
      "grad_norm": 1.9679985641118947e-08,
      "learning_rate": 0.17516243550485425,
      "loss": 0.0,
      "num_input_tokens_seen": 10141944,
      "step": 17855
    },
    {
      "epoch": 313.3362831858407,
      "grad_norm": 1.7207266722607528e-08,
      "learning_rate": 0.17510436340693478,
      "loss": 0.0,
      "num_input_tokens_seen": 10144392,
      "step": 17860
    },
    {
      "epoch": 313.42477876106193,
      "grad_norm": 4.036898459958138e-08,
      "learning_rate": 0.175046287437607,
      "loss": 0.0,
      "num_input_tokens_seen": 10146840,
      "step": 17865
    },
    {
      "epoch": 313.5132743362832,
      "grad_norm": 1.0488474089243027e-08,
      "learning_rate": 0.17498820760582695,
      "loss": 0.0,
      "num_input_tokens_seen": 10150104,
      "step": 17870
    },
    {
      "epoch": 313.60176991150445,
      "grad_norm": 2.1419173990011586e-08,
      "learning_rate": 0.1749301239205512,
      "loss": 0.0,
      "num_input_tokens_seen": 10152888,
      "step": 17875
    },
    {
      "epoch": 313.69026548672565,
      "grad_norm": 2.4286030964049132e-08,
      "learning_rate": 0.1748720363907371,
      "loss": 0.0,
      "num_input_tokens_seen": 10155192,
      "step": 17880
    },
    {
      "epoch": 313.7787610619469,
      "grad_norm": 2.9850330207636944e-08,
      "learning_rate": 0.17481394502534242,
      "loss": 0.0,
      "num_input_tokens_seen": 10157976,
      "step": 17885
    },
    {
      "epoch": 313.86725663716817,
      "grad_norm": 5.9233379090528615e-08,
      "learning_rate": 0.17475584983332562,
      "loss": 0.0,
      "num_input_tokens_seen": 10161144,
      "step": 17890
    },
    {
      "epoch": 313.95575221238937,
      "grad_norm": 3.2806294569809324e-08,
      "learning_rate": 0.17469775082364558,
      "loss": 0.0,
      "num_input_tokens_seen": 10164328,
      "step": 17895
    },
    {
      "epoch": 314.0353982300885,
      "grad_norm": 3.494783484825348e-08,
      "learning_rate": 0.17463964800526205,
      "loss": 0.0,
      "num_input_tokens_seen": 10167128,
      "step": 17900
    },
    {
      "epoch": 314.12389380530976,
      "grad_norm": 4.042754397914905e-08,
      "learning_rate": 0.17458154138713522,
      "loss": 0.0,
      "num_input_tokens_seen": 10169848,
      "step": 17905
    },
    {
      "epoch": 314.21238938053096,
      "grad_norm": 1.1233067809257591e-08,
      "learning_rate": 0.17452343097822576,
      "loss": 0.0,
      "num_input_tokens_seen": 10172920,
      "step": 17910
    },
    {
      "epoch": 314.3008849557522,
      "grad_norm": 4.779021267609096e-08,
      "learning_rate": 0.17446531678749497,
      "loss": 0.0,
      "num_input_tokens_seen": 10175576,
      "step": 17915
    },
    {
      "epoch": 314.3893805309734,
      "grad_norm": 1.8469007656563008e-08,
      "learning_rate": 0.17440719882390496,
      "loss": 0.0,
      "num_input_tokens_seen": 10178056,
      "step": 17920
    },
    {
      "epoch": 314.4778761061947,
      "grad_norm": 2.601538895419253e-08,
      "learning_rate": 0.17434907709641814,
      "loss": 0.0,
      "num_input_tokens_seen": 10181608,
      "step": 17925
    },
    {
      "epoch": 314.56637168141594,
      "grad_norm": 2.621306727235151e-08,
      "learning_rate": 0.17429095161399769,
      "loss": 0.0,
      "num_input_tokens_seen": 10184792,
      "step": 17930
    },
    {
      "epoch": 314.65486725663715,
      "grad_norm": 1.7393782414387715e-08,
      "learning_rate": 0.1742328223856072,
      "loss": 0.0,
      "num_input_tokens_seen": 10187576,
      "step": 17935
    },
    {
      "epoch": 314.7433628318584,
      "grad_norm": 2.6154546972634307e-08,
      "learning_rate": 0.174174689420211,
      "loss": 0.0,
      "num_input_tokens_seen": 10189928,
      "step": 17940
    },
    {
      "epoch": 314.83185840707966,
      "grad_norm": 3.161093786729907e-08,
      "learning_rate": 0.1741165527267739,
      "loss": 0.0,
      "num_input_tokens_seen": 10192616,
      "step": 17945
    },
    {
      "epoch": 314.92035398230087,
      "grad_norm": 2.1355209156581623e-08,
      "learning_rate": 0.17405841231426125,
      "loss": 0.0,
      "num_input_tokens_seen": 10195192,
      "step": 17950
    },
    {
      "epoch": 315.0,
      "grad_norm": 9.695042280100097e-08,
      "learning_rate": 0.1740002681916391,
      "loss": 0.0,
      "num_input_tokens_seen": 10197744,
      "step": 17955
    },
    {
      "epoch": 315.08849557522126,
      "grad_norm": 5.1600359540771024e-08,
      "learning_rate": 0.17394212036787401,
      "loss": 0.0,
      "num_input_tokens_seen": 10200736,
      "step": 17960
    },
    {
      "epoch": 315.17699115044246,
      "grad_norm": 2.231849371980843e-08,
      "learning_rate": 0.1738839688519331,
      "loss": 0.0,
      "num_input_tokens_seen": 10203952,
      "step": 17965
    },
    {
      "epoch": 315.2654867256637,
      "grad_norm": 5.298907623796367e-08,
      "learning_rate": 0.17382581365278402,
      "loss": 0.0,
      "num_input_tokens_seen": 10206896,
      "step": 17970
    },
    {
      "epoch": 315.353982300885,
      "grad_norm": 6.470151703297233e-08,
      "learning_rate": 0.17376765477939507,
      "loss": 0.0,
      "num_input_tokens_seen": 10210000,
      "step": 17975
    },
    {
      "epoch": 315.4424778761062,
      "grad_norm": 3.503372170143848e-08,
      "learning_rate": 0.1737094922407351,
      "loss": 0.0,
      "num_input_tokens_seen": 10212704,
      "step": 17980
    },
    {
      "epoch": 315.53097345132744,
      "grad_norm": 2.40850770438783e-08,
      "learning_rate": 0.1736513260457734,
      "loss": 0.0,
      "num_input_tokens_seen": 10215136,
      "step": 17985
    },
    {
      "epoch": 315.6194690265487,
      "grad_norm": 3.802251669071666e-08,
      "learning_rate": 0.17359315620348006,
      "loss": 0.0,
      "num_input_tokens_seen": 10218080,
      "step": 17990
    },
    {
      "epoch": 315.7079646017699,
      "grad_norm": 4.741281500741934e-08,
      "learning_rate": 0.17353498272282547,
      "loss": 0.0,
      "num_input_tokens_seen": 10220752,
      "step": 17995
    },
    {
      "epoch": 315.79646017699116,
      "grad_norm": 2.1896131130461072e-08,
      "learning_rate": 0.17347680561278087,
      "loss": 0.0,
      "num_input_tokens_seen": 10223648,
      "step": 18000
    },
    {
      "epoch": 315.79646017699116,
      "eval_loss": 0.6007505059242249,
      "eval_runtime": 0.9243,
      "eval_samples_per_second": 27.048,
      "eval_steps_per_second": 14.065,
      "num_input_tokens_seen": 10223648,
      "step": 18000
    },
    {
      "epoch": 315.88495575221236,
      "grad_norm": 2.2873168248338516e-08,
      "learning_rate": 0.1734186248823178,
      "loss": 0.0,
      "num_input_tokens_seen": 10226400,
      "step": 18005
    },
    {
      "epoch": 315.9734513274336,
      "grad_norm": 1.7366234672522296e-08,
      "learning_rate": 0.17336044054040844,
      "loss": 0.0,
      "num_input_tokens_seen": 10229536,
      "step": 18010
    },
    {
      "epoch": 316.05309734513276,
      "grad_norm": 1.516602843310011e-08,
      "learning_rate": 0.1733022525960256,
      "loss": 0.0,
      "num_input_tokens_seen": 10231656,
      "step": 18015
    },
    {
      "epoch": 316.14159292035396,
      "grad_norm": 2.2598372950710655e-08,
      "learning_rate": 0.1732440610581426,
      "loss": 0.0,
      "num_input_tokens_seen": 10234248,
      "step": 18020
    },
    {
      "epoch": 316.2300884955752,
      "grad_norm": 3.17091881640863e-08,
      "learning_rate": 0.17318586593573326,
      "loss": 0.0,
      "num_input_tokens_seen": 10236808,
      "step": 18025
    },
    {
      "epoch": 316.3185840707965,
      "grad_norm": 2.218844308288226e-08,
      "learning_rate": 0.17312766723777204,
      "loss": 0.0,
      "num_input_tokens_seen": 10239496,
      "step": 18030
    },
    {
      "epoch": 316.4070796460177,
      "grad_norm": 2.6775033745707333e-08,
      "learning_rate": 0.1730694649732339,
      "loss": 0.0,
      "num_input_tokens_seen": 10242392,
      "step": 18035
    },
    {
      "epoch": 316.49557522123894,
      "grad_norm": 1.6012176473623185e-08,
      "learning_rate": 0.17301125915109428,
      "loss": 0.0,
      "num_input_tokens_seen": 10245304,
      "step": 18040
    },
    {
      "epoch": 316.5840707964602,
      "grad_norm": 2.3501193879837956e-08,
      "learning_rate": 0.17295304978032938,
      "loss": 0.0,
      "num_input_tokens_seen": 10248088,
      "step": 18045
    },
    {
      "epoch": 316.6725663716814,
      "grad_norm": 3.90217174128793e-08,
      "learning_rate": 0.17289483686991577,
      "loss": 0.0,
      "num_input_tokens_seen": 10251464,
      "step": 18050
    },
    {
      "epoch": 316.76106194690266,
      "grad_norm": 1.768485269337816e-08,
      "learning_rate": 0.1728366204288306,
      "loss": 0.0,
      "num_input_tokens_seen": 10254424,
      "step": 18055
    },
    {
      "epoch": 316.8495575221239,
      "grad_norm": 2.1883304057723763e-08,
      "learning_rate": 0.17277840046605153,
      "loss": 0.0,
      "num_input_tokens_seen": 10257896,
      "step": 18060
    },
    {
      "epoch": 316.9380530973451,
      "grad_norm": 4.811251841374542e-08,
      "learning_rate": 0.17272017699055686,
      "loss": 0.0,
      "num_input_tokens_seen": 10260424,
      "step": 18065
    },
    {
      "epoch": 317.01769911504425,
      "grad_norm": 1.7620070735802074e-08,
      "learning_rate": 0.17266195001132542,
      "loss": 0.0,
      "num_input_tokens_seen": 10263120,
      "step": 18070
    },
    {
      "epoch": 317.1061946902655,
      "grad_norm": 1.9182470722967082e-08,
      "learning_rate": 0.17260371953733647,
      "loss": 0.0,
      "num_input_tokens_seen": 10265568,
      "step": 18075
    },
    {
      "epoch": 317.1946902654867,
      "grad_norm": 3.03300993209632e-08,
      "learning_rate": 0.1725454855775699,
      "loss": 0.0,
      "num_input_tokens_seen": 10268208,
      "step": 18080
    },
    {
      "epoch": 317.283185840708,
      "grad_norm": 4.363844041677112e-08,
      "learning_rate": 0.17248724814100616,
      "loss": 0.0,
      "num_input_tokens_seen": 10271376,
      "step": 18085
    },
    {
      "epoch": 317.37168141592923,
      "grad_norm": 1.250117787066074e-08,
      "learning_rate": 0.17242900723662619,
      "loss": 0.0,
      "num_input_tokens_seen": 10274176,
      "step": 18090
    },
    {
      "epoch": 317.46017699115043,
      "grad_norm": 1.922675707533017e-08,
      "learning_rate": 0.1723707628734114,
      "loss": 0.0,
      "num_input_tokens_seen": 10277264,
      "step": 18095
    },
    {
      "epoch": 317.5486725663717,
      "grad_norm": 3.803106096711417e-08,
      "learning_rate": 0.1723125150603438,
      "loss": 0.0,
      "num_input_tokens_seen": 10279744,
      "step": 18100
    },
    {
      "epoch": 317.6371681415929,
      "grad_norm": 3.4897112755061244e-08,
      "learning_rate": 0.1722542638064061,
      "loss": 0.0,
      "num_input_tokens_seen": 10283040,
      "step": 18105
    },
    {
      "epoch": 317.72566371681415,
      "grad_norm": 1.4367214973276532e-08,
      "learning_rate": 0.17219600912058117,
      "loss": 0.0,
      "num_input_tokens_seen": 10285536,
      "step": 18110
    },
    {
      "epoch": 317.8141592920354,
      "grad_norm": 2.9428907311057628e-08,
      "learning_rate": 0.17213775101185272,
      "loss": 0.0,
      "num_input_tokens_seen": 10288672,
      "step": 18115
    },
    {
      "epoch": 317.9026548672566,
      "grad_norm": 4.240757789375493e-08,
      "learning_rate": 0.17207948948920485,
      "loss": 0.0,
      "num_input_tokens_seen": 10291456,
      "step": 18120
    },
    {
      "epoch": 317.9911504424779,
      "grad_norm": 3.164161910262919e-08,
      "learning_rate": 0.17202122456162228,
      "loss": 0.0,
      "num_input_tokens_seen": 10294480,
      "step": 18125
    },
    {
      "epoch": 318.070796460177,
      "grad_norm": 3.797866554577922e-08,
      "learning_rate": 0.17196295623809013,
      "loss": 0.0,
      "num_input_tokens_seen": 10296688,
      "step": 18130
    },
    {
      "epoch": 318.1592920353982,
      "grad_norm": 3.1976064462924114e-08,
      "learning_rate": 0.1719046845275941,
      "loss": 0.0,
      "num_input_tokens_seen": 10299552,
      "step": 18135
    },
    {
      "epoch": 318.24778761061947,
      "grad_norm": 2.7210514730313662e-08,
      "learning_rate": 0.17184640943912044,
      "loss": 0.0,
      "num_input_tokens_seen": 10302240,
      "step": 18140
    },
    {
      "epoch": 318.3362831858407,
      "grad_norm": 1.3284125799373214e-08,
      "learning_rate": 0.1717881309816559,
      "loss": 0.0,
      "num_input_tokens_seen": 10305200,
      "step": 18145
    },
    {
      "epoch": 318.42477876106193,
      "grad_norm": 1.9616726021354225e-08,
      "learning_rate": 0.1717298491641878,
      "loss": 0.0,
      "num_input_tokens_seen": 10308688,
      "step": 18150
    },
    {
      "epoch": 318.5132743362832,
      "grad_norm": 2.881006189170421e-08,
      "learning_rate": 0.17167156399570385,
      "loss": 0.0,
      "num_input_tokens_seen": 10311328,
      "step": 18155
    },
    {
      "epoch": 318.60176991150445,
      "grad_norm": 2.909943219719935e-08,
      "learning_rate": 0.17161327548519242,
      "loss": 0.0,
      "num_input_tokens_seen": 10313952,
      "step": 18160
    },
    {
      "epoch": 318.69026548672565,
      "grad_norm": 2.3145776850697075e-08,
      "learning_rate": 0.1715549836416423,
      "loss": 0.0,
      "num_input_tokens_seen": 10316656,
      "step": 18165
    },
    {
      "epoch": 318.7787610619469,
      "grad_norm": 2.911355245771574e-08,
      "learning_rate": 0.17149668847404279,
      "loss": 0.0,
      "num_input_tokens_seen": 10319808,
      "step": 18170
    },
    {
      "epoch": 318.86725663716817,
      "grad_norm": 3.6916578238788134e-08,
      "learning_rate": 0.1714383899913838,
      "loss": 0.0,
      "num_input_tokens_seen": 10322336,
      "step": 18175
    },
    {
      "epoch": 318.95575221238937,
      "grad_norm": 4.312478552037646e-08,
      "learning_rate": 0.17138008820265563,
      "loss": 0.0,
      "num_input_tokens_seen": 10325248,
      "step": 18180
    },
    {
      "epoch": 319.0353982300885,
      "grad_norm": 1.5272995312898274e-08,
      "learning_rate": 0.17132178311684917,
      "loss": 0.0,
      "num_input_tokens_seen": 10327968,
      "step": 18185
    },
    {
      "epoch": 319.12389380530976,
      "grad_norm": 9.147719026714185e-09,
      "learning_rate": 0.1712634747429559,
      "loss": 0.0,
      "num_input_tokens_seen": 10330800,
      "step": 18190
    },
    {
      "epoch": 319.21238938053096,
      "grad_norm": 3.3911376817741257e-08,
      "learning_rate": 0.17120516308996753,
      "loss": 0.0,
      "num_input_tokens_seen": 10333424,
      "step": 18195
    },
    {
      "epoch": 319.3008849557522,
      "grad_norm": 6.868776836199686e-08,
      "learning_rate": 0.17114684816687653,
      "loss": 0.0,
      "num_input_tokens_seen": 10336864,
      "step": 18200
    },
    {
      "epoch": 319.3008849557522,
      "eval_loss": 0.6037719249725342,
      "eval_runtime": 0.9305,
      "eval_samples_per_second": 26.868,
      "eval_steps_per_second": 13.971,
      "num_input_tokens_seen": 10336864,
      "step": 18200
    },
    {
      "epoch": 319.3893805309734,
      "grad_norm": 2.157516654222036e-08,
      "learning_rate": 0.17108852998267585,
      "loss": 0.0,
      "num_input_tokens_seen": 10339680,
      "step": 18205
    },
    {
      "epoch": 319.4778761061947,
      "grad_norm": 2.9330811557315428e-08,
      "learning_rate": 0.17103020854635878,
      "loss": 0.0,
      "num_input_tokens_seen": 10342224,
      "step": 18210
    },
    {
      "epoch": 319.56637168141594,
      "grad_norm": 1.686703932080036e-08,
      "learning_rate": 0.1709718838669193,
      "loss": 0.0,
      "num_input_tokens_seen": 10344816,
      "step": 18215
    },
    {
      "epoch": 319.65486725663715,
      "grad_norm": 3.20881312632082e-08,
      "learning_rate": 0.17091355595335173,
      "loss": 0.0,
      "num_input_tokens_seen": 10347456,
      "step": 18220
    },
    {
      "epoch": 319.7433628318584,
      "grad_norm": 2.0023383839884445e-08,
      "learning_rate": 0.17085522481465107,
      "loss": 0.0,
      "num_input_tokens_seen": 10350592,
      "step": 18225
    },
    {
      "epoch": 319.83185840707966,
      "grad_norm": 2.1429780616699645e-08,
      "learning_rate": 0.17079689045981264,
      "loss": 0.0,
      "num_input_tokens_seen": 10353440,
      "step": 18230
    },
    {
      "epoch": 319.92035398230087,
      "grad_norm": 1.6587575757398554e-08,
      "learning_rate": 0.17073855289783238,
      "loss": 0.0,
      "num_input_tokens_seen": 10356976,
      "step": 18235
    },
    {
      "epoch": 320.0,
      "grad_norm": 1.5252583196456726e-08,
      "learning_rate": 0.1706802121377066,
      "loss": 0.0,
      "num_input_tokens_seen": 10359368,
      "step": 18240
    },
    {
      "epoch": 320.08849557522126,
      "grad_norm": 2.9072761975612593e-08,
      "learning_rate": 0.17062186818843225,
      "loss": 0.0,
      "num_input_tokens_seen": 10362504,
      "step": 18245
    },
    {
      "epoch": 320.17699115044246,
      "grad_norm": 1.779663705292478e-08,
      "learning_rate": 0.17056352105900668,
      "loss": 0.0,
      "num_input_tokens_seen": 10364920,
      "step": 18250
    },
    {
      "epoch": 320.2654867256637,
      "grad_norm": 4.430547662082063e-08,
      "learning_rate": 0.17050517075842772,
      "loss": 0.0,
      "num_input_tokens_seen": 10368104,
      "step": 18255
    },
    {
      "epoch": 320.353982300885,
      "grad_norm": 5.272287495472483e-08,
      "learning_rate": 0.17044681729569375,
      "loss": 0.0,
      "num_input_tokens_seen": 10370952,
      "step": 18260
    },
    {
      "epoch": 320.4424778761062,
      "grad_norm": 2.184851233266727e-08,
      "learning_rate": 0.17038846067980365,
      "loss": 0.0,
      "num_input_tokens_seen": 10373432,
      "step": 18265
    },
    {
      "epoch": 320.53097345132744,
      "grad_norm": 8.326046518902785e-09,
      "learning_rate": 0.17033010091975664,
      "loss": 0.0,
      "num_input_tokens_seen": 10376280,
      "step": 18270
    },
    {
      "epoch": 320.6194690265487,
      "grad_norm": 4.247291585102175e-08,
      "learning_rate": 0.17027173802455262,
      "loss": 0.0,
      "num_input_tokens_seen": 10378824,
      "step": 18275
    },
    {
      "epoch": 320.7079646017699,
      "grad_norm": 2.9208614193976246e-08,
      "learning_rate": 0.1702133720031918,
      "loss": 0.0,
      "num_input_tokens_seen": 10381816,
      "step": 18280
    },
    {
      "epoch": 320.79646017699116,
      "grad_norm": 4.420769172952532e-08,
      "learning_rate": 0.17015500286467503,
      "loss": 0.0,
      "num_input_tokens_seen": 10384568,
      "step": 18285
    },
    {
      "epoch": 320.88495575221236,
      "grad_norm": 2.2921271991549474e-08,
      "learning_rate": 0.17009663061800354,
      "loss": 0.0,
      "num_input_tokens_seen": 10387528,
      "step": 18290
    },
    {
      "epoch": 320.9734513274336,
      "grad_norm": 2.6509443529221244e-08,
      "learning_rate": 0.17003825527217903,
      "loss": 0.0,
      "num_input_tokens_seen": 10390376,
      "step": 18295
    },
    {
      "epoch": 321.05309734513276,
      "grad_norm": 4.469436021281581e-08,
      "learning_rate": 0.16997987683620377,
      "loss": 0.0,
      "num_input_tokens_seen": 10393016,
      "step": 18300
    },
    {
      "epoch": 321.14159292035396,
      "grad_norm": 2.14803232978511e-08,
      "learning_rate": 0.16992149531908043,
      "loss": 0.0,
      "num_input_tokens_seen": 10396008,
      "step": 18305
    },
    {
      "epoch": 321.2300884955752,
      "grad_norm": 3.721454788774281e-08,
      "learning_rate": 0.16986311072981214,
      "loss": 0.0,
      "num_input_tokens_seen": 10399080,
      "step": 18310
    },
    {
      "epoch": 321.3185840707965,
      "grad_norm": 4.749123760916518e-08,
      "learning_rate": 0.16980472307740255,
      "loss": 0.0,
      "num_input_tokens_seen": 10402072,
      "step": 18315
    },
    {
      "epoch": 321.4070796460177,
      "grad_norm": 2.9910861343296347e-08,
      "learning_rate": 0.1697463323708558,
      "loss": 0.0,
      "num_input_tokens_seen": 10405032,
      "step": 18320
    },
    {
      "epoch": 321.49557522123894,
      "grad_norm": 3.643105372930222e-08,
      "learning_rate": 0.16968793861917641,
      "loss": 0.0,
      "num_input_tokens_seen": 10407528,
      "step": 18325
    },
    {
      "epoch": 321.5840707964602,
      "grad_norm": 2.713463942427552e-08,
      "learning_rate": 0.16962954183136952,
      "loss": 0.0,
      "num_input_tokens_seen": 10410184,
      "step": 18330
    },
    {
      "epoch": 321.6725663716814,
      "grad_norm": 2.501503182372744e-08,
      "learning_rate": 0.16957114201644058,
      "loss": 0.0,
      "num_input_tokens_seen": 10413288,
      "step": 18335
    },
    {
      "epoch": 321.76106194690266,
      "grad_norm": 1.3412101651510966e-08,
      "learning_rate": 0.16951273918339563,
      "loss": 0.0,
      "num_input_tokens_seen": 10416376,
      "step": 18340
    },
    {
      "epoch": 321.8495575221239,
      "grad_norm": 3.0019773333833655e-08,
      "learning_rate": 0.16945433334124105,
      "loss": 0.0,
      "num_input_tokens_seen": 10419656,
      "step": 18345
    },
    {
      "epoch": 321.9380530973451,
      "grad_norm": 2.1327151600303296e-08,
      "learning_rate": 0.1693959244989838,
      "loss": 0.0,
      "num_input_tokens_seen": 10422376,
      "step": 18350
    },
    {
      "epoch": 322.01769911504425,
      "grad_norm": 3.289980909926271e-08,
      "learning_rate": 0.16933751266563127,
      "loss": 0.0,
      "num_input_tokens_seen": 10424320,
      "step": 18355
    },
    {
      "epoch": 322.1061946902655,
      "grad_norm": 1.595761034423049e-08,
      "learning_rate": 0.16927909785019118,
      "loss": 0.0,
      "num_input_tokens_seen": 10427376,
      "step": 18360
    },
    {
      "epoch": 322.1946902654867,
      "grad_norm": 1.6576729322537176e-08,
      "learning_rate": 0.169220680061672,
      "loss": 0.0,
      "num_input_tokens_seen": 10430240,
      "step": 18365
    },
    {
      "epoch": 322.283185840708,
      "grad_norm": 2.0480182882920417e-08,
      "learning_rate": 0.16916225930908244,
      "loss": 0.0,
      "num_input_tokens_seen": 10433248,
      "step": 18370
    },
    {
      "epoch": 322.37168141592923,
      "grad_norm": 1.5170508405049077e-08,
      "learning_rate": 0.16910383560143163,
      "loss": 0.0,
      "num_input_tokens_seen": 10436400,
      "step": 18375
    },
    {
      "epoch": 322.46017699115043,
      "grad_norm": 1.777240754563536e-08,
      "learning_rate": 0.16904540894772935,
      "loss": 0.0,
      "num_input_tokens_seen": 10439600,
      "step": 18380
    },
    {
      "epoch": 322.5486725663717,
      "grad_norm": 2.9498496090241133e-08,
      "learning_rate": 0.16898697935698562,
      "loss": 0.0,
      "num_input_tokens_seen": 10442448,
      "step": 18385
    },
    {
      "epoch": 322.6371681415929,
      "grad_norm": 1.5407220388397036e-08,
      "learning_rate": 0.1689285468382111,
      "loss": 0.0,
      "num_input_tokens_seen": 10444864,
      "step": 18390
    },
    {
      "epoch": 322.72566371681415,
      "grad_norm": 1.279815986521271e-08,
      "learning_rate": 0.16887011140041677,
      "loss": 0.0,
      "num_input_tokens_seen": 10447424,
      "step": 18395
    },
    {
      "epoch": 322.8141592920354,
      "grad_norm": 2.8369344207135327e-08,
      "learning_rate": 0.1688116730526141,
      "loss": 0.0,
      "num_input_tokens_seen": 10450688,
      "step": 18400
    },
    {
      "epoch": 322.8141592920354,
      "eval_loss": 0.6033881902694702,
      "eval_runtime": 0.9271,
      "eval_samples_per_second": 26.965,
      "eval_steps_per_second": 14.022,
      "num_input_tokens_seen": 10450688,
      "step": 18400
    },
    {
      "epoch": 322.9026548672566,
      "grad_norm": 1.8517864575073872e-08,
      "learning_rate": 0.1687532318038151,
      "loss": 0.0,
      "num_input_tokens_seen": 10453200,
      "step": 18405
    },
    {
      "epoch": 322.9911504424779,
      "grad_norm": 8.757428560102198e-09,
      "learning_rate": 0.16869478766303206,
      "loss": 0.0,
      "num_input_tokens_seen": 10456016,
      "step": 18410
    },
    {
      "epoch": 323.070796460177,
      "grad_norm": 3.0667496986325204e-08,
      "learning_rate": 0.16863634063927788,
      "loss": 0.0,
      "num_input_tokens_seen": 10458352,
      "step": 18415
    },
    {
      "epoch": 323.1592920353982,
      "grad_norm": 1.9337992540613413e-08,
      "learning_rate": 0.16857789074156568,
      "loss": 0.0,
      "num_input_tokens_seen": 10461456,
      "step": 18420
    },
    {
      "epoch": 323.24778761061947,
      "grad_norm": 2.2829510726296576e-08,
      "learning_rate": 0.16851943797890928,
      "loss": 0.0,
      "num_input_tokens_seen": 10464256,
      "step": 18425
    },
    {
      "epoch": 323.3362831858407,
      "grad_norm": 2.4196689096811497e-08,
      "learning_rate": 0.16846098236032284,
      "loss": 0.0,
      "num_input_tokens_seen": 10467488,
      "step": 18430
    },
    {
      "epoch": 323.42477876106193,
      "grad_norm": 3.079833277297439e-08,
      "learning_rate": 0.16840252389482097,
      "loss": 0.0,
      "num_input_tokens_seen": 10470112,
      "step": 18435
    },
    {
      "epoch": 323.5132743362832,
      "grad_norm": 6.676538788497055e-08,
      "learning_rate": 0.16834406259141857,
      "loss": 0.0,
      "num_input_tokens_seen": 10473312,
      "step": 18440
    },
    {
      "epoch": 323.60176991150445,
      "grad_norm": 2.845042956778343e-08,
      "learning_rate": 0.16828559845913124,
      "loss": 0.0,
      "num_input_tokens_seen": 10476192,
      "step": 18445
    },
    {
      "epoch": 323.69026548672565,
      "grad_norm": 1.6741799058195284e-08,
      "learning_rate": 0.16822713150697488,
      "loss": 0.0,
      "num_input_tokens_seen": 10478960,
      "step": 18450
    },
    {
      "epoch": 323.7787610619469,
      "grad_norm": 2.862888415222642e-08,
      "learning_rate": 0.16816866174396575,
      "loss": 0.0,
      "num_input_tokens_seen": 10481904,
      "step": 18455
    },
    {
      "epoch": 323.86725663716817,
      "grad_norm": 2.8268834384448382e-08,
      "learning_rate": 0.16811018917912057,
      "loss": 0.0,
      "num_input_tokens_seen": 10484672,
      "step": 18460
    },
    {
      "epoch": 323.95575221238937,
      "grad_norm": 2.943059307369822e-08,
      "learning_rate": 0.16805171382145673,
      "loss": 0.0,
      "num_input_tokens_seen": 10487344,
      "step": 18465
    },
    {
      "epoch": 324.0353982300885,
      "grad_norm": 5.1053955729685185e-08,
      "learning_rate": 0.16799323567999175,
      "loss": 0.0,
      "num_input_tokens_seen": 10489664,
      "step": 18470
    },
    {
      "epoch": 324.12389380530976,
      "grad_norm": 1.8258516476521436e-08,
      "learning_rate": 0.16793475476374367,
      "loss": 0.0,
      "num_input_tokens_seen": 10492464,
      "step": 18475
    },
    {
      "epoch": 324.21238938053096,
      "grad_norm": 3.0729133015938714e-08,
      "learning_rate": 0.1678762710817311,
      "loss": 0.0,
      "num_input_tokens_seen": 10495344,
      "step": 18480
    },
    {
      "epoch": 324.3008849557522,
      "grad_norm": 2.1398889771262475e-08,
      "learning_rate": 0.1678177846429728,
      "loss": 0.0,
      "num_input_tokens_seen": 10497984,
      "step": 18485
    },
    {
      "epoch": 324.3893805309734,
      "grad_norm": 1.4857591601469267e-08,
      "learning_rate": 0.16775929545648827,
      "loss": 0.0,
      "num_input_tokens_seen": 10501120,
      "step": 18490
    },
    {
      "epoch": 324.4778761061947,
      "grad_norm": 3.281862248627476e-08,
      "learning_rate": 0.16770080353129715,
      "loss": 0.0,
      "num_input_tokens_seen": 10503968,
      "step": 18495
    },
    {
      "epoch": 324.56637168141594,
      "grad_norm": 2.8873810009599765e-08,
      "learning_rate": 0.16764230887641968,
      "loss": 0.0,
      "num_input_tokens_seen": 10506880,
      "step": 18500
    },
    {
      "epoch": 324.65486725663715,
      "grad_norm": 2.3578916596989075e-08,
      "learning_rate": 0.1675838115008765,
      "loss": 0.0,
      "num_input_tokens_seen": 10509520,
      "step": 18505
    },
    {
      "epoch": 324.7433628318584,
      "grad_norm": 3.416533900235663e-08,
      "learning_rate": 0.1675253114136886,
      "loss": 0.0,
      "num_input_tokens_seen": 10512192,
      "step": 18510
    },
    {
      "epoch": 324.83185840707966,
      "grad_norm": 1.9839870191162845e-08,
      "learning_rate": 0.16746680862387747,
      "loss": 0.0,
      "num_input_tokens_seen": 10514640,
      "step": 18515
    },
    {
      "epoch": 324.92035398230087,
      "grad_norm": 3.1701414826557084e-08,
      "learning_rate": 0.16740830314046493,
      "loss": 0.0,
      "num_input_tokens_seen": 10517840,
      "step": 18520
    },
    {
      "epoch": 325.0,
      "grad_norm": 1.905334201524056e-08,
      "learning_rate": 0.1673497949724733,
      "loss": 0.0,
      "num_input_tokens_seen": 10520464,
      "step": 18525
    },
    {
      "epoch": 325.08849557522126,
      "grad_norm": 1.7232489213370172e-08,
      "learning_rate": 0.16729128412892522,
      "loss": 0.0,
      "num_input_tokens_seen": 10523584,
      "step": 18530
    },
    {
      "epoch": 325.17699115044246,
      "grad_norm": 1.5284872034726504e-08,
      "learning_rate": 0.16723277061884384,
      "loss": 0.0,
      "num_input_tokens_seen": 10526144,
      "step": 18535
    },
    {
      "epoch": 325.2654867256637,
      "grad_norm": 4.376538598194202e-08,
      "learning_rate": 0.16717425445125267,
      "loss": 0.0,
      "num_input_tokens_seen": 10528656,
      "step": 18540
    },
    {
      "epoch": 325.353982300885,
      "grad_norm": 1.6088705478978227e-08,
      "learning_rate": 0.16711573563517565,
      "loss": 0.0,
      "num_input_tokens_seen": 10531824,
      "step": 18545
    },
    {
      "epoch": 325.4424778761062,
      "grad_norm": 3.7275132314107395e-08,
      "learning_rate": 0.1670572141796371,
      "loss": 0.0,
      "num_input_tokens_seen": 10534576,
      "step": 18550
    },
    {
      "epoch": 325.53097345132744,
      "grad_norm": 4.1745451540009526e-08,
      "learning_rate": 0.16699869009366175,
      "loss": 0.0,
      "num_input_tokens_seen": 10537520,
      "step": 18555
    },
    {
      "epoch": 325.6194690265487,
      "grad_norm": 2.419516675900013e-08,
      "learning_rate": 0.1669401633862748,
      "loss": 0.0,
      "num_input_tokens_seen": 10540640,
      "step": 18560
    },
    {
      "epoch": 325.7079646017699,
      "grad_norm": 3.436367279618935e-08,
      "learning_rate": 0.16688163406650178,
      "loss": 0.0,
      "num_input_tokens_seen": 10543808,
      "step": 18565
    },
    {
      "epoch": 325.79646017699116,
      "grad_norm": 3.65207988295424e-08,
      "learning_rate": 0.1668231021433686,
      "loss": 0.0,
      "num_input_tokens_seen": 10546896,
      "step": 18570
    },
    {
      "epoch": 325.88495575221236,
      "grad_norm": 3.108631929649164e-08,
      "learning_rate": 0.1667645676259017,
      "loss": 0.0,
      "num_input_tokens_seen": 10549776,
      "step": 18575
    },
    {
      "epoch": 325.9734513274336,
      "grad_norm": 1.7208764191423143e-08,
      "learning_rate": 0.1667060305231277,
      "loss": 0.0,
      "num_input_tokens_seen": 10552192,
      "step": 18580
    },
    {
      "epoch": 326.05309734513276,
      "grad_norm": 3.619719635139518e-08,
      "learning_rate": 0.16664749084407396,
      "loss": 0.0,
      "num_input_tokens_seen": 10554360,
      "step": 18585
    },
    {
      "epoch": 326.14159292035396,
      "grad_norm": 3.5454466029705145e-08,
      "learning_rate": 0.16658894859776788,
      "loss": 0.0,
      "num_input_tokens_seen": 10557224,
      "step": 18590
    },
    {
      "epoch": 326.2300884955752,
      "grad_norm": 3.0140228091113386e-08,
      "learning_rate": 0.16653040379323752,
      "loss": 0.0,
      "num_input_tokens_seen": 10559768,
      "step": 18595
    },
    {
      "epoch": 326.3185840707965,
      "grad_norm": 2.39052919681626e-08,
      "learning_rate": 0.16647185643951107,
      "loss": 0.0,
      "num_input_tokens_seen": 10563128,
      "step": 18600
    },
    {
      "epoch": 326.3185840707965,
      "eval_loss": 0.6052086353302002,
      "eval_runtime": 0.9402,
      "eval_samples_per_second": 26.589,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 10563128,
      "step": 18600
    },
    {
      "epoch": 326.4070796460177,
      "grad_norm": 7.47773576392774e-09,
      "learning_rate": 0.1664133065456174,
      "loss": 0.0,
      "num_input_tokens_seen": 10566088,
      "step": 18605
    },
    {
      "epoch": 326.49557522123894,
      "grad_norm": 2.7061195950750516e-08,
      "learning_rate": 0.1663547541205856,
      "loss": 0.0,
      "num_input_tokens_seen": 10569016,
      "step": 18610
    },
    {
      "epoch": 326.5840707964602,
      "grad_norm": 3.464355557980525e-08,
      "learning_rate": 0.16629619917344518,
      "loss": 0.0,
      "num_input_tokens_seen": 10572008,
      "step": 18615
    },
    {
      "epoch": 326.6725663716814,
      "grad_norm": 2.0922890087149426e-08,
      "learning_rate": 0.16623764171322605,
      "loss": 0.0,
      "num_input_tokens_seen": 10574696,
      "step": 18620
    },
    {
      "epoch": 326.76106194690266,
      "grad_norm": 4.027326028221978e-08,
      "learning_rate": 0.1661790817489585,
      "loss": 0.0,
      "num_input_tokens_seen": 10577720,
      "step": 18625
    },
    {
      "epoch": 326.8495575221239,
      "grad_norm": 2.403479015811172e-08,
      "learning_rate": 0.16612051928967328,
      "loss": 0.0,
      "num_input_tokens_seen": 10580696,
      "step": 18630
    },
    {
      "epoch": 326.9380530973451,
      "grad_norm": 5.503138567064525e-08,
      "learning_rate": 0.16606195434440138,
      "loss": 0.0,
      "num_input_tokens_seen": 10583208,
      "step": 18635
    },
    {
      "epoch": 327.01769911504425,
      "grad_norm": 2.0568158731748554e-08,
      "learning_rate": 0.16600338692217426,
      "loss": 0.0,
      "num_input_tokens_seen": 10586016,
      "step": 18640
    },
    {
      "epoch": 327.1061946902655,
      "grad_norm": 2.5329068620294493e-08,
      "learning_rate": 0.16594481703202374,
      "loss": 0.0,
      "num_input_tokens_seen": 10588768,
      "step": 18645
    },
    {
      "epoch": 327.1946902654867,
      "grad_norm": 2.5592937547003203e-08,
      "learning_rate": 0.1658862446829821,
      "loss": 0.0,
      "num_input_tokens_seen": 10592224,
      "step": 18650
    },
    {
      "epoch": 327.283185840708,
      "grad_norm": 2.4911503970770354e-08,
      "learning_rate": 0.16582766988408187,
      "loss": 0.0,
      "num_input_tokens_seen": 10595264,
      "step": 18655
    },
    {
      "epoch": 327.37168141592923,
      "grad_norm": 2.9429919834456086e-08,
      "learning_rate": 0.16576909264435608,
      "loss": 0.0,
      "num_input_tokens_seen": 10597792,
      "step": 18660
    },
    {
      "epoch": 327.46017699115043,
      "grad_norm": 3.623468813884756e-08,
      "learning_rate": 0.16571051297283798,
      "loss": 0.0,
      "num_input_tokens_seen": 10600688,
      "step": 18665
    },
    {
      "epoch": 327.5486725663717,
      "grad_norm": 3.173963136759994e-08,
      "learning_rate": 0.16565193087856137,
      "loss": 0.0,
      "num_input_tokens_seen": 10603504,
      "step": 18670
    },
    {
      "epoch": 327.6371681415929,
      "grad_norm": 1.4884436794204703e-08,
      "learning_rate": 0.16559334637056033,
      "loss": 0.0,
      "num_input_tokens_seen": 10606112,
      "step": 18675
    },
    {
      "epoch": 327.72566371681415,
      "grad_norm": 2.6731184377126738e-08,
      "learning_rate": 0.16553475945786933,
      "loss": 0.0,
      "num_input_tokens_seen": 10609136,
      "step": 18680
    },
    {
      "epoch": 327.8141592920354,
      "grad_norm": 2.9496682429908105e-08,
      "learning_rate": 0.16547617014952318,
      "loss": 0.0,
      "num_input_tokens_seen": 10611664,
      "step": 18685
    },
    {
      "epoch": 327.9026548672566,
      "grad_norm": 1.2481264022312644e-08,
      "learning_rate": 0.1654175784545571,
      "loss": 0.0,
      "num_input_tokens_seen": 10614624,
      "step": 18690
    },
    {
      "epoch": 327.9911504424779,
      "grad_norm": 3.401573422934234e-08,
      "learning_rate": 0.1653589843820067,
      "loss": 0.0,
      "num_input_tokens_seen": 10618032,
      "step": 18695
    },
    {
      "epoch": 328.070796460177,
      "grad_norm": 2.8303857035893998e-08,
      "learning_rate": 0.1653003879409079,
      "loss": 0.0,
      "num_input_tokens_seen": 10619992,
      "step": 18700
    },
    {
      "epoch": 328.1592920353982,
      "grad_norm": 1.689766548906846e-08,
      "learning_rate": 0.165241789140297,
      "loss": 0.0,
      "num_input_tokens_seen": 10622616,
      "step": 18705
    },
    {
      "epoch": 328.24778761061947,
      "grad_norm": 4.515500862112276e-08,
      "learning_rate": 0.16518318798921064,
      "loss": 0.0,
      "num_input_tokens_seen": 10625608,
      "step": 18710
    },
    {
      "epoch": 328.3362831858407,
      "grad_norm": 3.6354894206169774e-08,
      "learning_rate": 0.16512458449668593,
      "loss": 0.0,
      "num_input_tokens_seen": 10629160,
      "step": 18715
    },
    {
      "epoch": 328.42477876106193,
      "grad_norm": 2.0987142690387373e-08,
      "learning_rate": 0.1650659786717602,
      "loss": 0.0,
      "num_input_tokens_seen": 10632072,
      "step": 18720
    },
    {
      "epoch": 328.5132743362832,
      "grad_norm": 1.5597187541516178e-08,
      "learning_rate": 0.1650073705234712,
      "loss": 0.0,
      "num_input_tokens_seen": 10634872,
      "step": 18725
    },
    {
      "epoch": 328.60176991150445,
      "grad_norm": 2.249777786289542e-08,
      "learning_rate": 0.16494876006085712,
      "loss": 0.0,
      "num_input_tokens_seen": 10637656,
      "step": 18730
    },
    {
      "epoch": 328.69026548672565,
      "grad_norm": 5.32590460622373e-09,
      "learning_rate": 0.16489014729295634,
      "loss": 0.0,
      "num_input_tokens_seen": 10640456,
      "step": 18735
    },
    {
      "epoch": 328.7787610619469,
      "grad_norm": 9.699520298056541e-09,
      "learning_rate": 0.16483153222880775,
      "loss": 0.0,
      "num_input_tokens_seen": 10643512,
      "step": 18740
    },
    {
      "epoch": 328.86725663716817,
      "grad_norm": 2.450302361012291e-08,
      "learning_rate": 0.16477291487745052,
      "loss": 0.0,
      "num_input_tokens_seen": 10646280,
      "step": 18745
    },
    {
      "epoch": 328.95575221238937,
      "grad_norm": 2.507936613938e-08,
      "learning_rate": 0.16471429524792416,
      "loss": 0.0,
      "num_input_tokens_seen": 10649288,
      "step": 18750
    },
    {
      "epoch": 329.0353982300885,
      "grad_norm": 2.77280101101951e-08,
      "learning_rate": 0.16465567334926856,
      "loss": 0.0,
      "num_input_tokens_seen": 10651880,
      "step": 18755
    },
    {
      "epoch": 329.12389380530976,
      "grad_norm": 2.1497992719332615e-08,
      "learning_rate": 0.16459704919052395,
      "loss": 0.0,
      "num_input_tokens_seen": 10654648,
      "step": 18760
    },
    {
      "epoch": 329.21238938053096,
      "grad_norm": 1.4224113442651287e-08,
      "learning_rate": 0.16453842278073086,
      "loss": 0.0,
      "num_input_tokens_seen": 10657496,
      "step": 18765
    },
    {
      "epoch": 329.3008849557522,
      "grad_norm": 2.60869779111772e-08,
      "learning_rate": 0.16447979412893038,
      "loss": 0.0,
      "num_input_tokens_seen": 10660216,
      "step": 18770
    },
    {
      "epoch": 329.3893805309734,
      "grad_norm": 3.541189386169208e-08,
      "learning_rate": 0.16442116324416367,
      "loss": 0.0,
      "num_input_tokens_seen": 10663112,
      "step": 18775
    },
    {
      "epoch": 329.4778761061947,
      "grad_norm": 1.625785017722592e-08,
      "learning_rate": 0.1643625301354723,
      "loss": 0.0,
      "num_input_tokens_seen": 10665928,
      "step": 18780
    },
    {
      "epoch": 329.56637168141594,
      "grad_norm": 9.734854700127471e-09,
      "learning_rate": 0.16430389481189828,
      "loss": 0.0,
      "num_input_tokens_seen": 10669080,
      "step": 18785
    },
    {
      "epoch": 329.65486725663715,
      "grad_norm": 2.31233077130355e-08,
      "learning_rate": 0.164245257282484,
      "loss": 0.0,
      "num_input_tokens_seen": 10672296,
      "step": 18790
    },
    {
      "epoch": 329.7433628318584,
      "grad_norm": 8.971987597305997e-09,
      "learning_rate": 0.16418661755627195,
      "loss": 0.0,
      "num_input_tokens_seen": 10674904,
      "step": 18795
    },
    {
      "epoch": 329.83185840707966,
      "grad_norm": 6.985271738813026e-08,
      "learning_rate": 0.16412797564230527,
      "loss": 0.0,
      "num_input_tokens_seen": 10677928,
      "step": 18800
    },
    {
      "epoch": 329.83185840707966,
      "eval_loss": 0.6042895317077637,
      "eval_runtime": 0.9431,
      "eval_samples_per_second": 26.508,
      "eval_steps_per_second": 13.784,
      "num_input_tokens_seen": 10677928,
      "step": 18800
    },
    {
      "epoch": 329.92035398230087,
      "grad_norm": 2.2507819608108548e-08,
      "learning_rate": 0.16406933154962713,
      "loss": 0.0,
      "num_input_tokens_seen": 10681080,
      "step": 18805
    },
    {
      "epoch": 330.0,
      "grad_norm": 8.662195938313744e-08,
      "learning_rate": 0.16401068528728133,
      "loss": 0.0,
      "num_input_tokens_seen": 10683296,
      "step": 18810
    },
    {
      "epoch": 330.08849557522126,
      "grad_norm": 1.6302728056416527e-08,
      "learning_rate": 0.16395203686431173,
      "loss": 0.0,
      "num_input_tokens_seen": 10686096,
      "step": 18815
    },
    {
      "epoch": 330.17699115044246,
      "grad_norm": 2.054228609438269e-08,
      "learning_rate": 0.16389338628976277,
      "loss": 0.0,
      "num_input_tokens_seen": 10689152,
      "step": 18820
    },
    {
      "epoch": 330.2654867256637,
      "grad_norm": 2.999177795004471e-08,
      "learning_rate": 0.163834733572679,
      "loss": 0.0,
      "num_input_tokens_seen": 10691856,
      "step": 18825
    },
    {
      "epoch": 330.353982300885,
      "grad_norm": 1.6097574828677352e-08,
      "learning_rate": 0.16377607872210545,
      "loss": 0.0,
      "num_input_tokens_seen": 10694768,
      "step": 18830
    },
    {
      "epoch": 330.4424778761062,
      "grad_norm": 3.54420386372567e-08,
      "learning_rate": 0.16371742174708748,
      "loss": 0.0,
      "num_input_tokens_seen": 10697328,
      "step": 18835
    },
    {
      "epoch": 330.53097345132744,
      "grad_norm": 2.2753216200044335e-08,
      "learning_rate": 0.16365876265667065,
      "loss": 0.0,
      "num_input_tokens_seen": 10699616,
      "step": 18840
    },
    {
      "epoch": 330.6194690265487,
      "grad_norm": 2.75057594478767e-08,
      "learning_rate": 0.163600101459901,
      "loss": 0.0,
      "num_input_tokens_seen": 10702736,
      "step": 18845
    },
    {
      "epoch": 330.7079646017699,
      "grad_norm": 1.7950901209928816e-08,
      "learning_rate": 0.16354143816582484,
      "loss": 0.0,
      "num_input_tokens_seen": 10705728,
      "step": 18850
    },
    {
      "epoch": 330.79646017699116,
      "grad_norm": 9.76764447102596e-09,
      "learning_rate": 0.1634827727834887,
      "loss": 0.0,
      "num_input_tokens_seen": 10708784,
      "step": 18855
    },
    {
      "epoch": 330.88495575221236,
      "grad_norm": 8.428438391661075e-09,
      "learning_rate": 0.16342410532193954,
      "loss": 0.0,
      "num_input_tokens_seen": 10711296,
      "step": 18860
    },
    {
      "epoch": 330.9734513274336,
      "grad_norm": 1.0533491412445528e-08,
      "learning_rate": 0.16336543579022464,
      "loss": 0.0,
      "num_input_tokens_seen": 10714576,
      "step": 18865
    },
    {
      "epoch": 331.05309734513276,
      "grad_norm": 2.7337248909020673e-08,
      "learning_rate": 0.16330676419739157,
      "loss": 0.0,
      "num_input_tokens_seen": 10717048,
      "step": 18870
    },
    {
      "epoch": 331.14159292035396,
      "grad_norm": 4.3176875408335036e-08,
      "learning_rate": 0.1632480905524883,
      "loss": 0.0,
      "num_input_tokens_seen": 10719976,
      "step": 18875
    },
    {
      "epoch": 331.2300884955752,
      "grad_norm": 1.4761477373781418e-08,
      "learning_rate": 0.16318941486456293,
      "loss": 0.0,
      "num_input_tokens_seen": 10722696,
      "step": 18880
    },
    {
      "epoch": 331.3185840707965,
      "grad_norm": 1.800334459289843e-08,
      "learning_rate": 0.16313073714266405,
      "loss": 0.0,
      "num_input_tokens_seen": 10725816,
      "step": 18885
    },
    {
      "epoch": 331.4070796460177,
      "grad_norm": 2.4001693077480013e-08,
      "learning_rate": 0.16307205739584052,
      "loss": 0.0,
      "num_input_tokens_seen": 10728728,
      "step": 18890
    },
    {
      "epoch": 331.49557522123894,
      "grad_norm": 2.090265383003498e-08,
      "learning_rate": 0.16301337563314144,
      "loss": 0.0,
      "num_input_tokens_seen": 10732424,
      "step": 18895
    },
    {
      "epoch": 331.5840707964602,
      "grad_norm": 2.8104771843118215e-08,
      "learning_rate": 0.1629546918636163,
      "loss": 0.0,
      "num_input_tokens_seen": 10735064,
      "step": 18900
    },
    {
      "epoch": 331.6725663716814,
      "grad_norm": 1.9228640013579934e-08,
      "learning_rate": 0.16289600609631485,
      "loss": 0.0,
      "num_input_tokens_seen": 10737944,
      "step": 18905
    },
    {
      "epoch": 331.76106194690266,
      "grad_norm": 1.2864041387672387e-08,
      "learning_rate": 0.16283731834028722,
      "loss": 0.0,
      "num_input_tokens_seen": 10740568,
      "step": 18910
    },
    {
      "epoch": 331.8495575221239,
      "grad_norm": 1.2928751402796479e-08,
      "learning_rate": 0.16277862860458378,
      "loss": 0.0,
      "num_input_tokens_seen": 10743384,
      "step": 18915
    },
    {
      "epoch": 331.9380530973451,
      "grad_norm": 3.7404952024644444e-08,
      "learning_rate": 0.16271993689825526,
      "loss": 0.0,
      "num_input_tokens_seen": 10745736,
      "step": 18920
    },
    {
      "epoch": 332.01769911504425,
      "grad_norm": 1.4463913622364544e-08,
      "learning_rate": 0.1626612432303526,
      "loss": 0.0,
      "num_input_tokens_seen": 10748000,
      "step": 18925
    },
    {
      "epoch": 332.1061946902655,
      "grad_norm": 2.20449081211882e-08,
      "learning_rate": 0.1626025476099271,
      "loss": 0.0,
      "num_input_tokens_seen": 10750432,
      "step": 18930
    },
    {
      "epoch": 332.1946902654867,
      "grad_norm": 2.639871254928039e-08,
      "learning_rate": 0.1625438500460304,
      "loss": 0.0,
      "num_input_tokens_seen": 10753488,
      "step": 18935
    },
    {
      "epoch": 332.283185840708,
      "grad_norm": 2.6042364709155663e-08,
      "learning_rate": 0.16248515054771442,
      "loss": 0.0,
      "num_input_tokens_seen": 10756592,
      "step": 18940
    },
    {
      "epoch": 332.37168141592923,
      "grad_norm": 1.2616030886647422e-08,
      "learning_rate": 0.16242644912403123,
      "loss": 0.0,
      "num_input_tokens_seen": 10759424,
      "step": 18945
    },
    {
      "epoch": 332.46017699115043,
      "grad_norm": 1.9444636123466807e-08,
      "learning_rate": 0.1623677457840335,
      "loss": 0.0,
      "num_input_tokens_seen": 10762192,
      "step": 18950
    },
    {
      "epoch": 332.5486725663717,
      "grad_norm": 1.3134291876326643e-08,
      "learning_rate": 0.16230904053677397,
      "loss": 0.0,
      "num_input_tokens_seen": 10764912,
      "step": 18955
    },
    {
      "epoch": 332.6371681415929,
      "grad_norm": 1.075432098929241e-08,
      "learning_rate": 0.16225033339130568,
      "loss": 0.0,
      "num_input_tokens_seen": 10767872,
      "step": 18960
    },
    {
      "epoch": 332.72566371681415,
      "grad_norm": 2.923427722123506e-08,
      "learning_rate": 0.16219162435668197,
      "loss": 0.0,
      "num_input_tokens_seen": 10771152,
      "step": 18965
    },
    {
      "epoch": 332.8141592920354,
      "grad_norm": 4.91202420960235e-08,
      "learning_rate": 0.16213291344195666,
      "loss": 0.0,
      "num_input_tokens_seen": 10774000,
      "step": 18970
    },
    {
      "epoch": 332.9026548672566,
      "grad_norm": 2.3349350897206023e-08,
      "learning_rate": 0.16207420065618358,
      "loss": 0.0,
      "num_input_tokens_seen": 10776688,
      "step": 18975
    },
    {
      "epoch": 332.9911504424779,
      "grad_norm": 2.0367577846513996e-08,
      "learning_rate": 0.16201548600841706,
      "loss": 0.0,
      "num_input_tokens_seen": 10779600,
      "step": 18980
    },
    {
      "epoch": 333.070796460177,
      "grad_norm": 1.9519244887078457e-08,
      "learning_rate": 0.16195676950771154,
      "loss": 0.0,
      "num_input_tokens_seen": 10781936,
      "step": 18985
    },
    {
      "epoch": 333.1592920353982,
      "grad_norm": 1.3237437812563257e-08,
      "learning_rate": 0.16189805116312198,
      "loss": 0.0,
      "num_input_tokens_seen": 10784768,
      "step": 18990
    },
    {
      "epoch": 333.24778761061947,
      "grad_norm": 1.146504224891487e-08,
      "learning_rate": 0.16183933098370337,
      "loss": 0.0,
      "num_input_tokens_seen": 10788048,
      "step": 18995
    },
    {
      "epoch": 333.3362831858407,
      "grad_norm": 1.7447741029741337e-08,
      "learning_rate": 0.16178060897851115,
      "loss": 0.0,
      "num_input_tokens_seen": 10790896,
      "step": 19000
    },
    {
      "epoch": 333.3362831858407,
      "eval_loss": 0.6169363260269165,
      "eval_runtime": 0.9341,
      "eval_samples_per_second": 26.763,
      "eval_steps_per_second": 13.917,
      "num_input_tokens_seen": 10790896,
      "step": 19000
    },
    {
      "epoch": 333.42477876106193,
      "grad_norm": 1.712177954971139e-08,
      "learning_rate": 0.16172188515660096,
      "loss": 0.0,
      "num_input_tokens_seen": 10793696,
      "step": 19005
    },
    {
      "epoch": 333.5132743362832,
      "grad_norm": 1.1515526310290625e-08,
      "learning_rate": 0.16166315952702878,
      "loss": 0.0,
      "num_input_tokens_seen": 10796880,
      "step": 19010
    },
    {
      "epoch": 333.60176991150445,
      "grad_norm": 5.3441581826518814e-08,
      "learning_rate": 0.16160443209885084,
      "loss": 0.0,
      "num_input_tokens_seen": 10799472,
      "step": 19015
    },
    {
      "epoch": 333.69026548672565,
      "grad_norm": 2.322113878960863e-08,
      "learning_rate": 0.16154570288112363,
      "loss": 0.0,
      "num_input_tokens_seen": 10801856,
      "step": 19020
    },
    {
      "epoch": 333.7787610619469,
      "grad_norm": 3.452591812447281e-08,
      "learning_rate": 0.16148697188290395,
      "loss": 0.0,
      "num_input_tokens_seen": 10804864,
      "step": 19025
    },
    {
      "epoch": 333.86725663716817,
      "grad_norm": 4.346452087133912e-08,
      "learning_rate": 0.16142823911324888,
      "loss": 0.0,
      "num_input_tokens_seen": 10807968,
      "step": 19030
    },
    {
      "epoch": 333.95575221238937,
      "grad_norm": 1.8056674377930904e-08,
      "learning_rate": 0.16136950458121568,
      "loss": 0.0,
      "num_input_tokens_seen": 10811104,
      "step": 19035
    },
    {
      "epoch": 334.0353982300885,
      "grad_norm": 9.695314773239261e-09,
      "learning_rate": 0.16131076829586205,
      "loss": 0.0,
      "num_input_tokens_seen": 10813792,
      "step": 19040
    },
    {
      "epoch": 334.12389380530976,
      "grad_norm": 3.456059971540526e-08,
      "learning_rate": 0.1612520302662457,
      "loss": 0.0,
      "num_input_tokens_seen": 10816576,
      "step": 19045
    },
    {
      "epoch": 334.21238938053096,
      "grad_norm": 1.5560857491436764e-08,
      "learning_rate": 0.16119329050142497,
      "loss": 0.0,
      "num_input_tokens_seen": 10819632,
      "step": 19050
    },
    {
      "epoch": 334.3008849557522,
      "grad_norm": 3.725168795654099e-08,
      "learning_rate": 0.16113454901045818,
      "loss": 0.0,
      "num_input_tokens_seen": 10822992,
      "step": 19055
    },
    {
      "epoch": 334.3893805309734,
      "grad_norm": 2.8653236228137757e-08,
      "learning_rate": 0.16107580580240397,
      "loss": 0.0,
      "num_input_tokens_seen": 10825696,
      "step": 19060
    },
    {
      "epoch": 334.4778761061947,
      "grad_norm": 2.8585825262439357e-08,
      "learning_rate": 0.16101706088632134,
      "loss": 0.0,
      "num_input_tokens_seen": 10828432,
      "step": 19065
    },
    {
      "epoch": 334.56637168141594,
      "grad_norm": 1.8479852315067546e-08,
      "learning_rate": 0.16095831427126947,
      "loss": 0.0,
      "num_input_tokens_seen": 10831520,
      "step": 19070
    },
    {
      "epoch": 334.65486725663715,
      "grad_norm": 1.965136320336569e-08,
      "learning_rate": 0.16089956596630783,
      "loss": 0.0,
      "num_input_tokens_seen": 10833904,
      "step": 19075
    },
    {
      "epoch": 334.7433628318584,
      "grad_norm": 1.7765525939239524e-08,
      "learning_rate": 0.16084081598049618,
      "loss": 0.0,
      "num_input_tokens_seen": 10836848,
      "step": 19080
    },
    {
      "epoch": 334.83185840707966,
      "grad_norm": 1.0501519653871583e-08,
      "learning_rate": 0.1607820643228944,
      "loss": 0.0,
      "num_input_tokens_seen": 10839440,
      "step": 19085
    },
    {
      "epoch": 334.92035398230087,
      "grad_norm": 2.1693297824754154e-08,
      "learning_rate": 0.16072331100256285,
      "loss": 0.0,
      "num_input_tokens_seen": 10842144,
      "step": 19090
    },
    {
      "epoch": 335.0,
      "grad_norm": 1.0613177892082604e-08,
      "learning_rate": 0.16066455602856197,
      "loss": 0.0,
      "num_input_tokens_seen": 10844392,
      "step": 19095
    },
    {
      "epoch": 335.08849557522126,
      "grad_norm": 2.5091472011240512e-08,
      "learning_rate": 0.16060579940995257,
      "loss": 0.0,
      "num_input_tokens_seen": 10847320,
      "step": 19100
    },
    {
      "epoch": 335.17699115044246,
      "grad_norm": 3.494075073717795e-08,
      "learning_rate": 0.16054704115579557,
      "loss": 0.0,
      "num_input_tokens_seen": 10850632,
      "step": 19105
    },
    {
      "epoch": 335.2654867256637,
      "grad_norm": 4.471290537821915e-08,
      "learning_rate": 0.1604882812751523,
      "loss": 0.0,
      "num_input_tokens_seen": 10853352,
      "step": 19110
    },
    {
      "epoch": 335.353982300885,
      "grad_norm": 4.138549769550082e-08,
      "learning_rate": 0.16042951977708425,
      "loss": 0.0,
      "num_input_tokens_seen": 10855960,
      "step": 19115
    },
    {
      "epoch": 335.4424778761062,
      "grad_norm": 3.909099177690223e-08,
      "learning_rate": 0.16037075667065318,
      "loss": 0.0,
      "num_input_tokens_seen": 10858568,
      "step": 19120
    },
    {
      "epoch": 335.53097345132744,
      "grad_norm": 1.2282856509671092e-08,
      "learning_rate": 0.1603119919649211,
      "loss": 0.0,
      "num_input_tokens_seen": 10861432,
      "step": 19125
    },
    {
      "epoch": 335.6194690265487,
      "grad_norm": 2.1421820761702293e-08,
      "learning_rate": 0.16025322566895028,
      "loss": 0.0,
      "num_input_tokens_seen": 10864664,
      "step": 19130
    },
    {
      "epoch": 335.7079646017699,
      "grad_norm": 4.600947178801107e-09,
      "learning_rate": 0.16019445779180322,
      "loss": 0.0,
      "num_input_tokens_seen": 10867512,
      "step": 19135
    },
    {
      "epoch": 335.79646017699116,
      "grad_norm": 2.4813884280661114e-08,
      "learning_rate": 0.16013568834254271,
      "loss": 0.0,
      "num_input_tokens_seen": 10870888,
      "step": 19140
    },
    {
      "epoch": 335.88495575221236,
      "grad_norm": 1.515654624029139e-08,
      "learning_rate": 0.1600769173302316,
      "loss": 0.0,
      "num_input_tokens_seen": 10873768,
      "step": 19145
    },
    {
      "epoch": 335.9734513274336,
      "grad_norm": 5.063694175078126e-08,
      "learning_rate": 0.16001814476393322,
      "loss": 0.0,
      "num_input_tokens_seen": 10876248,
      "step": 19150
    },
    {
      "epoch": 336.05309734513276,
      "grad_norm": 2.102964913319738e-08,
      "learning_rate": 0.15995937065271104,
      "loss": 0.0,
      "num_input_tokens_seen": 10878952,
      "step": 19155
    },
    {
      "epoch": 336.14159292035396,
      "grad_norm": 1.4781268653507595e-08,
      "learning_rate": 0.15990059500562873,
      "loss": 0.0,
      "num_input_tokens_seen": 10881896,
      "step": 19160
    },
    {
      "epoch": 336.2300884955752,
      "grad_norm": 1.9366163783729462e-08,
      "learning_rate": 0.15984181783175025,
      "loss": 0.0,
      "num_input_tokens_seen": 10885224,
      "step": 19165
    },
    {
      "epoch": 336.3185840707965,
      "grad_norm": 1.1827260060215394e-08,
      "learning_rate": 0.1597830391401398,
      "loss": 0.0,
      "num_input_tokens_seen": 10888120,
      "step": 19170
    },
    {
      "epoch": 336.4070796460177,
      "grad_norm": 1.7759303361231105e-08,
      "learning_rate": 0.15972425893986178,
      "loss": 0.0,
      "num_input_tokens_seen": 10890808,
      "step": 19175
    },
    {
      "epoch": 336.49557522123894,
      "grad_norm": 2.9346018948217534e-08,
      "learning_rate": 0.15966547723998084,
      "loss": 0.0,
      "num_input_tokens_seen": 10893752,
      "step": 19180
    },
    {
      "epoch": 336.5840707964602,
      "grad_norm": 3.681959626078424e-08,
      "learning_rate": 0.15960669404956176,
      "loss": 0.0,
      "num_input_tokens_seen": 10896536,
      "step": 19185
    },
    {
      "epoch": 336.6725663716814,
      "grad_norm": 1.6344456454930878e-08,
      "learning_rate": 0.1595479093776698,
      "loss": 0.0,
      "num_input_tokens_seen": 10899192,
      "step": 19190
    },
    {
      "epoch": 336.76106194690266,
      "grad_norm": 8.803603179785568e-09,
      "learning_rate": 0.15948912323337022,
      "loss": 0.0,
      "num_input_tokens_seen": 10902008,
      "step": 19195
    },
    {
      "epoch": 336.8495575221239,
      "grad_norm": 1.953277362076733e-08,
      "learning_rate": 0.1594303356257286,
      "loss": 0.0,
      "num_input_tokens_seen": 10904600,
      "step": 19200
    },
    {
      "epoch": 336.8495575221239,
      "eval_loss": 0.6139432787895203,
      "eval_runtime": 0.9369,
      "eval_samples_per_second": 26.683,
      "eval_steps_per_second": 13.875,
      "num_input_tokens_seen": 10904600,
      "step": 19200
    },
    {
      "epoch": 336.9380530973451,
      "grad_norm": 1.0277122264312766e-08,
      "learning_rate": 0.15937154656381072,
      "loss": 0.0,
      "num_input_tokens_seen": 10907928,
      "step": 19205
    },
    {
      "epoch": 337.01769911504425,
      "grad_norm": 1.3423130162948382e-08,
      "learning_rate": 0.15931275605668258,
      "loss": 0.0,
      "num_input_tokens_seen": 10910000,
      "step": 19210
    },
    {
      "epoch": 337.1061946902655,
      "grad_norm": 2.5080714394221104e-08,
      "learning_rate": 0.1592539641134104,
      "loss": 0.0,
      "num_input_tokens_seen": 10912960,
      "step": 19215
    },
    {
      "epoch": 337.1946902654867,
      "grad_norm": 3.823129191005137e-08,
      "learning_rate": 0.1591951707430607,
      "loss": 0.0,
      "num_input_tokens_seen": 10915760,
      "step": 19220
    },
    {
      "epoch": 337.283185840708,
      "grad_norm": 4.2422207968684233e-08,
      "learning_rate": 0.15913637595470007,
      "loss": 0.0,
      "num_input_tokens_seen": 10919184,
      "step": 19225
    },
    {
      "epoch": 337.37168141592923,
      "grad_norm": 1.2117606473793785e-08,
      "learning_rate": 0.15907757975739548,
      "loss": 0.0,
      "num_input_tokens_seen": 10922128,
      "step": 19230
    },
    {
      "epoch": 337.46017699115043,
      "grad_norm": 1.880126099251811e-08,
      "learning_rate": 0.159018782160214,
      "loss": 0.0,
      "num_input_tokens_seen": 10924864,
      "step": 19235
    },
    {
      "epoch": 337.5486725663717,
      "grad_norm": 2.8546976338361674e-08,
      "learning_rate": 0.158959983172223,
      "loss": 0.0,
      "num_input_tokens_seen": 10927936,
      "step": 19240
    },
    {
      "epoch": 337.6371681415929,
      "grad_norm": 1.688643713748661e-08,
      "learning_rate": 0.15890118280249,
      "loss": 0.0,
      "num_input_tokens_seen": 10930928,
      "step": 19245
    },
    {
      "epoch": 337.72566371681415,
      "grad_norm": 8.83623840763903e-09,
      "learning_rate": 0.15884238106008275,
      "loss": 0.0,
      "num_input_tokens_seen": 10933792,
      "step": 19250
    },
    {
      "epoch": 337.8141592920354,
      "grad_norm": 2.998285353328356e-08,
      "learning_rate": 0.15878357795406922,
      "loss": 0.0,
      "num_input_tokens_seen": 10936432,
      "step": 19255
    },
    {
      "epoch": 337.9026548672566,
      "grad_norm": 1.776307989587167e-08,
      "learning_rate": 0.15872477349351757,
      "loss": 0.0,
      "num_input_tokens_seen": 10939152,
      "step": 19260
    },
    {
      "epoch": 337.9911504424779,
      "grad_norm": 1.2952593664294909e-08,
      "learning_rate": 0.15866596768749622,
      "loss": 0.0,
      "num_input_tokens_seen": 10941792,
      "step": 19265
    },
    {
      "epoch": 338.070796460177,
      "grad_norm": 2.2288478618293084e-08,
      "learning_rate": 0.15860716054507373,
      "loss": 0.0,
      "num_input_tokens_seen": 10944112,
      "step": 19270
    },
    {
      "epoch": 338.1592920353982,
      "grad_norm": 3.8557551818030333e-08,
      "learning_rate": 0.1585483520753189,
      "loss": 0.0,
      "num_input_tokens_seen": 10947408,
      "step": 19275
    },
    {
      "epoch": 338.24778761061947,
      "grad_norm": 4.5522348557369696e-08,
      "learning_rate": 0.1584895422873008,
      "loss": 0.0,
      "num_input_tokens_seen": 10950048,
      "step": 19280
    },
    {
      "epoch": 338.3362831858407,
      "grad_norm": 1.1729605731147785e-08,
      "learning_rate": 0.1584307311900886,
      "loss": 0.0,
      "num_input_tokens_seen": 10953424,
      "step": 19285
    },
    {
      "epoch": 338.42477876106193,
      "grad_norm": 4.8105718519764196e-08,
      "learning_rate": 0.1583719187927517,
      "loss": 0.0,
      "num_input_tokens_seen": 10956016,
      "step": 19290
    },
    {
      "epoch": 338.5132743362832,
      "grad_norm": 1.662078652486798e-08,
      "learning_rate": 0.15831310510435967,
      "loss": 0.0,
      "num_input_tokens_seen": 10958960,
      "step": 19295
    },
    {
      "epoch": 338.60176991150445,
      "grad_norm": 2.2359538220939612e-08,
      "learning_rate": 0.15825429013398243,
      "loss": 0.0,
      "num_input_tokens_seen": 10961824,
      "step": 19300
    },
    {
      "epoch": 338.69026548672565,
      "grad_norm": 1.1256501508682959e-08,
      "learning_rate": 0.15819547389068986,
      "loss": 0.0,
      "num_input_tokens_seen": 10964368,
      "step": 19305
    },
    {
      "epoch": 338.7787610619469,
      "grad_norm": 2.226046902364942e-08,
      "learning_rate": 0.1581366563835522,
      "loss": 0.0,
      "num_input_tokens_seen": 10967520,
      "step": 19310
    },
    {
      "epoch": 338.86725663716817,
      "grad_norm": 3.578966811801365e-08,
      "learning_rate": 0.15807783762163993,
      "loss": 0.0,
      "num_input_tokens_seen": 10970288,
      "step": 19315
    },
    {
      "epoch": 338.95575221238937,
      "grad_norm": 1.7430570764531694e-08,
      "learning_rate": 0.15801901761402365,
      "loss": 0.0,
      "num_input_tokens_seen": 10973344,
      "step": 19320
    },
    {
      "epoch": 339.0353982300885,
      "grad_norm": 2.0082385532305125e-08,
      "learning_rate": 0.157960196369774,
      "loss": 0.0,
      "num_input_tokens_seen": 10975544,
      "step": 19325
    },
    {
      "epoch": 339.12389380530976,
      "grad_norm": 1.963833362594869e-08,
      "learning_rate": 0.157901373897962,
      "loss": 0.0,
      "num_input_tokens_seen": 10978328,
      "step": 19330
    },
    {
      "epoch": 339.21238938053096,
      "grad_norm": 2.0540603884455777e-08,
      "learning_rate": 0.15784255020765892,
      "loss": 0.0,
      "num_input_tokens_seen": 10980952,
      "step": 19335
    },
    {
      "epoch": 339.3008849557522,
      "grad_norm": 1.5691943744400305e-08,
      "learning_rate": 0.157783725307936,
      "loss": 0.0,
      "num_input_tokens_seen": 10983928,
      "step": 19340
    },
    {
      "epoch": 339.3893805309734,
      "grad_norm": 2.02158041417988e-08,
      "learning_rate": 0.15772489920786484,
      "loss": 0.0,
      "num_input_tokens_seen": 10987000,
      "step": 19345
    },
    {
      "epoch": 339.4778761061947,
      "grad_norm": 1.8629103593070795e-08,
      "learning_rate": 0.15766607191651713,
      "loss": 0.0,
      "num_input_tokens_seen": 10990520,
      "step": 19350
    },
    {
      "epoch": 339.56637168141594,
      "grad_norm": 2.3975617935434457e-08,
      "learning_rate": 0.1576072434429648,
      "loss": 0.0,
      "num_input_tokens_seen": 10993288,
      "step": 19355
    },
    {
      "epoch": 339.65486725663715,
      "grad_norm": 1.4793481994956892e-08,
      "learning_rate": 0.15754841379627998,
      "loss": 0.0,
      "num_input_tokens_seen": 10996008,
      "step": 19360
    },
    {
      "epoch": 339.7433628318584,
      "grad_norm": 9.334805817218239e-09,
      "learning_rate": 0.15748958298553484,
      "loss": 0.0,
      "num_input_tokens_seen": 10999176,
      "step": 19365
    },
    {
      "epoch": 339.83185840707966,
      "grad_norm": 7.770646348603805e-09,
      "learning_rate": 0.1574307510198019,
      "loss": 0.0,
      "num_input_tokens_seen": 11001864,
      "step": 19370
    },
    {
      "epoch": 339.92035398230087,
      "grad_norm": 2.655172082199897e-08,
      "learning_rate": 0.15737191790815375,
      "loss": 0.0,
      "num_input_tokens_seen": 11004408,
      "step": 19375
    },
    {
      "epoch": 340.0,
      "grad_norm": 2.5776570211633043e-08,
      "learning_rate": 0.15731308365966323,
      "loss": 0.0,
      "num_input_tokens_seen": 11006656,
      "step": 19380
    },
    {
      "epoch": 340.08849557522126,
      "grad_norm": 2.184219916046004e-08,
      "learning_rate": 0.15725424828340331,
      "loss": 0.0,
      "num_input_tokens_seen": 11009664,
      "step": 19385
    },
    {
      "epoch": 340.17699115044246,
      "grad_norm": 1.2979357144615733e-08,
      "learning_rate": 0.15719541178844715,
      "loss": 0.0,
      "num_input_tokens_seen": 11012592,
      "step": 19390
    },
    {
      "epoch": 340.2654867256637,
      "grad_norm": 2.179431035642665e-08,
      "learning_rate": 0.15713657418386806,
      "loss": 0.0,
      "num_input_tokens_seen": 11015328,
      "step": 19395
    },
    {
      "epoch": 340.353982300885,
      "grad_norm": 2.4013679933432286e-08,
      "learning_rate": 0.15707773547873957,
      "loss": 0.0,
      "num_input_tokens_seen": 11018112,
      "step": 19400
    },
    {
      "epoch": 340.353982300885,
      "eval_loss": 0.6213271617889404,
      "eval_runtime": 0.9383,
      "eval_samples_per_second": 26.644,
      "eval_steps_per_second": 13.855,
      "num_input_tokens_seen": 11018112,
      "step": 19400
    },
    {
      "epoch": 340.4424778761062,
      "grad_norm": 5.659961743731401e-08,
      "learning_rate": 0.1570188956821353,
      "loss": 0.0,
      "num_input_tokens_seen": 11021200,
      "step": 19405
    },
    {
      "epoch": 340.53097345132744,
      "grad_norm": 2.9240696974852654e-08,
      "learning_rate": 0.1569600548031291,
      "loss": 0.0,
      "num_input_tokens_seen": 11024320,
      "step": 19410
    },
    {
      "epoch": 340.6194690265487,
      "grad_norm": 2.9712312610286062e-08,
      "learning_rate": 0.156901212850795,
      "loss": 0.0,
      "num_input_tokens_seen": 11027104,
      "step": 19415
    },
    {
      "epoch": 340.7079646017699,
      "grad_norm": 1.9466442680027285e-08,
      "learning_rate": 0.15684236983420716,
      "loss": 0.0,
      "num_input_tokens_seen": 11029856,
      "step": 19420
    },
    {
      "epoch": 340.79646017699116,
      "grad_norm": 3.4599061393691954e-08,
      "learning_rate": 0.1567835257624399,
      "loss": 0.0,
      "num_input_tokens_seen": 11032640,
      "step": 19425
    },
    {
      "epoch": 340.88495575221236,
      "grad_norm": 2.9608376195255914e-08,
      "learning_rate": 0.1567246806445677,
      "loss": 0.0,
      "num_input_tokens_seen": 11035648,
      "step": 19430
    },
    {
      "epoch": 340.9734513274336,
      "grad_norm": 1.4636702516668265e-08,
      "learning_rate": 0.15666583448966526,
      "loss": 0.0,
      "num_input_tokens_seen": 11038416,
      "step": 19435
    },
    {
      "epoch": 341.05309734513276,
      "grad_norm": 2.1981334086262905e-08,
      "learning_rate": 0.1566069873068074,
      "loss": 0.0,
      "num_input_tokens_seen": 11040640,
      "step": 19440
    },
    {
      "epoch": 341.14159292035396,
      "grad_norm": 2.9149621383339763e-08,
      "learning_rate": 0.156548139105069,
      "loss": 0.0,
      "num_input_tokens_seen": 11043136,
      "step": 19445
    },
    {
      "epoch": 341.2300884955752,
      "grad_norm": 2.1244167314193874e-08,
      "learning_rate": 0.15648928989352529,
      "loss": 0.0,
      "num_input_tokens_seen": 11046432,
      "step": 19450
    },
    {
      "epoch": 341.3185840707965,
      "grad_norm": 1.7466796009557584e-08,
      "learning_rate": 0.15643043968125156,
      "loss": 0.0,
      "num_input_tokens_seen": 11049520,
      "step": 19455
    },
    {
      "epoch": 341.4070796460177,
      "grad_norm": 1.759700474224246e-08,
      "learning_rate": 0.15637158847732316,
      "loss": 0.0,
      "num_input_tokens_seen": 11052496,
      "step": 19460
    },
    {
      "epoch": 341.49557522123894,
      "grad_norm": 2.002167853731862e-08,
      "learning_rate": 0.15631273629081582,
      "loss": 0.0,
      "num_input_tokens_seen": 11055472,
      "step": 19465
    },
    {
      "epoch": 341.5840707964602,
      "grad_norm": 1.6389485324452835e-08,
      "learning_rate": 0.15625388313080518,
      "loss": 0.0,
      "num_input_tokens_seen": 11058000,
      "step": 19470
    },
    {
      "epoch": 341.6725663716814,
      "grad_norm": 1.341398636611757e-08,
      "learning_rate": 0.15619502900636714,
      "loss": 0.0,
      "num_input_tokens_seen": 11060640,
      "step": 19475
    },
    {
      "epoch": 341.76106194690266,
      "grad_norm": 1.9540433271458824e-08,
      "learning_rate": 0.15613617392657783,
      "loss": 0.0,
      "num_input_tokens_seen": 11064000,
      "step": 19480
    },
    {
      "epoch": 341.8495575221239,
      "grad_norm": 2.485287531328595e-08,
      "learning_rate": 0.15607731790051335,
      "loss": 0.0,
      "num_input_tokens_seen": 11067120,
      "step": 19485
    },
    {
      "epoch": 341.9380530973451,
      "grad_norm": 1.7041770661307964e-08,
      "learning_rate": 0.15601846093725008,
      "loss": 0.0,
      "num_input_tokens_seen": 11069984,
      "step": 19490
    },
    {
      "epoch": 342.01769911504425,
      "grad_norm": 2.1504101610503312e-08,
      "learning_rate": 0.1559596030458645,
      "loss": 0.0,
      "num_input_tokens_seen": 11072296,
      "step": 19495
    },
    {
      "epoch": 342.1061946902655,
      "grad_norm": 1.1361498408746229e-08,
      "learning_rate": 0.1559007442354333,
      "loss": 0.0,
      "num_input_tokens_seen": 11075384,
      "step": 19500
    },
    {
      "epoch": 342.1946902654867,
      "grad_norm": 2.0563241776017094e-08,
      "learning_rate": 0.15584188451503314,
      "loss": 0.0,
      "num_input_tokens_seen": 11078440,
      "step": 19505
    },
    {
      "epoch": 342.283185840708,
      "grad_norm": 2.5246661650157876e-08,
      "learning_rate": 0.15578302389374094,
      "loss": 0.0,
      "num_input_tokens_seen": 11081256,
      "step": 19510
    },
    {
      "epoch": 342.37168141592923,
      "grad_norm": 2.2373413344212167e-08,
      "learning_rate": 0.1557241623806338,
      "loss": 0.0,
      "num_input_tokens_seen": 11084200,
      "step": 19515
    },
    {
      "epoch": 342.46017699115043,
      "grad_norm": 1.2857360509599403e-08,
      "learning_rate": 0.15566529998478887,
      "loss": 0.0,
      "num_input_tokens_seen": 11087176,
      "step": 19520
    },
    {
      "epoch": 342.5486725663717,
      "grad_norm": 2.5074287535176154e-08,
      "learning_rate": 0.15560643671528354,
      "loss": 0.0,
      "num_input_tokens_seen": 11089848,
      "step": 19525
    },
    {
      "epoch": 342.6371681415929,
      "grad_norm": 1.3175473156934459e-08,
      "learning_rate": 0.15554757258119514,
      "loss": 0.0,
      "num_input_tokens_seen": 11092376,
      "step": 19530
    },
    {
      "epoch": 342.72566371681415,
      "grad_norm": 2.702178747426842e-08,
      "learning_rate": 0.1554887075916014,
      "loss": 0.0,
      "num_input_tokens_seen": 11095128,
      "step": 19535
    },
    {
      "epoch": 342.8141592920354,
      "grad_norm": 1.5285881005411284e-08,
      "learning_rate": 0.15542984175558,
      "loss": 0.0,
      "num_input_tokens_seen": 11098024,
      "step": 19540
    },
    {
      "epoch": 342.9026548672566,
      "grad_norm": 8.54390425075735e-09,
      "learning_rate": 0.1553709750822087,
      "loss": 0.0,
      "num_input_tokens_seen": 11100968,
      "step": 19545
    },
    {
      "epoch": 342.9911504424779,
      "grad_norm": 1.1288265433506695e-08,
      "learning_rate": 0.15531210758056554,
      "loss": 0.0,
      "num_input_tokens_seen": 11103816,
      "step": 19550
    },
    {
      "epoch": 343.070796460177,
      "grad_norm": 2.508608787366029e-08,
      "learning_rate": 0.15525323925972867,
      "loss": 0.0,
      "num_input_tokens_seen": 11106368,
      "step": 19555
    },
    {
      "epoch": 343.1592920353982,
      "grad_norm": 1.7419495179638034e-08,
      "learning_rate": 0.15519437012877627,
      "loss": 0.0,
      "num_input_tokens_seen": 11109088,
      "step": 19560
    },
    {
      "epoch": 343.24778761061947,
      "grad_norm": 3.9332448409368226e-08,
      "learning_rate": 0.15513550019678676,
      "loss": 0.0,
      "num_input_tokens_seen": 11111712,
      "step": 19565
    },
    {
      "epoch": 343.3362831858407,
      "grad_norm": 2.485559846832075e-08,
      "learning_rate": 0.15507662947283854,
      "loss": 0.0,
      "num_input_tokens_seen": 11114560,
      "step": 19570
    },
    {
      "epoch": 343.42477876106193,
      "grad_norm": 5.5668074594450445e-08,
      "learning_rate": 0.15501775796601028,
      "loss": 0.0,
      "num_input_tokens_seen": 11117568,
      "step": 19575
    },
    {
      "epoch": 343.5132743362832,
      "grad_norm": 1.198928689660761e-08,
      "learning_rate": 0.15495888568538066,
      "loss": 0.0,
      "num_input_tokens_seen": 11120320,
      "step": 19580
    },
    {
      "epoch": 343.60176991150445,
      "grad_norm": 2.1079257450651312e-08,
      "learning_rate": 0.1549000126400286,
      "loss": 0.0,
      "num_input_tokens_seen": 11123248,
      "step": 19585
    },
    {
      "epoch": 343.69026548672565,
      "grad_norm": 1.4345579835151057e-08,
      "learning_rate": 0.15484113883903294,
      "loss": 0.0,
      "num_input_tokens_seen": 11125840,
      "step": 19590
    },
    {
      "epoch": 343.7787610619469,
      "grad_norm": 1.3653644437283674e-08,
      "learning_rate": 0.15478226429147288,
      "loss": 0.0,
      "num_input_tokens_seen": 11128672,
      "step": 19595
    },
    {
      "epoch": 343.86725663716817,
      "grad_norm": 1.8581886251922697e-08,
      "learning_rate": 0.15472338900642757,
      "loss": 0.0,
      "num_input_tokens_seen": 11131712,
      "step": 19600
    },
    {
      "epoch": 343.86725663716817,
      "eval_loss": 0.6173127889633179,
      "eval_runtime": 0.9391,
      "eval_samples_per_second": 26.62,
      "eval_steps_per_second": 13.843,
      "num_input_tokens_seen": 11131712,
      "step": 19600
    },
    {
      "epoch": 343.95575221238937,
      "grad_norm": 1.731320686815252e-08,
      "learning_rate": 0.15466451299297632,
      "loss": 0.0,
      "num_input_tokens_seen": 11134928,
      "step": 19605
    },
    {
      "epoch": 344.0353982300885,
      "grad_norm": 1.7944948638159985e-08,
      "learning_rate": 0.15460563626019852,
      "loss": 0.0,
      "num_input_tokens_seen": 11137648,
      "step": 19610
    },
    {
      "epoch": 344.12389380530976,
      "grad_norm": 2.3812226856989582e-08,
      "learning_rate": 0.15454675881717375,
      "loss": 0.0,
      "num_input_tokens_seen": 11140272,
      "step": 19615
    },
    {
      "epoch": 344.21238938053096,
      "grad_norm": 3.3350289641020936e-08,
      "learning_rate": 0.1544878806729816,
      "loss": 0.0,
      "num_input_tokens_seen": 11142896,
      "step": 19620
    },
    {
      "epoch": 344.3008849557522,
      "grad_norm": 2.3984101815699432e-08,
      "learning_rate": 0.1544290018367019,
      "loss": 0.0,
      "num_input_tokens_seen": 11146208,
      "step": 19625
    },
    {
      "epoch": 344.3893805309734,
      "grad_norm": 1.4134702297496915e-08,
      "learning_rate": 0.15437012231741445,
      "loss": 0.0,
      "num_input_tokens_seen": 11149584,
      "step": 19630
    },
    {
      "epoch": 344.4778761061947,
      "grad_norm": 1.702203178410855e-08,
      "learning_rate": 0.1543112421241992,
      "loss": 0.0,
      "num_input_tokens_seen": 11152160,
      "step": 19635
    },
    {
      "epoch": 344.56637168141594,
      "grad_norm": 1.7721630385381104e-08,
      "learning_rate": 0.15425236126613626,
      "loss": 0.0,
      "num_input_tokens_seen": 11154768,
      "step": 19640
    },
    {
      "epoch": 344.65486725663715,
      "grad_norm": 1.619093481508571e-08,
      "learning_rate": 0.15419347975230577,
      "loss": 0.0,
      "num_input_tokens_seen": 11157456,
      "step": 19645
    },
    {
      "epoch": 344.7433628318584,
      "grad_norm": 9.433186676233163e-09,
      "learning_rate": 0.154134597591788,
      "loss": 0.0,
      "num_input_tokens_seen": 11160576,
      "step": 19650
    },
    {
      "epoch": 344.83185840707966,
      "grad_norm": 2.232643403488055e-08,
      "learning_rate": 0.1540757147936633,
      "loss": 0.0,
      "num_input_tokens_seen": 11163520,
      "step": 19655
    },
    {
      "epoch": 344.92035398230087,
      "grad_norm": 1.3529916742527348e-08,
      "learning_rate": 0.1540168313670122,
      "loss": 0.0,
      "num_input_tokens_seen": 11166160,
      "step": 19660
    },
    {
      "epoch": 345.0,
      "grad_norm": 9.568119452296742e-08,
      "learning_rate": 0.1539579473209152,
      "loss": 0.0,
      "num_input_tokens_seen": 11168448,
      "step": 19665
    },
    {
      "epoch": 345.08849557522126,
      "grad_norm": 1.3423012035218562e-08,
      "learning_rate": 0.15389906266445294,
      "loss": 0.0,
      "num_input_tokens_seen": 11170944,
      "step": 19670
    },
    {
      "epoch": 345.17699115044246,
      "grad_norm": 1.3072654958534713e-08,
      "learning_rate": 0.15384017740670627,
      "loss": 0.0,
      "num_input_tokens_seen": 11173616,
      "step": 19675
    },
    {
      "epoch": 345.2654867256637,
      "grad_norm": 2.2589210502133028e-08,
      "learning_rate": 0.15378129155675602,
      "loss": 0.0,
      "num_input_tokens_seen": 11176368,
      "step": 19680
    },
    {
      "epoch": 345.353982300885,
      "grad_norm": 2.5367063116732425e-08,
      "learning_rate": 0.15372240512368307,
      "loss": 0.0,
      "num_input_tokens_seen": 11179408,
      "step": 19685
    },
    {
      "epoch": 345.4424778761062,
      "grad_norm": 1.9143579166325253e-08,
      "learning_rate": 0.1536635181165684,
      "loss": 0.0,
      "num_input_tokens_seen": 11182304,
      "step": 19690
    },
    {
      "epoch": 345.53097345132744,
      "grad_norm": 2.174751401184949e-08,
      "learning_rate": 0.15360463054449328,
      "loss": 0.0,
      "num_input_tokens_seen": 11185616,
      "step": 19695
    },
    {
      "epoch": 345.6194690265487,
      "grad_norm": 2.037923430009414e-08,
      "learning_rate": 0.1535457424165388,
      "loss": 0.0,
      "num_input_tokens_seen": 11188656,
      "step": 19700
    },
    {
      "epoch": 345.7079646017699,
      "grad_norm": 1.679571504098476e-08,
      "learning_rate": 0.15348685374178628,
      "loss": 0.0,
      "num_input_tokens_seen": 11191712,
      "step": 19705
    },
    {
      "epoch": 345.79646017699116,
      "grad_norm": 8.979041510315255e-09,
      "learning_rate": 0.1534279645293171,
      "loss": 0.0,
      "num_input_tokens_seen": 11194480,
      "step": 19710
    },
    {
      "epoch": 345.88495575221236,
      "grad_norm": 1.1783996889391801e-08,
      "learning_rate": 0.1533690747882127,
      "loss": 0.0,
      "num_input_tokens_seen": 11197456,
      "step": 19715
    },
    {
      "epoch": 345.9734513274336,
      "grad_norm": 3.792064262597705e-08,
      "learning_rate": 0.15331018452755465,
      "loss": 0.0,
      "num_input_tokens_seen": 11200352,
      "step": 19720
    },
    {
      "epoch": 346.05309734513276,
      "grad_norm": 2.3152407990778556e-08,
      "learning_rate": 0.15325129375642457,
      "loss": 0.0,
      "num_input_tokens_seen": 11202480,
      "step": 19725
    },
    {
      "epoch": 346.14159292035396,
      "grad_norm": 2.0203787087780256e-08,
      "learning_rate": 0.15319240248390406,
      "loss": 0.0,
      "num_input_tokens_seen": 11204992,
      "step": 19730
    },
    {
      "epoch": 346.2300884955752,
      "grad_norm": 2.3706277829660394e-08,
      "learning_rate": 0.153133510719075,
      "loss": 0.0,
      "num_input_tokens_seen": 11208208,
      "step": 19735
    },
    {
      "epoch": 346.3185840707965,
      "grad_norm": 2.5576639473001705e-08,
      "learning_rate": 0.15307461847101922,
      "loss": 0.0,
      "num_input_tokens_seen": 11211472,
      "step": 19740
    },
    {
      "epoch": 346.4070796460177,
      "grad_norm": 2.164278534166897e-08,
      "learning_rate": 0.15301572574881864,
      "loss": 0.0,
      "num_input_tokens_seen": 11214624,
      "step": 19745
    },
    {
      "epoch": 346.49557522123894,
      "grad_norm": 1.7428677168140894e-08,
      "learning_rate": 0.15295683256155523,
      "loss": 0.0,
      "num_input_tokens_seen": 11217072,
      "step": 19750
    },
    {
      "epoch": 346.5840707964602,
      "grad_norm": 1.8294086245873586e-08,
      "learning_rate": 0.15289793891831113,
      "loss": 0.0,
      "num_input_tokens_seen": 11220288,
      "step": 19755
    },
    {
      "epoch": 346.6725663716814,
      "grad_norm": 1.531275195532089e-08,
      "learning_rate": 0.15283904482816837,
      "loss": 0.0,
      "num_input_tokens_seen": 11222928,
      "step": 19760
    },
    {
      "epoch": 346.76106194690266,
      "grad_norm": 3.167113504787267e-08,
      "learning_rate": 0.15278015030020928,
      "loss": 0.0,
      "num_input_tokens_seen": 11225856,
      "step": 19765
    },
    {
      "epoch": 346.8495575221239,
      "grad_norm": 1.8796445289126495e-08,
      "learning_rate": 0.152721255343516,
      "loss": 0.0,
      "num_input_tokens_seen": 11228656,
      "step": 19770
    },
    {
      "epoch": 346.9380530973451,
      "grad_norm": 3.922051572402552e-09,
      "learning_rate": 0.15266235996717098,
      "loss": 0.0,
      "num_input_tokens_seen": 11231744,
      "step": 19775
    },
    {
      "epoch": 347.01769911504425,
      "grad_norm": 1.8081482977549967e-08,
      "learning_rate": 0.15260346418025664,
      "loss": 0.0,
      "num_input_tokens_seen": 11233888,
      "step": 19780
    },
    {
      "epoch": 347.1061946902655,
      "grad_norm": 1.9048654209541382e-08,
      "learning_rate": 0.15254456799185537,
      "loss": 0.0,
      "num_input_tokens_seen": 11236544,
      "step": 19785
    },
    {
      "epoch": 347.1946902654867,
      "grad_norm": 2.109490537804959e-08,
      "learning_rate": 0.15248567141104974,
      "loss": 0.0,
      "num_input_tokens_seen": 11239392,
      "step": 19790
    },
    {
      "epoch": 347.283185840708,
      "grad_norm": 1.963394424819853e-08,
      "learning_rate": 0.15242677444692232,
      "loss": 0.0,
      "num_input_tokens_seen": 11242960,
      "step": 19795
    },
    {
      "epoch": 347.37168141592923,
      "grad_norm": 2.583110614295947e-08,
      "learning_rate": 0.15236787710855584,
      "loss": 0.0,
      "num_input_tokens_seen": 11245728,
      "step": 19800
    },
    {
      "epoch": 347.37168141592923,
      "eval_loss": 0.6093894243240356,
      "eval_runtime": 0.9347,
      "eval_samples_per_second": 26.748,
      "eval_steps_per_second": 13.909,
      "num_input_tokens_seen": 11245728,
      "step": 19800
    },
    {
      "epoch": 347.46017699115043,
      "grad_norm": 3.744959542473225e-08,
      "learning_rate": 0.1523089794050329,
      "loss": 0.0,
      "num_input_tokens_seen": 11248576,
      "step": 19805
    },
    {
      "epoch": 347.5486725663717,
      "grad_norm": 9.21286513744235e-09,
      "learning_rate": 0.15225008134543633,
      "loss": 0.0,
      "num_input_tokens_seen": 11251264,
      "step": 19810
    },
    {
      "epoch": 347.6371681415929,
      "grad_norm": 1.8884090735582504e-08,
      "learning_rate": 0.15219118293884895,
      "loss": 0.0,
      "num_input_tokens_seen": 11254272,
      "step": 19815
    },
    {
      "epoch": 347.72566371681415,
      "grad_norm": 4.1639449221975156e-08,
      "learning_rate": 0.15213228419435362,
      "loss": 0.0,
      "num_input_tokens_seen": 11257008,
      "step": 19820
    },
    {
      "epoch": 347.8141592920354,
      "grad_norm": 2.6339314729284524e-08,
      "learning_rate": 0.15207338512103327,
      "loss": 0.0,
      "num_input_tokens_seen": 11259760,
      "step": 19825
    },
    {
      "epoch": 347.9026548672566,
      "grad_norm": 2.030880175141192e-08,
      "learning_rate": 0.1520144857279709,
      "loss": 0.0,
      "num_input_tokens_seen": 11262688,
      "step": 19830
    },
    {
      "epoch": 347.9911504424779,
      "grad_norm": 2.2875488170370772e-08,
      "learning_rate": 0.1519555860242495,
      "loss": 0.0,
      "num_input_tokens_seen": 11265312,
      "step": 19835
    },
    {
      "epoch": 348.070796460177,
      "grad_norm": 3.7658036688981156e-08,
      "learning_rate": 0.15189668601895218,
      "loss": 0.0,
      "num_input_tokens_seen": 11267912,
      "step": 19840
    },
    {
      "epoch": 348.1592920353982,
      "grad_norm": 1.1165218083419859e-08,
      "learning_rate": 0.151837785721162,
      "loss": 0.0,
      "num_input_tokens_seen": 11270536,
      "step": 19845
    },
    {
      "epoch": 348.24778761061947,
      "grad_norm": 1.9594860845018047e-08,
      "learning_rate": 0.15177888513996218,
      "loss": 0.0,
      "num_input_tokens_seen": 11273512,
      "step": 19850
    },
    {
      "epoch": 348.3362831858407,
      "grad_norm": 2.6501448147087103e-08,
      "learning_rate": 0.15171998428443592,
      "loss": 0.0,
      "num_input_tokens_seen": 11276392,
      "step": 19855
    },
    {
      "epoch": 348.42477876106193,
      "grad_norm": 2.46508431445136e-08,
      "learning_rate": 0.1516610831636665,
      "loss": 0.0,
      "num_input_tokens_seen": 11279480,
      "step": 19860
    },
    {
      "epoch": 348.5132743362832,
      "grad_norm": 1.497066648425971e-08,
      "learning_rate": 0.15160218178673715,
      "loss": 0.0,
      "num_input_tokens_seen": 11282088,
      "step": 19865
    },
    {
      "epoch": 348.60176991150445,
      "grad_norm": 2.2987851622247035e-08,
      "learning_rate": 0.15154328016273122,
      "loss": 0.0,
      "num_input_tokens_seen": 11285400,
      "step": 19870
    },
    {
      "epoch": 348.69026548672565,
      "grad_norm": 1.2522995973540674e-08,
      "learning_rate": 0.1514843783007321,
      "loss": 0.0,
      "num_input_tokens_seen": 11288392,
      "step": 19875
    },
    {
      "epoch": 348.7787610619469,
      "grad_norm": 2.1935191441002644e-08,
      "learning_rate": 0.15142547620982322,
      "loss": 0.0,
      "num_input_tokens_seen": 11291192,
      "step": 19880
    },
    {
      "epoch": 348.86725663716817,
      "grad_norm": 1.0540124328883849e-08,
      "learning_rate": 0.15136657389908797,
      "loss": 0.0,
      "num_input_tokens_seen": 11293768,
      "step": 19885
    },
    {
      "epoch": 348.95575221238937,
      "grad_norm": 1.754111700336125e-08,
      "learning_rate": 0.15130767137760986,
      "loss": 0.0,
      "num_input_tokens_seen": 11296952,
      "step": 19890
    },
    {
      "epoch": 349.0353982300885,
      "grad_norm": 1.5903479422263445e-08,
      "learning_rate": 0.15124876865447243,
      "loss": 0.0,
      "num_input_tokens_seen": 11299280,
      "step": 19895
    },
    {
      "epoch": 349.12389380530976,
      "grad_norm": 2.12195612192545e-08,
      "learning_rate": 0.15118986573875912,
      "loss": 0.0,
      "num_input_tokens_seen": 11302400,
      "step": 19900
    },
    {
      "epoch": 349.21238938053096,
      "grad_norm": 1.8483973462934955e-08,
      "learning_rate": 0.15113096263955358,
      "loss": 0.0,
      "num_input_tokens_seen": 11305104,
      "step": 19905
    },
    {
      "epoch": 349.3008849557522,
      "grad_norm": 3.3185944658953304e-08,
      "learning_rate": 0.1510720593659394,
      "loss": 0.0,
      "num_input_tokens_seen": 11308352,
      "step": 19910
    },
    {
      "epoch": 349.3893805309734,
      "grad_norm": 2.1092908752962103e-08,
      "learning_rate": 0.15101315592700015,
      "loss": 0.0,
      "num_input_tokens_seen": 11311328,
      "step": 19915
    },
    {
      "epoch": 349.4778761061947,
      "grad_norm": 1.3774053897464e-08,
      "learning_rate": 0.15095425233181956,
      "loss": 0.0,
      "num_input_tokens_seen": 11313552,
      "step": 19920
    },
    {
      "epoch": 349.56637168141594,
      "grad_norm": 1.319958009560196e-08,
      "learning_rate": 0.15089534858948128,
      "loss": 0.0,
      "num_input_tokens_seen": 11316496,
      "step": 19925
    },
    {
      "epoch": 349.65486725663715,
      "grad_norm": 1.371769364766351e-08,
      "learning_rate": 0.15083644470906898,
      "loss": 0.0,
      "num_input_tokens_seen": 11318912,
      "step": 19930
    },
    {
      "epoch": 349.7433628318584,
      "grad_norm": 1.1879137673531659e-08,
      "learning_rate": 0.1507775406996664,
      "loss": 0.0,
      "num_input_tokens_seen": 11322416,
      "step": 19935
    },
    {
      "epoch": 349.83185840707966,
      "grad_norm": 7.435073001715864e-09,
      "learning_rate": 0.15071863657035725,
      "loss": 0.0,
      "num_input_tokens_seen": 11325248,
      "step": 19940
    },
    {
      "epoch": 349.92035398230087,
      "grad_norm": 1.2842073182639524e-08,
      "learning_rate": 0.15065973233022534,
      "loss": 0.0,
      "num_input_tokens_seen": 11327872,
      "step": 19945
    },
    {
      "epoch": 350.0,
      "grad_norm": 6.898752236139671e-09,
      "learning_rate": 0.15060082798835442,
      "loss": 0.0,
      "num_input_tokens_seen": 11330576,
      "step": 19950
    },
    {
      "epoch": 350.08849557522126,
      "grad_norm": 2.050330571989889e-08,
      "learning_rate": 0.15054192355382823,
      "loss": 0.0,
      "num_input_tokens_seen": 11333232,
      "step": 19955
    },
    {
      "epoch": 350.17699115044246,
      "grad_norm": 1.883869948926531e-08,
      "learning_rate": 0.15048301903573066,
      "loss": 0.0,
      "num_input_tokens_seen": 11335952,
      "step": 19960
    },
    {
      "epoch": 350.2654867256637,
      "grad_norm": 2.6548921283620075e-08,
      "learning_rate": 0.15042411444314546,
      "loss": 0.0,
      "num_input_tokens_seen": 11339136,
      "step": 19965
    },
    {
      "epoch": 350.353982300885,
      "grad_norm": 2.8345725766598662e-08,
      "learning_rate": 0.1503652097851565,
      "loss": 0.0,
      "num_input_tokens_seen": 11341648,
      "step": 19970
    },
    {
      "epoch": 350.4424778761062,
      "grad_norm": 1.2640960278531566e-08,
      "learning_rate": 0.15030630507084758,
      "loss": 0.0,
      "num_input_tokens_seen": 11344272,
      "step": 19975
    },
    {
      "epoch": 350.53097345132744,
      "grad_norm": 1.07394830806129e-08,
      "learning_rate": 0.1502474003093026,
      "loss": 0.0,
      "num_input_tokens_seen": 11346896,
      "step": 19980
    },
    {
      "epoch": 350.6194690265487,
      "grad_norm": 8.41992253697299e-09,
      "learning_rate": 0.15018849550960536,
      "loss": 0.0,
      "num_input_tokens_seen": 11349904,
      "step": 19985
    },
    {
      "epoch": 350.7079646017699,
      "grad_norm": 2.4330423897822584e-08,
      "learning_rate": 0.15012959068083975,
      "loss": 0.0,
      "num_input_tokens_seen": 11352720,
      "step": 19990
    },
    {
      "epoch": 350.79646017699116,
      "grad_norm": 1.13100506737851e-08,
      "learning_rate": 0.1500706858320896,
      "loss": 0.0,
      "num_input_tokens_seen": 11356320,
      "step": 19995
    },
    {
      "epoch": 350.88495575221236,
      "grad_norm": 1.6672924374461218e-08,
      "learning_rate": 0.15001178097243886,
      "loss": 0.0,
      "num_input_tokens_seen": 11358800,
      "step": 20000
    },
    {
      "epoch": 350.88495575221236,
      "eval_loss": 0.6228640675544739,
      "eval_runtime": 0.9433,
      "eval_samples_per_second": 26.501,
      "eval_steps_per_second": 13.781,
      "num_input_tokens_seen": 11358800,
      "step": 20000
    },
    {
      "epoch": 350.9734513274336,
      "grad_norm": 1.2977445784656538e-08,
      "learning_rate": 0.1499528761109713,
      "loss": 0.0,
      "num_input_tokens_seen": 11362048,
      "step": 20005
    },
    {
      "epoch": 351.05309734513276,
      "grad_norm": 2.023635126136014e-08,
      "learning_rate": 0.14989397125677087,
      "loss": 0.0,
      "num_input_tokens_seen": 11364464,
      "step": 20010
    },
    {
      "epoch": 351.14159292035396,
      "grad_norm": 5.89170490172819e-09,
      "learning_rate": 0.14983506641892141,
      "loss": 0.0,
      "num_input_tokens_seen": 11367328,
      "step": 20015
    },
    {
      "epoch": 351.2300884955752,
      "grad_norm": 3.4013510230579413e-08,
      "learning_rate": 0.14977616160650672,
      "loss": 0.0,
      "num_input_tokens_seen": 11370336,
      "step": 20020
    },
    {
      "epoch": 351.3185840707965,
      "grad_norm": 2.0559735247616118e-08,
      "learning_rate": 0.14971725682861076,
      "loss": 0.0,
      "num_input_tokens_seen": 11373200,
      "step": 20025
    },
    {
      "epoch": 351.4070796460177,
      "grad_norm": 1.4595137542983139e-08,
      "learning_rate": 0.14965835209431738,
      "loss": 0.0,
      "num_input_tokens_seen": 11375888,
      "step": 20030
    },
    {
      "epoch": 351.49557522123894,
      "grad_norm": 2.2873990701555158e-08,
      "learning_rate": 0.14959944741271036,
      "loss": 0.0,
      "num_input_tokens_seen": 11379152,
      "step": 20035
    },
    {
      "epoch": 351.5840707964602,
      "grad_norm": 3.106718438061762e-08,
      "learning_rate": 0.14954054279287363,
      "loss": 0.0,
      "num_input_tokens_seen": 11382288,
      "step": 20040
    },
    {
      "epoch": 351.6725663716814,
      "grad_norm": 2.885827754539605e-08,
      "learning_rate": 0.14948163824389094,
      "loss": 0.0,
      "num_input_tokens_seen": 11385344,
      "step": 20045
    },
    {
      "epoch": 351.76106194690266,
      "grad_norm": 1.5551080423392705e-08,
      "learning_rate": 0.14942273377484613,
      "loss": 0.0,
      "num_input_tokens_seen": 11387920,
      "step": 20050
    },
    {
      "epoch": 351.8495575221239,
      "grad_norm": 1.496023749325559e-08,
      "learning_rate": 0.1493638293948231,
      "loss": 0.0,
      "num_input_tokens_seen": 11390768,
      "step": 20055
    },
    {
      "epoch": 351.9380530973451,
      "grad_norm": 2.177559288440989e-08,
      "learning_rate": 0.14930492511290547,
      "loss": 0.0,
      "num_input_tokens_seen": 11393536,
      "step": 20060
    },
    {
      "epoch": 352.01769911504425,
      "grad_norm": 3.477894949810434e-08,
      "learning_rate": 0.14924602093817715,
      "loss": 0.0,
      "num_input_tokens_seen": 11396296,
      "step": 20065
    },
    {
      "epoch": 352.1061946902655,
      "grad_norm": 1.0943250750017341e-08,
      "learning_rate": 0.14918711687972194,
      "loss": 0.0,
      "num_input_tokens_seen": 11398824,
      "step": 20070
    },
    {
      "epoch": 352.1946902654867,
      "grad_norm": 1.57301709435842e-08,
      "learning_rate": 0.14912821294662346,
      "loss": 0.0,
      "num_input_tokens_seen": 11401512,
      "step": 20075
    },
    {
      "epoch": 352.283185840708,
      "grad_norm": 3.671521042747372e-08,
      "learning_rate": 0.14906930914796554,
      "loss": 0.0,
      "num_input_tokens_seen": 11404296,
      "step": 20080
    },
    {
      "epoch": 352.37168141592923,
      "grad_norm": 9.178738658022212e-09,
      "learning_rate": 0.14901040549283182,
      "loss": 0.0,
      "num_input_tokens_seen": 11407624,
      "step": 20085
    },
    {
      "epoch": 352.46017699115043,
      "grad_norm": 2.327243997513051e-08,
      "learning_rate": 0.148951501990306,
      "loss": 0.0,
      "num_input_tokens_seen": 11410984,
      "step": 20090
    },
    {
      "epoch": 352.5486725663717,
      "grad_norm": 2.5867796793477282e-08,
      "learning_rate": 0.14889259864947177,
      "loss": 0.0,
      "num_input_tokens_seen": 11413432,
      "step": 20095
    },
    {
      "epoch": 352.6371681415929,
      "grad_norm": 2.2985474856795918e-08,
      "learning_rate": 0.14883369547941272,
      "loss": 0.0,
      "num_input_tokens_seen": 11416440,
      "step": 20100
    },
    {
      "epoch": 352.72566371681415,
      "grad_norm": 1.8896523457101466e-08,
      "learning_rate": 0.14877479248921247,
      "loss": 0.0,
      "num_input_tokens_seen": 11419496,
      "step": 20105
    },
    {
      "epoch": 352.8141592920354,
      "grad_norm": 2.7942634872601957e-08,
      "learning_rate": 0.14871588968795468,
      "loss": 0.0,
      "num_input_tokens_seen": 11422424,
      "step": 20110
    },
    {
      "epoch": 352.9026548672566,
      "grad_norm": 1.1993463999715459e-08,
      "learning_rate": 0.1486569870847228,
      "loss": 0.0,
      "num_input_tokens_seen": 11425432,
      "step": 20115
    },
    {
      "epoch": 352.9911504424779,
      "grad_norm": 2.5993870167440036e-08,
      "learning_rate": 0.1485980846886004,
      "loss": 0.0,
      "num_input_tokens_seen": 11427976,
      "step": 20120
    },
    {
      "epoch": 353.070796460177,
      "grad_norm": 3.382660906936508e-08,
      "learning_rate": 0.14853918250867096,
      "loss": 0.0,
      "num_input_tokens_seen": 11430200,
      "step": 20125
    },
    {
      "epoch": 353.1592920353982,
      "grad_norm": 1.622153078528754e-08,
      "learning_rate": 0.1484802805540179,
      "loss": 0.0,
      "num_input_tokens_seen": 11432936,
      "step": 20130
    },
    {
      "epoch": 353.24778761061947,
      "grad_norm": 2.3705801766027434e-08,
      "learning_rate": 0.14842137883372472,
      "loss": 0.0,
      "num_input_tokens_seen": 11435432,
      "step": 20135
    },
    {
      "epoch": 353.3362831858407,
      "grad_norm": 1.5101049299914848e-08,
      "learning_rate": 0.14836247735687474,
      "loss": 0.0,
      "num_input_tokens_seen": 11438088,
      "step": 20140
    },
    {
      "epoch": 353.42477876106193,
      "grad_norm": 1.514597336438328e-08,
      "learning_rate": 0.14830357613255132,
      "loss": 0.0,
      "num_input_tokens_seen": 11441128,
      "step": 20145
    },
    {
      "epoch": 353.5132743362832,
      "grad_norm": 1.516309566795826e-08,
      "learning_rate": 0.1482446751698378,
      "loss": 0.0,
      "num_input_tokens_seen": 11444248,
      "step": 20150
    },
    {
      "epoch": 353.60176991150445,
      "grad_norm": 2.0525597221876524e-08,
      "learning_rate": 0.14818577447781744,
      "loss": 0.0,
      "num_input_tokens_seen": 11447304,
      "step": 20155
    },
    {
      "epoch": 353.69026548672565,
      "grad_norm": 1.3730550918467088e-08,
      "learning_rate": 0.14812687406557346,
      "loss": 0.0,
      "num_input_tokens_seen": 11450680,
      "step": 20160
    },
    {
      "epoch": 353.7787610619469,
      "grad_norm": 1.5615817972047807e-08,
      "learning_rate": 0.14806797394218899,
      "loss": 0.0,
      "num_input_tokens_seen": 11453080,
      "step": 20165
    },
    {
      "epoch": 353.86725663716817,
      "grad_norm": 1.3216395089443722e-08,
      "learning_rate": 0.1480090741167472,
      "loss": 0.0,
      "num_input_tokens_seen": 11456312,
      "step": 20170
    },
    {
      "epoch": 353.95575221238937,
      "grad_norm": 1.9768100045780557e-08,
      "learning_rate": 0.1479501745983313,
      "loss": 0.0,
      "num_input_tokens_seen": 11459368,
      "step": 20175
    },
    {
      "epoch": 354.0353982300885,
      "grad_norm": 1.852753150899389e-08,
      "learning_rate": 0.14789127539602415,
      "loss": 0.0,
      "num_input_tokens_seen": 11461496,
      "step": 20180
    },
    {
      "epoch": 354.12389380530976,
      "grad_norm": 1.1314675418816478e-08,
      "learning_rate": 0.14783237651890885,
      "loss": 0.0,
      "num_input_tokens_seen": 11464104,
      "step": 20185
    },
    {
      "epoch": 354.21238938053096,
      "grad_norm": 3.445992646788909e-08,
      "learning_rate": 0.14777347797606838,
      "loss": 0.0,
      "num_input_tokens_seen": 11466936,
      "step": 20190
    },
    {
      "epoch": 354.3008849557522,
      "grad_norm": 1.269147809068727e-08,
      "learning_rate": 0.14771457977658553,
      "loss": 0.0,
      "num_input_tokens_seen": 11469336,
      "step": 20195
    },
    {
      "epoch": 354.3893805309734,
      "grad_norm": 2.4480131699533558e-08,
      "learning_rate": 0.14765568192954326,
      "loss": 0.0,
      "num_input_tokens_seen": 11471832,
      "step": 20200
    },
    {
      "epoch": 354.3893805309734,
      "eval_loss": 0.6238388419151306,
      "eval_runtime": 0.93,
      "eval_samples_per_second": 26.883,
      "eval_steps_per_second": 13.979,
      "num_input_tokens_seen": 11471832,
      "step": 20200
    },
    {
      "epoch": 354.4778761061947,
      "grad_norm": 2.1439962694103087e-08,
      "learning_rate": 0.14759678444402421,
      "loss": 0.0,
      "num_input_tokens_seen": 11475144,
      "step": 20205
    },
    {
      "epoch": 354.56637168141594,
      "grad_norm": 2.163777956809554e-08,
      "learning_rate": 0.14753788732911122,
      "loss": 0.0,
      "num_input_tokens_seen": 11478120,
      "step": 20210
    },
    {
      "epoch": 354.65486725663715,
      "grad_norm": 9.657230570780939e-09,
      "learning_rate": 0.147478990593887,
      "loss": 0.0,
      "num_input_tokens_seen": 11481352,
      "step": 20215
    },
    {
      "epoch": 354.7433628318584,
      "grad_norm": 1.3311927560266668e-08,
      "learning_rate": 0.14742009424743405,
      "loss": 0.0,
      "num_input_tokens_seen": 11484456,
      "step": 20220
    },
    {
      "epoch": 354.83185840707966,
      "grad_norm": 1.3496157080794546e-08,
      "learning_rate": 0.14736119829883504,
      "loss": 0.0,
      "num_input_tokens_seen": 11487720,
      "step": 20225
    },
    {
      "epoch": 354.92035398230087,
      "grad_norm": 2.0098564590398382e-08,
      "learning_rate": 0.14730230275717243,
      "loss": 0.0,
      "num_input_tokens_seen": 11490536,
      "step": 20230
    },
    {
      "epoch": 355.0,
      "grad_norm": 9.99363347631288e-09,
      "learning_rate": 0.14724340763152854,
      "loss": 0.0,
      "num_input_tokens_seen": 11492832,
      "step": 20235
    },
    {
      "epoch": 355.08849557522126,
      "grad_norm": 9.158651614882274e-09,
      "learning_rate": 0.14718451293098594,
      "loss": 0.0,
      "num_input_tokens_seen": 11496256,
      "step": 20240
    },
    {
      "epoch": 355.17699115044246,
      "grad_norm": 2.4637675011263127e-08,
      "learning_rate": 0.14712561866462676,
      "loss": 0.0,
      "num_input_tokens_seen": 11499040,
      "step": 20245
    },
    {
      "epoch": 355.2654867256637,
      "grad_norm": 1.3754963390510966e-08,
      "learning_rate": 0.1470667248415333,
      "loss": 0.0,
      "num_input_tokens_seen": 11501456,
      "step": 20250
    },
    {
      "epoch": 355.353982300885,
      "grad_norm": 2.2510526775931794e-08,
      "learning_rate": 0.1470078314707878,
      "loss": 0.0,
      "num_input_tokens_seen": 11504528,
      "step": 20255
    },
    {
      "epoch": 355.4424778761062,
      "grad_norm": 7.898822040885989e-09,
      "learning_rate": 0.14694893856147223,
      "loss": 0.0,
      "num_input_tokens_seen": 11507200,
      "step": 20260
    },
    {
      "epoch": 355.53097345132744,
      "grad_norm": 1.7335350932512483e-08,
      "learning_rate": 0.14689004612266868,
      "loss": 0.0,
      "num_input_tokens_seen": 11510096,
      "step": 20265
    },
    {
      "epoch": 355.6194690265487,
      "grad_norm": 3.0791362348736584e-08,
      "learning_rate": 0.14683115416345913,
      "loss": 0.0,
      "num_input_tokens_seen": 11513056,
      "step": 20270
    },
    {
      "epoch": 355.7079646017699,
      "grad_norm": 9.169610848402954e-09,
      "learning_rate": 0.1467722626929254,
      "loss": 0.0,
      "num_input_tokens_seen": 11515824,
      "step": 20275
    },
    {
      "epoch": 355.79646017699116,
      "grad_norm": 1.8513620858584545e-08,
      "learning_rate": 0.14671337172014937,
      "loss": 0.0,
      "num_input_tokens_seen": 11519072,
      "step": 20280
    },
    {
      "epoch": 355.88495575221236,
      "grad_norm": 1.9270988360631236e-08,
      "learning_rate": 0.14665448125421265,
      "loss": 0.0,
      "num_input_tokens_seen": 11521392,
      "step": 20285
    },
    {
      "epoch": 355.9734513274336,
      "grad_norm": 3.6038255046833e-08,
      "learning_rate": 0.146595591304197,
      "loss": 0.0,
      "num_input_tokens_seen": 11524080,
      "step": 20290
    },
    {
      "epoch": 356.05309734513276,
      "grad_norm": 1.063880716856147e-08,
      "learning_rate": 0.14653670187918397,
      "loss": 0.0,
      "num_input_tokens_seen": 11527064,
      "step": 20295
    },
    {
      "epoch": 356.14159292035396,
      "grad_norm": 1.5941923336981745e-08,
      "learning_rate": 0.14647781298825502,
      "loss": 0.0,
      "num_input_tokens_seen": 11529688,
      "step": 20300
    },
    {
      "epoch": 356.2300884955752,
      "grad_norm": 1.2657390691117598e-08,
      "learning_rate": 0.14641892464049153,
      "loss": 0.0,
      "num_input_tokens_seen": 11532456,
      "step": 20305
    },
    {
      "epoch": 356.3185840707965,
      "grad_norm": 2.55356660261441e-08,
      "learning_rate": 0.14636003684497495,
      "loss": 0.0,
      "num_input_tokens_seen": 11535272,
      "step": 20310
    },
    {
      "epoch": 356.4070796460177,
      "grad_norm": 1.1450010717339865e-08,
      "learning_rate": 0.14630114961078636,
      "loss": 0.0,
      "num_input_tokens_seen": 11537848,
      "step": 20315
    },
    {
      "epoch": 356.49557522123894,
      "grad_norm": 2.804111076670779e-08,
      "learning_rate": 0.14624226294700704,
      "loss": 0.0,
      "num_input_tokens_seen": 11541256,
      "step": 20320
    },
    {
      "epoch": 356.5840707964602,
      "grad_norm": 2.2149244216507213e-08,
      "learning_rate": 0.14618337686271793,
      "loss": 0.0,
      "num_input_tokens_seen": 11544008,
      "step": 20325
    },
    {
      "epoch": 356.6725663716814,
      "grad_norm": 3.9670840834560295e-08,
      "learning_rate": 0.1461244913670001,
      "loss": 0.0,
      "num_input_tokens_seen": 11546904,
      "step": 20330
    },
    {
      "epoch": 356.76106194690266,
      "grad_norm": 2.5760337862834604e-08,
      "learning_rate": 0.1460656064689344,
      "loss": 0.0,
      "num_input_tokens_seen": 11550104,
      "step": 20335
    },
    {
      "epoch": 356.8495575221239,
      "grad_norm": 2.5297724803863275e-08,
      "learning_rate": 0.14600672217760163,
      "loss": 0.0,
      "num_input_tokens_seen": 11553176,
      "step": 20340
    },
    {
      "epoch": 356.9380530973451,
      "grad_norm": 2.0139134804253445e-08,
      "learning_rate": 0.14594783850208248,
      "loss": 0.0,
      "num_input_tokens_seen": 11556024,
      "step": 20345
    },
    {
      "epoch": 357.01769911504425,
      "grad_norm": 1.9212611945818026e-08,
      "learning_rate": 0.14588895545145758,
      "loss": 0.0,
      "num_input_tokens_seen": 11558240,
      "step": 20350
    },
    {
      "epoch": 357.1061946902655,
      "grad_norm": 2.1487098322836573e-08,
      "learning_rate": 0.14583007303480738,
      "loss": 0.0,
      "num_input_tokens_seen": 11561456,
      "step": 20355
    },
    {
      "epoch": 357.1946902654867,
      "grad_norm": 3.394022485281312e-08,
      "learning_rate": 0.14577119126121235,
      "loss": 0.0,
      "num_input_tokens_seen": 11563968,
      "step": 20360
    },
    {
      "epoch": 357.283185840708,
      "grad_norm": 1.5895770033580447e-08,
      "learning_rate": 0.14571231013975272,
      "loss": 0.0,
      "num_input_tokens_seen": 11566320,
      "step": 20365
    },
    {
      "epoch": 357.37168141592923,
      "grad_norm": 1.2906404833756824e-08,
      "learning_rate": 0.1456534296795088,
      "loss": 0.0,
      "num_input_tokens_seen": 11569376,
      "step": 20370
    },
    {
      "epoch": 357.46017699115043,
      "grad_norm": 1.385275183451995e-08,
      "learning_rate": 0.14559454988956066,
      "loss": 0.0,
      "num_input_tokens_seen": 11572320,
      "step": 20375
    },
    {
      "epoch": 357.5486725663717,
      "grad_norm": 1.1197165861176472e-08,
      "learning_rate": 0.1455356707789882,
      "loss": 0.0,
      "num_input_tokens_seen": 11575168,
      "step": 20380
    },
    {
      "epoch": 357.6371681415929,
      "grad_norm": 2.6624070059710903e-08,
      "learning_rate": 0.14547679235687147,
      "loss": 0.0,
      "num_input_tokens_seen": 11577792,
      "step": 20385
    },
    {
      "epoch": 357.72566371681415,
      "grad_norm": 2.4388182140455683e-08,
      "learning_rate": 0.14541791463229023,
      "loss": 0.0,
      "num_input_tokens_seen": 11580224,
      "step": 20390
    },
    {
      "epoch": 357.8141592920354,
      "grad_norm": 1.6642278666267885e-08,
      "learning_rate": 0.14535903761432406,
      "loss": 0.0,
      "num_input_tokens_seen": 11583024,
      "step": 20395
    },
    {
      "epoch": 357.9026548672566,
      "grad_norm": 1.5433137434683886e-08,
      "learning_rate": 0.1453001613120527,
      "loss": 0.0,
      "num_input_tokens_seen": 11586368,
      "step": 20400
    },
    {
      "epoch": 357.9026548672566,
      "eval_loss": 0.6087567210197449,
      "eval_runtime": 0.9178,
      "eval_samples_per_second": 27.239,
      "eval_steps_per_second": 14.164,
      "num_input_tokens_seen": 11586368,
      "step": 20400
    },
    {
      "epoch": 357.9911504424779,
      "grad_norm": 1.1285513856762464e-08,
      "learning_rate": 0.14524128573455547,
      "loss": 0.0,
      "num_input_tokens_seen": 11589504,
      "step": 20405
    },
    {
      "epoch": 358.070796460177,
      "grad_norm": 7.447683803007976e-09,
      "learning_rate": 0.14518241089091177,
      "loss": 0.0,
      "num_input_tokens_seen": 11592016,
      "step": 20410
    },
    {
      "epoch": 358.1592920353982,
      "grad_norm": 2.1202788857976884e-08,
      "learning_rate": 0.1451235367902009,
      "loss": 0.0,
      "num_input_tokens_seen": 11595216,
      "step": 20415
    },
    {
      "epoch": 358.24778761061947,
      "grad_norm": 1.1243178832387457e-08,
      "learning_rate": 0.1450646634415019,
      "loss": 0.0,
      "num_input_tokens_seen": 11597664,
      "step": 20420
    },
    {
      "epoch": 358.3362831858407,
      "grad_norm": 7.382387146037672e-09,
      "learning_rate": 0.1450057908538938,
      "loss": 0.0,
      "num_input_tokens_seen": 11600880,
      "step": 20425
    },
    {
      "epoch": 358.42477876106193,
      "grad_norm": 3.167939510717588e-08,
      "learning_rate": 0.14494691903645557,
      "loss": 0.0,
      "num_input_tokens_seen": 11603632,
      "step": 20430
    },
    {
      "epoch": 358.5132743362832,
      "grad_norm": 8.832552467197274e-09,
      "learning_rate": 0.14488804799826588,
      "loss": 0.0,
      "num_input_tokens_seen": 11606288,
      "step": 20435
    },
    {
      "epoch": 358.60176991150445,
      "grad_norm": 1.685761930048102e-08,
      "learning_rate": 0.14482917774840348,
      "loss": 0.0,
      "num_input_tokens_seen": 11609392,
      "step": 20440
    },
    {
      "epoch": 358.69026548672565,
      "grad_norm": 2.694374323652937e-08,
      "learning_rate": 0.14477030829594684,
      "loss": 0.0,
      "num_input_tokens_seen": 11612576,
      "step": 20445
    },
    {
      "epoch": 358.7787610619469,
      "grad_norm": 2.065279325336178e-08,
      "learning_rate": 0.14471143964997432,
      "loss": 0.0,
      "num_input_tokens_seen": 11615040,
      "step": 20450
    },
    {
      "epoch": 358.86725663716817,
      "grad_norm": 7.151824465978507e-09,
      "learning_rate": 0.14465257181956434,
      "loss": 0.0,
      "num_input_tokens_seen": 11618288,
      "step": 20455
    },
    {
      "epoch": 358.95575221238937,
      "grad_norm": 2.2720939796272432e-08,
      "learning_rate": 0.1445937048137949,
      "loss": 0.0,
      "num_input_tokens_seen": 11621008,
      "step": 20460
    },
    {
      "epoch": 359.0353982300885,
      "grad_norm": 1.068551824801034e-08,
      "learning_rate": 0.14453483864174416,
      "loss": 0.0,
      "num_input_tokens_seen": 11623472,
      "step": 20465
    },
    {
      "epoch": 359.12389380530976,
      "grad_norm": 1.9408155083056045e-08,
      "learning_rate": 0.14447597331249,
      "loss": 0.0,
      "num_input_tokens_seen": 11626176,
      "step": 20470
    },
    {
      "epoch": 359.21238938053096,
      "grad_norm": 1.6891400278495894e-08,
      "learning_rate": 0.1444171088351102,
      "loss": 0.0,
      "num_input_tokens_seen": 11629216,
      "step": 20475
    },
    {
      "epoch": 359.3008849557522,
      "grad_norm": 2.1678845385508794e-08,
      "learning_rate": 0.14435824521868235,
      "loss": 0.0,
      "num_input_tokens_seen": 11632352,
      "step": 20480
    },
    {
      "epoch": 359.3893805309734,
      "grad_norm": 1.0348856882558266e-08,
      "learning_rate": 0.14429938247228397,
      "loss": 0.0,
      "num_input_tokens_seen": 11635024,
      "step": 20485
    },
    {
      "epoch": 359.4778761061947,
      "grad_norm": 9.239399467730891e-09,
      "learning_rate": 0.14424052060499243,
      "loss": 0.0,
      "num_input_tokens_seen": 11637376,
      "step": 20490
    },
    {
      "epoch": 359.56637168141594,
      "grad_norm": 3.583798502404534e-08,
      "learning_rate": 0.14418165962588506,
      "loss": 0.0,
      "num_input_tokens_seen": 11640704,
      "step": 20495
    },
    {
      "epoch": 359.65486725663715,
      "grad_norm": 1.3518898889230968e-08,
      "learning_rate": 0.1441227995440388,
      "loss": 0.0,
      "num_input_tokens_seen": 11643760,
      "step": 20500
    },
    {
      "epoch": 359.7433628318584,
      "grad_norm": 2.2768530172356805e-08,
      "learning_rate": 0.14406394036853082,
      "loss": 0.0,
      "num_input_tokens_seen": 11646656,
      "step": 20505
    },
    {
      "epoch": 359.83185840707966,
      "grad_norm": 1.662072790509228e-08,
      "learning_rate": 0.14400508210843774,
      "loss": 0.0,
      "num_input_tokens_seen": 11649552,
      "step": 20510
    },
    {
      "epoch": 359.92035398230087,
      "grad_norm": 2.6292163113339484e-08,
      "learning_rate": 0.1439462247728364,
      "loss": 0.0,
      "num_input_tokens_seen": 11652464,
      "step": 20515
    },
    {
      "epoch": 360.0,
      "grad_norm": 9.84210579701994e-09,
      "learning_rate": 0.14388736837080326,
      "loss": 0.0,
      "num_input_tokens_seen": 11655312,
      "step": 20520
    },
    {
      "epoch": 360.08849557522126,
      "grad_norm": 1.759813628154916e-08,
      "learning_rate": 0.14382851291141469,
      "loss": 0.0,
      "num_input_tokens_seen": 11657760,
      "step": 20525
    },
    {
      "epoch": 360.17699115044246,
      "grad_norm": 9.746605300620104e-09,
      "learning_rate": 0.14376965840374697,
      "loss": 0.0,
      "num_input_tokens_seen": 11660656,
      "step": 20530
    },
    {
      "epoch": 360.2654867256637,
      "grad_norm": 6.28666407820333e-09,
      "learning_rate": 0.14371080485687632,
      "loss": 0.0,
      "num_input_tokens_seen": 11663344,
      "step": 20535
    },
    {
      "epoch": 360.353982300885,
      "grad_norm": 2.008805921605017e-08,
      "learning_rate": 0.1436519522798785,
      "loss": 0.0,
      "num_input_tokens_seen": 11666432,
      "step": 20540
    },
    {
      "epoch": 360.4424778761062,
      "grad_norm": 9.438396730843124e-09,
      "learning_rate": 0.14359310068182948,
      "loss": 0.0,
      "num_input_tokens_seen": 11669008,
      "step": 20545
    },
    {
      "epoch": 360.53097345132744,
      "grad_norm": 1.4786361468566156e-08,
      "learning_rate": 0.14353425007180484,
      "loss": 0.0,
      "num_input_tokens_seen": 11672096,
      "step": 20550
    },
    {
      "epoch": 360.6194690265487,
      "grad_norm": 2.2518527487136453e-08,
      "learning_rate": 0.14347540045888005,
      "loss": 0.0,
      "num_input_tokens_seen": 11675552,
      "step": 20555
    },
    {
      "epoch": 360.7079646017699,
      "grad_norm": 1.297989982163017e-08,
      "learning_rate": 0.14341655185213056,
      "loss": 0.0,
      "num_input_tokens_seen": 11678272,
      "step": 20560
    },
    {
      "epoch": 360.79646017699116,
      "grad_norm": 1.2962787288017807e-08,
      "learning_rate": 0.14335770426063144,
      "loss": 0.0,
      "num_input_tokens_seen": 11681472,
      "step": 20565
    },
    {
      "epoch": 360.88495575221236,
      "grad_norm": 2.027835854789828e-08,
      "learning_rate": 0.1432988576934578,
      "loss": 0.0,
      "num_input_tokens_seen": 11684272,
      "step": 20570
    },
    {
      "epoch": 360.9734513274336,
      "grad_norm": 3.5156059396967976e-08,
      "learning_rate": 0.14324001215968457,
      "loss": 0.0,
      "num_input_tokens_seen": 11687008,
      "step": 20575
    },
    {
      "epoch": 361.05309734513276,
      "grad_norm": 2.0488860386080887e-08,
      "learning_rate": 0.14318116766838637,
      "loss": 0.0,
      "num_input_tokens_seen": 11688944,
      "step": 20580
    },
    {
      "epoch": 361.14159292035396,
      "grad_norm": 8.345012680877062e-09,
      "learning_rate": 0.14312232422863788,
      "loss": 0.0,
      "num_input_tokens_seen": 11692240,
      "step": 20585
    },
    {
      "epoch": 361.2300884955752,
      "grad_norm": 2.1546897599478143e-08,
      "learning_rate": 0.14306348184951334,
      "loss": 0.0,
      "num_input_tokens_seen": 11694816,
      "step": 20590
    },
    {
      "epoch": 361.3185840707965,
      "grad_norm": 1.5939853881263844e-08,
      "learning_rate": 0.1430046405400871,
      "loss": 0.0,
      "num_input_tokens_seen": 11697648,
      "step": 20595
    },
    {
      "epoch": 361.4070796460177,
      "grad_norm": 3.3308676705701146e-08,
      "learning_rate": 0.14294580030943324,
      "loss": 0.0,
      "num_input_tokens_seen": 11700176,
      "step": 20600
    },
    {
      "epoch": 361.4070796460177,
      "eval_loss": 0.6196402311325073,
      "eval_runtime": 0.9291,
      "eval_samples_per_second": 26.907,
      "eval_steps_per_second": 13.992,
      "num_input_tokens_seen": 11700176,
      "step": 20600
    },
    {
      "epoch": 361.49557522123894,
      "grad_norm": 1.69791132265118e-08,
      "learning_rate": 0.14288696116662553,
      "loss": 0.0,
      "num_input_tokens_seen": 11703296,
      "step": 20605
    },
    {
      "epoch": 361.5840707964602,
      "grad_norm": 1.6609629227559708e-08,
      "learning_rate": 0.1428281231207378,
      "loss": 0.0,
      "num_input_tokens_seen": 11705904,
      "step": 20610
    },
    {
      "epoch": 361.6725663716814,
      "grad_norm": 1.3674884336012383e-08,
      "learning_rate": 0.1427692861808437,
      "loss": 0.0,
      "num_input_tokens_seen": 11708496,
      "step": 20615
    },
    {
      "epoch": 361.76106194690266,
      "grad_norm": 1.4731148745283917e-08,
      "learning_rate": 0.1427104503560165,
      "loss": 0.0,
      "num_input_tokens_seen": 11711728,
      "step": 20620
    },
    {
      "epoch": 361.8495575221239,
      "grad_norm": 2.2985393144381305e-08,
      "learning_rate": 0.14265161565532947,
      "loss": 0.0,
      "num_input_tokens_seen": 11714848,
      "step": 20625
    },
    {
      "epoch": 361.9380530973451,
      "grad_norm": 1.4958857263991376e-08,
      "learning_rate": 0.14259278208785564,
      "loss": 0.0,
      "num_input_tokens_seen": 11718352,
      "step": 20630
    },
    {
      "epoch": 362.01769911504425,
      "grad_norm": 7.226466092191686e-09,
      "learning_rate": 0.14253394966266789,
      "loss": 0.0,
      "num_input_tokens_seen": 11720688,
      "step": 20635
    },
    {
      "epoch": 362.1061946902655,
      "grad_norm": 3.1422491275634457e-08,
      "learning_rate": 0.14247511838883894,
      "loss": 0.0,
      "num_input_tokens_seen": 11723040,
      "step": 20640
    },
    {
      "epoch": 362.1946902654867,
      "grad_norm": 5.127037461249984e-09,
      "learning_rate": 0.14241628827544126,
      "loss": 0.0,
      "num_input_tokens_seen": 11725872,
      "step": 20645
    },
    {
      "epoch": 362.283185840708,
      "grad_norm": 2.8290012110687712e-08,
      "learning_rate": 0.14235745933154723,
      "loss": 0.0,
      "num_input_tokens_seen": 11728960,
      "step": 20650
    },
    {
      "epoch": 362.37168141592923,
      "grad_norm": 2.0837447323174274e-08,
      "learning_rate": 0.14229863156622907,
      "loss": 0.0,
      "num_input_tokens_seen": 11732144,
      "step": 20655
    },
    {
      "epoch": 362.46017699115043,
      "grad_norm": 1.6062312369058418e-08,
      "learning_rate": 0.14223980498855868,
      "loss": 0.0,
      "num_input_tokens_seen": 11735424,
      "step": 20660
    },
    {
      "epoch": 362.5486725663717,
      "grad_norm": 8.104629856120482e-09,
      "learning_rate": 0.14218097960760792,
      "loss": 0.0,
      "num_input_tokens_seen": 11738144,
      "step": 20665
    },
    {
      "epoch": 362.6371681415929,
      "grad_norm": 1.0961080043614402e-08,
      "learning_rate": 0.1421221554324483,
      "loss": 0.0,
      "num_input_tokens_seen": 11740528,
      "step": 20670
    },
    {
      "epoch": 362.72566371681415,
      "grad_norm": 1.638603563947072e-08,
      "learning_rate": 0.1420633324721513,
      "loss": 0.0,
      "num_input_tokens_seen": 11743344,
      "step": 20675
    },
    {
      "epoch": 362.8141592920354,
      "grad_norm": 4.821150412226416e-08,
      "learning_rate": 0.14200451073578824,
      "loss": 0.0,
      "num_input_tokens_seen": 11746624,
      "step": 20680
    },
    {
      "epoch": 362.9026548672566,
      "grad_norm": 8.894887493227088e-09,
      "learning_rate": 0.14194569023243003,
      "loss": 0.0,
      "num_input_tokens_seen": 11749920,
      "step": 20685
    },
    {
      "epoch": 362.9911504424779,
      "grad_norm": 6.4860552484447e-09,
      "learning_rate": 0.14188687097114766,
      "loss": 0.0,
      "num_input_tokens_seen": 11752432,
      "step": 20690
    },
    {
      "epoch": 363.070796460177,
      "grad_norm": 1.8422387171312948e-08,
      "learning_rate": 0.14182805296101172,
      "loss": 0.0,
      "num_input_tokens_seen": 11754384,
      "step": 20695
    },
    {
      "epoch": 363.1592920353982,
      "grad_norm": 1.2956578032685684e-08,
      "learning_rate": 0.14176923621109272,
      "loss": 0.0,
      "num_input_tokens_seen": 11756912,
      "step": 20700
    },
    {
      "epoch": 363.24778761061947,
      "grad_norm": 1.1690784340601112e-08,
      "learning_rate": 0.14171042073046097,
      "loss": 0.0,
      "num_input_tokens_seen": 11760048,
      "step": 20705
    },
    {
      "epoch": 363.3362831858407,
      "grad_norm": 1.245918124226364e-08,
      "learning_rate": 0.14165160652818642,
      "loss": 0.0,
      "num_input_tokens_seen": 11762960,
      "step": 20710
    },
    {
      "epoch": 363.42477876106193,
      "grad_norm": 1.4610498588751852e-08,
      "learning_rate": 0.14159279361333907,
      "loss": 0.0,
      "num_input_tokens_seen": 11766320,
      "step": 20715
    },
    {
      "epoch": 363.5132743362832,
      "grad_norm": 9.162421044095481e-09,
      "learning_rate": 0.14153398199498868,
      "loss": 0.0,
      "num_input_tokens_seen": 11769264,
      "step": 20720
    },
    {
      "epoch": 363.60176991150445,
      "grad_norm": 1.8742936092053242e-08,
      "learning_rate": 0.14147517168220458,
      "loss": 0.0,
      "num_input_tokens_seen": 11771776,
      "step": 20725
    },
    {
      "epoch": 363.69026548672565,
      "grad_norm": 1.258982518237417e-08,
      "learning_rate": 0.14141636268405616,
      "loss": 0.0,
      "num_input_tokens_seen": 11774432,
      "step": 20730
    },
    {
      "epoch": 363.7787610619469,
      "grad_norm": 1.524202986047385e-08,
      "learning_rate": 0.14135755500961253,
      "loss": 0.0,
      "num_input_tokens_seen": 11778000,
      "step": 20735
    },
    {
      "epoch": 363.86725663716817,
      "grad_norm": 1.4824502514443338e-08,
      "learning_rate": 0.14129874866794245,
      "loss": 0.0,
      "num_input_tokens_seen": 11781072,
      "step": 20740
    },
    {
      "epoch": 363.95575221238937,
      "grad_norm": 1.343240718654215e-08,
      "learning_rate": 0.14123994366811476,
      "loss": 0.0,
      "num_input_tokens_seen": 11783888,
      "step": 20745
    },
    {
      "epoch": 364.0353982300885,
      "grad_norm": 1.400528581996241e-08,
      "learning_rate": 0.14118114001919774,
      "loss": 0.0,
      "num_input_tokens_seen": 11786048,
      "step": 20750
    },
    {
      "epoch": 364.12389380530976,
      "grad_norm": 1.634225910152054e-08,
      "learning_rate": 0.14112233773025978,
      "loss": 0.0,
      "num_input_tokens_seen": 11788784,
      "step": 20755
    },
    {
      "epoch": 364.21238938053096,
      "grad_norm": 1.6255725654445996e-08,
      "learning_rate": 0.14106353681036896,
      "loss": 0.0,
      "num_input_tokens_seen": 11791712,
      "step": 20760
    },
    {
      "epoch": 364.3008849557522,
      "grad_norm": 2.4176024737698754e-08,
      "learning_rate": 0.14100473726859303,
      "loss": 0.0,
      "num_input_tokens_seen": 11794320,
      "step": 20765
    },
    {
      "epoch": 364.3893805309734,
      "grad_norm": 1.9456436461950943e-08,
      "learning_rate": 0.14094593911399964,
      "loss": 0.0,
      "num_input_tokens_seen": 11796768,
      "step": 20770
    },
    {
      "epoch": 364.4778761061947,
      "grad_norm": 1.141694117023917e-08,
      "learning_rate": 0.14088714235565625,
      "loss": 0.0,
      "num_input_tokens_seen": 11799904,
      "step": 20775
    },
    {
      "epoch": 364.56637168141594,
      "grad_norm": 2.160816059415538e-08,
      "learning_rate": 0.14082834700263,
      "loss": 0.0,
      "num_input_tokens_seen": 11802240,
      "step": 20780
    },
    {
      "epoch": 364.65486725663715,
      "grad_norm": 1.620752243525203e-08,
      "learning_rate": 0.14076955306398795,
      "loss": 0.0,
      "num_input_tokens_seen": 11805440,
      "step": 20785
    },
    {
      "epoch": 364.7433628318584,
      "grad_norm": 2.049148584148952e-08,
      "learning_rate": 0.14071076054879675,
      "loss": 0.0,
      "num_input_tokens_seen": 11808352,
      "step": 20790
    },
    {
      "epoch": 364.83185840707966,
      "grad_norm": 2.029076107135097e-08,
      "learning_rate": 0.14065196946612302,
      "loss": 0.0,
      "num_input_tokens_seen": 11811280,
      "step": 20795
    },
    {
      "epoch": 364.92035398230087,
      "grad_norm": 1.4978921214492402e-08,
      "learning_rate": 0.1405931798250331,
      "loss": 0.0,
      "num_input_tokens_seen": 11814304,
      "step": 20800
    },
    {
      "epoch": 364.92035398230087,
      "eval_loss": 0.6082650423049927,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 26.718,
      "eval_steps_per_second": 13.893,
      "num_input_tokens_seen": 11814304,
      "step": 20800
    },
    {
      "epoch": 365.0,
      "grad_norm": 3.75432840371559e-08,
      "learning_rate": 0.14053439163459308,
      "loss": 0.0,
      "num_input_tokens_seen": 11817184,
      "step": 20805
    },
    {
      "epoch": 365.08849557522126,
      "grad_norm": 1.164159524336128e-08,
      "learning_rate": 0.14047560490386876,
      "loss": 0.0,
      "num_input_tokens_seen": 11819888,
      "step": 20810
    },
    {
      "epoch": 365.17699115044246,
      "grad_norm": 1.2403885030209949e-08,
      "learning_rate": 0.14041681964192593,
      "loss": 0.0,
      "num_input_tokens_seen": 11822624,
      "step": 20815
    },
    {
      "epoch": 365.2654867256637,
      "grad_norm": 9.751704332927602e-09,
      "learning_rate": 0.14035803585782988,
      "loss": 0.0,
      "num_input_tokens_seen": 11825344,
      "step": 20820
    },
    {
      "epoch": 365.353982300885,
      "grad_norm": 2.8926244510785182e-08,
      "learning_rate": 0.14029925356064593,
      "loss": 0.0,
      "num_input_tokens_seen": 11828384,
      "step": 20825
    },
    {
      "epoch": 365.4424778761062,
      "grad_norm": 1.834217222551615e-08,
      "learning_rate": 0.1402404727594389,
      "loss": 0.0,
      "num_input_tokens_seen": 11831248,
      "step": 20830
    },
    {
      "epoch": 365.53097345132744,
      "grad_norm": 1.491800638575569e-08,
      "learning_rate": 0.1401816934632737,
      "loss": 0.0,
      "num_input_tokens_seen": 11834224,
      "step": 20835
    },
    {
      "epoch": 365.6194690265487,
      "grad_norm": 2.5070377773772634e-08,
      "learning_rate": 0.1401229156812147,
      "loss": 0.0,
      "num_input_tokens_seen": 11837504,
      "step": 20840
    },
    {
      "epoch": 365.7079646017699,
      "grad_norm": 7.69395036570586e-09,
      "learning_rate": 0.14006413942232626,
      "loss": 0.0,
      "num_input_tokens_seen": 11840288,
      "step": 20845
    },
    {
      "epoch": 365.79646017699116,
      "grad_norm": 1.3463584913608884e-08,
      "learning_rate": 0.14000536469567235,
      "loss": 0.0,
      "num_input_tokens_seen": 11842832,
      "step": 20850
    },
    {
      "epoch": 365.88495575221236,
      "grad_norm": 2.059219283978564e-08,
      "learning_rate": 0.13994659151031685,
      "loss": 0.0,
      "num_input_tokens_seen": 11846192,
      "step": 20855
    },
    {
      "epoch": 365.9734513274336,
      "grad_norm": 1.409799210705387e-08,
      "learning_rate": 0.13988781987532323,
      "loss": 0.0,
      "num_input_tokens_seen": 11849232,
      "step": 20860
    },
    {
      "epoch": 366.05309734513276,
      "grad_norm": 1.047312192525851e-08,
      "learning_rate": 0.1398290497997549,
      "loss": 0.0,
      "num_input_tokens_seen": 11851592,
      "step": 20865
    },
    {
      "epoch": 366.14159292035396,
      "grad_norm": 1.5370671846426376e-08,
      "learning_rate": 0.13977028129267488,
      "loss": 0.0,
      "num_input_tokens_seen": 11854664,
      "step": 20870
    },
    {
      "epoch": 366.2300884955752,
      "grad_norm": 1.2457318732117528e-08,
      "learning_rate": 0.13971151436314605,
      "loss": 0.0,
      "num_input_tokens_seen": 11857544,
      "step": 20875
    },
    {
      "epoch": 366.3185840707965,
      "grad_norm": 1.4188492158950794e-08,
      "learning_rate": 0.13965274902023103,
      "loss": 0.0,
      "num_input_tokens_seen": 11860536,
      "step": 20880
    },
    {
      "epoch": 366.4070796460177,
      "grad_norm": 2.4096118877992012e-08,
      "learning_rate": 0.13959398527299208,
      "loss": 0.0,
      "num_input_tokens_seen": 11863272,
      "step": 20885
    },
    {
      "epoch": 366.49557522123894,
      "grad_norm": 2.1936724436955046e-08,
      "learning_rate": 0.13953522313049138,
      "loss": 0.0,
      "num_input_tokens_seen": 11865944,
      "step": 20890
    },
    {
      "epoch": 366.5840707964602,
      "grad_norm": 2.3663115911176646e-08,
      "learning_rate": 0.13947646260179083,
      "loss": 0.0,
      "num_input_tokens_seen": 11868776,
      "step": 20895
    },
    {
      "epoch": 366.6725663716814,
      "grad_norm": 1.8187884975873203e-08,
      "learning_rate": 0.13941770369595194,
      "loss": 0.0,
      "num_input_tokens_seen": 11871816,
      "step": 20900
    },
    {
      "epoch": 366.76106194690266,
      "grad_norm": 2.0509236975385647e-08,
      "learning_rate": 0.1393589464220362,
      "loss": 0.0,
      "num_input_tokens_seen": 11874520,
      "step": 20905
    },
    {
      "epoch": 366.8495575221239,
      "grad_norm": 3.256969449694225e-08,
      "learning_rate": 0.13930019078910455,
      "loss": 0.0,
      "num_input_tokens_seen": 11877288,
      "step": 20910
    },
    {
      "epoch": 366.9380530973451,
      "grad_norm": 1.2754302503026338e-08,
      "learning_rate": 0.139241436806218,
      "loss": 0.0,
      "num_input_tokens_seen": 11880328,
      "step": 20915
    },
    {
      "epoch": 367.01769911504425,
      "grad_norm": 3.205655119131734e-08,
      "learning_rate": 0.13918268448243712,
      "loss": 0.0,
      "num_input_tokens_seen": 11882952,
      "step": 20920
    },
    {
      "epoch": 367.1061946902655,
      "grad_norm": 3.270785242648344e-08,
      "learning_rate": 0.13912393382682217,
      "loss": 0.0,
      "num_input_tokens_seen": 11885848,
      "step": 20925
    },
    {
      "epoch": 367.1946902654867,
      "grad_norm": 8.971103859778395e-09,
      "learning_rate": 0.1390651848484333,
      "loss": 0.0,
      "num_input_tokens_seen": 11888424,
      "step": 20930
    },
    {
      "epoch": 367.283185840708,
      "grad_norm": 2.079596228554692e-08,
      "learning_rate": 0.1390064375563304,
      "loss": 0.0,
      "num_input_tokens_seen": 11891224,
      "step": 20935
    },
    {
      "epoch": 367.37168141592923,
      "grad_norm": 1.870449750640546e-08,
      "learning_rate": 0.13894769195957293,
      "loss": 0.0,
      "num_input_tokens_seen": 11894264,
      "step": 20940
    },
    {
      "epoch": 367.46017699115043,
      "grad_norm": 1.5185953827767662e-08,
      "learning_rate": 0.13888894806722032,
      "loss": 0.0,
      "num_input_tokens_seen": 11896984,
      "step": 20945
    },
    {
      "epoch": 367.5486725663717,
      "grad_norm": 2.0985217119573463e-08,
      "learning_rate": 0.1388302058883315,
      "loss": 0.0,
      "num_input_tokens_seen": 11900184,
      "step": 20950
    },
    {
      "epoch": 367.6371681415929,
      "grad_norm": 2.3187835651583555e-08,
      "learning_rate": 0.13877146543196528,
      "loss": 0.0,
      "num_input_tokens_seen": 11903464,
      "step": 20955
    },
    {
      "epoch": 367.72566371681415,
      "grad_norm": 1.342171707108264e-08,
      "learning_rate": 0.13871272670718027,
      "loss": 0.0,
      "num_input_tokens_seen": 11906056,
      "step": 20960
    },
    {
      "epoch": 367.8141592920354,
      "grad_norm": 1.3380082819480776e-08,
      "learning_rate": 0.13865398972303455,
      "loss": 0.0,
      "num_input_tokens_seen": 11908888,
      "step": 20965
    },
    {
      "epoch": 367.9026548672566,
      "grad_norm": 1.1497662377735196e-08,
      "learning_rate": 0.13859525448858623,
      "loss": 0.0,
      "num_input_tokens_seen": 11911704,
      "step": 20970
    },
    {
      "epoch": 367.9911504424779,
      "grad_norm": 2.6607247960441782e-08,
      "learning_rate": 0.13853652101289304,
      "loss": 0.0,
      "num_input_tokens_seen": 11914248,
      "step": 20975
    },
    {
      "epoch": 368.070796460177,
      "grad_norm": 2.7784947675968397e-08,
      "learning_rate": 0.13847778930501234,
      "loss": 0.0,
      "num_input_tokens_seen": 11916344,
      "step": 20980
    },
    {
      "epoch": 368.1592920353982,
      "grad_norm": 1.4356341893062563e-08,
      "learning_rate": 0.1384190593740013,
      "loss": 0.0,
      "num_input_tokens_seen": 11919304,
      "step": 20985
    },
    {
      "epoch": 368.24778761061947,
      "grad_norm": 2.1583788978318807e-08,
      "learning_rate": 0.13836033122891686,
      "loss": 0.0,
      "num_input_tokens_seen": 11922376,
      "step": 20990
    },
    {
      "epoch": 368.3362831858407,
      "grad_norm": 4.6563082278794354e-09,
      "learning_rate": 0.1383016048788156,
      "loss": 0.0,
      "num_input_tokens_seen": 11924888,
      "step": 20995
    },
    {
      "epoch": 368.42477876106193,
      "grad_norm": 2.4134251930263417e-08,
      "learning_rate": 0.13824288033275392,
      "loss": 0.0,
      "num_input_tokens_seen": 11927464,
      "step": 21000
    },
    {
      "epoch": 368.42477876106193,
      "eval_loss": 0.6165170669555664,
      "eval_runtime": 0.9408,
      "eval_samples_per_second": 26.574,
      "eval_steps_per_second": 13.819,
      "num_input_tokens_seen": 11927464,
      "step": 21000
    },
    {
      "epoch": 368.5132743362832,
      "grad_norm": 1.5906845618474108e-08,
      "learning_rate": 0.1381841575997878,
      "loss": 0.0,
      "num_input_tokens_seen": 11930888,
      "step": 21005
    },
    {
      "epoch": 368.60176991150445,
      "grad_norm": 1.0400287742129422e-08,
      "learning_rate": 0.13812543668897306,
      "loss": 0.0,
      "num_input_tokens_seen": 11933976,
      "step": 21010
    },
    {
      "epoch": 368.69026548672565,
      "grad_norm": 7.253219358460683e-09,
      "learning_rate": 0.13806671760936526,
      "loss": 0.0,
      "num_input_tokens_seen": 11936936,
      "step": 21015
    },
    {
      "epoch": 368.7787610619469,
      "grad_norm": 2.7450271389284353e-08,
      "learning_rate": 0.13800800037001956,
      "loss": 0.0,
      "num_input_tokens_seen": 11939688,
      "step": 21020
    },
    {
      "epoch": 368.86725663716817,
      "grad_norm": 1.5099494987680373e-08,
      "learning_rate": 0.13794928497999087,
      "loss": 0.0,
      "num_input_tokens_seen": 11942616,
      "step": 21025
    },
    {
      "epoch": 368.95575221238937,
      "grad_norm": 1.4149569516064275e-08,
      "learning_rate": 0.1378905714483339,
      "loss": 0.0,
      "num_input_tokens_seen": 11945304,
      "step": 21030
    },
    {
      "epoch": 369.0353982300885,
      "grad_norm": 1.2661311110662155e-08,
      "learning_rate": 0.13783185978410295,
      "loss": 0.0,
      "num_input_tokens_seen": 11947520,
      "step": 21035
    },
    {
      "epoch": 369.12389380530976,
      "grad_norm": 1.2589754128100594e-08,
      "learning_rate": 0.13777314999635218,
      "loss": 0.0,
      "num_input_tokens_seen": 11951280,
      "step": 21040
    },
    {
      "epoch": 369.21238938053096,
      "grad_norm": 8.283661756536276e-09,
      "learning_rate": 0.1377144420941353,
      "loss": 0.0,
      "num_input_tokens_seen": 11954384,
      "step": 21045
    },
    {
      "epoch": 369.3008849557522,
      "grad_norm": 1.2869382892688463e-08,
      "learning_rate": 0.13765573608650586,
      "loss": 0.0,
      "num_input_tokens_seen": 11956880,
      "step": 21050
    },
    {
      "epoch": 369.3893805309734,
      "grad_norm": 2.7905651123205644e-08,
      "learning_rate": 0.13759703198251702,
      "loss": 0.0,
      "num_input_tokens_seen": 11959744,
      "step": 21055
    },
    {
      "epoch": 369.4778761061947,
      "grad_norm": 1.786570535955434e-08,
      "learning_rate": 0.13753832979122174,
      "loss": 0.0,
      "num_input_tokens_seen": 11962656,
      "step": 21060
    },
    {
      "epoch": 369.56637168141594,
      "grad_norm": 1.0682538409412246e-08,
      "learning_rate": 0.13747962952167264,
      "loss": 0.0,
      "num_input_tokens_seen": 11965648,
      "step": 21065
    },
    {
      "epoch": 369.65486725663715,
      "grad_norm": 2.3083101652332516e-08,
      "learning_rate": 0.13742093118292192,
      "loss": 0.0,
      "num_input_tokens_seen": 11968752,
      "step": 21070
    },
    {
      "epoch": 369.7433628318584,
      "grad_norm": 1.9371547921309684e-08,
      "learning_rate": 0.13736223478402174,
      "loss": 0.0,
      "num_input_tokens_seen": 11971440,
      "step": 21075
    },
    {
      "epoch": 369.83185840707966,
      "grad_norm": 2.1973098895955445e-08,
      "learning_rate": 0.1373035403340238,
      "loss": 0.0,
      "num_input_tokens_seen": 11974064,
      "step": 21080
    },
    {
      "epoch": 369.92035398230087,
      "grad_norm": 2.394252440751643e-08,
      "learning_rate": 0.13724484784197943,
      "loss": 0.0,
      "num_input_tokens_seen": 11976816,
      "step": 21085
    },
    {
      "epoch": 370.0,
      "grad_norm": 4.17807655139768e-09,
      "learning_rate": 0.13718615731693987,
      "loss": 0.0,
      "num_input_tokens_seen": 11978976,
      "step": 21090
    },
    {
      "epoch": 370.08849557522126,
      "grad_norm": 2.0835162928278805e-08,
      "learning_rate": 0.13712746876795587,
      "loss": 0.0,
      "num_input_tokens_seen": 11981376,
      "step": 21095
    },
    {
      "epoch": 370.17699115044246,
      "grad_norm": 2.188935432911876e-08,
      "learning_rate": 0.13706878220407792,
      "loss": 0.0,
      "num_input_tokens_seen": 11984576,
      "step": 21100
    },
    {
      "epoch": 370.2654867256637,
      "grad_norm": 1.5311197643086416e-08,
      "learning_rate": 0.13701009763435631,
      "loss": 0.0,
      "num_input_tokens_seen": 11987328,
      "step": 21105
    },
    {
      "epoch": 370.353982300885,
      "grad_norm": 1.0007565443004296e-08,
      "learning_rate": 0.13695141506784084,
      "loss": 0.0,
      "num_input_tokens_seen": 11990720,
      "step": 21110
    },
    {
      "epoch": 370.4424778761062,
      "grad_norm": 1.3415443866904297e-08,
      "learning_rate": 0.13689273451358114,
      "loss": 0.0,
      "num_input_tokens_seen": 11993440,
      "step": 21115
    },
    {
      "epoch": 370.53097345132744,
      "grad_norm": 2.2467157023697837e-08,
      "learning_rate": 0.13683405598062653,
      "loss": 0.0,
      "num_input_tokens_seen": 11996112,
      "step": 21120
    },
    {
      "epoch": 370.6194690265487,
      "grad_norm": 1.4291983596592672e-08,
      "learning_rate": 0.1367753794780259,
      "loss": 0.0,
      "num_input_tokens_seen": 11998688,
      "step": 21125
    },
    {
      "epoch": 370.7079646017699,
      "grad_norm": 2.5027432570823294e-08,
      "learning_rate": 0.13671670501482802,
      "loss": 0.0,
      "num_input_tokens_seen": 12001520,
      "step": 21130
    },
    {
      "epoch": 370.79646017699116,
      "grad_norm": 6.299257560016258e-09,
      "learning_rate": 0.1366580326000811,
      "loss": 0.0,
      "num_input_tokens_seen": 12004768,
      "step": 21135
    },
    {
      "epoch": 370.88495575221236,
      "grad_norm": 1.87700077702857e-08,
      "learning_rate": 0.1365993622428332,
      "loss": 0.0,
      "num_input_tokens_seen": 12007552,
      "step": 21140
    },
    {
      "epoch": 370.9734513274336,
      "grad_norm": 1.1979889080748762e-08,
      "learning_rate": 0.13654069395213211,
      "loss": 0.0,
      "num_input_tokens_seen": 12010352,
      "step": 21145
    },
    {
      "epoch": 371.05309734513276,
      "grad_norm": 2.019115896700896e-08,
      "learning_rate": 0.13648202773702509,
      "loss": 0.0,
      "num_input_tokens_seen": 12012648,
      "step": 21150
    },
    {
      "epoch": 371.14159292035396,
      "grad_norm": 1.9877182566574447e-08,
      "learning_rate": 0.13642336360655927,
      "loss": 0.0,
      "num_input_tokens_seen": 12016232,
      "step": 21155
    },
    {
      "epoch": 371.2300884955752,
      "grad_norm": 1.6089570564759015e-08,
      "learning_rate": 0.13636470156978145,
      "loss": 0.0,
      "num_input_tokens_seen": 12018872,
      "step": 21160
    },
    {
      "epoch": 371.3185840707965,
      "grad_norm": 1.3394274134270745e-08,
      "learning_rate": 0.13630604163573798,
      "loss": 0.0,
      "num_input_tokens_seen": 12021144,
      "step": 21165
    },
    {
      "epoch": 371.4070796460177,
      "grad_norm": 8.188746569715022e-09,
      "learning_rate": 0.13624738381347495,
      "loss": 0.0,
      "num_input_tokens_seen": 12023752,
      "step": 21170
    },
    {
      "epoch": 371.49557522123894,
      "grad_norm": 2.343839078378096e-08,
      "learning_rate": 0.1361887281120382,
      "loss": 0.0,
      "num_input_tokens_seen": 12026648,
      "step": 21175
    },
    {
      "epoch": 371.5840707964602,
      "grad_norm": 1.5552259924334066e-08,
      "learning_rate": 0.13613007454047307,
      "loss": 0.0,
      "num_input_tokens_seen": 12029368,
      "step": 21180
    },
    {
      "epoch": 371.6725663716814,
      "grad_norm": 2.2851432746051614e-08,
      "learning_rate": 0.13607142310782486,
      "loss": 0.0,
      "num_input_tokens_seen": 12032120,
      "step": 21185
    },
    {
      "epoch": 371.76106194690266,
      "grad_norm": 1.3518519637045756e-08,
      "learning_rate": 0.13601277382313814,
      "loss": 0.0,
      "num_input_tokens_seen": 12035000,
      "step": 21190
    },
    {
      "epoch": 371.8495575221239,
      "grad_norm": 2.3419456596229793e-08,
      "learning_rate": 0.1359541266954575,
      "loss": 0.0,
      "num_input_tokens_seen": 12038024,
      "step": 21195
    },
    {
      "epoch": 371.9380530973451,
      "grad_norm": 2.00007761463894e-08,
      "learning_rate": 0.13589548173382707,
      "loss": 0.0,
      "num_input_tokens_seen": 12041416,
      "step": 21200
    },
    {
      "epoch": 371.9380530973451,
      "eval_loss": 0.638322114944458,
      "eval_runtime": 0.9182,
      "eval_samples_per_second": 27.226,
      "eval_steps_per_second": 14.158,
      "num_input_tokens_seen": 12041416,
      "step": 21200
    },
    {
      "epoch": 372.01769911504425,
      "grad_norm": 2.171546142903935e-08,
      "learning_rate": 0.1358368389472906,
      "loss": 0.0,
      "num_input_tokens_seen": 12044088,
      "step": 21205
    },
    {
      "epoch": 372.1061946902655,
      "grad_norm": 1.4542909987369512e-08,
      "learning_rate": 0.13577819834489155,
      "loss": 0.0,
      "num_input_tokens_seen": 12046792,
      "step": 21210
    },
    {
      "epoch": 372.1946902654867,
      "grad_norm": 9.069131223782279e-09,
      "learning_rate": 0.135719559935673,
      "loss": 0.0,
      "num_input_tokens_seen": 12049480,
      "step": 21215
    },
    {
      "epoch": 372.283185840708,
      "grad_norm": 2.242401997420984e-08,
      "learning_rate": 0.13566092372867775,
      "loss": 0.0,
      "num_input_tokens_seen": 12052552,
      "step": 21220
    },
    {
      "epoch": 372.37168141592923,
      "grad_norm": 1.5746298487329113e-08,
      "learning_rate": 0.13560228973294833,
      "loss": 0.0,
      "num_input_tokens_seen": 12055480,
      "step": 21225
    },
    {
      "epoch": 372.46017699115043,
      "grad_norm": 1.7373873006931717e-08,
      "learning_rate": 0.13554365795752668,
      "loss": 0.0,
      "num_input_tokens_seen": 12058232,
      "step": 21230
    },
    {
      "epoch": 372.5486725663717,
      "grad_norm": 1.7579480982021778e-08,
      "learning_rate": 0.1354850284114547,
      "loss": 0.0,
      "num_input_tokens_seen": 12060808,
      "step": 21235
    },
    {
      "epoch": 372.6371681415929,
      "grad_norm": 2.5899092648273836e-08,
      "learning_rate": 0.13542640110377374,
      "loss": 0.0,
      "num_input_tokens_seen": 12063896,
      "step": 21240
    },
    {
      "epoch": 372.72566371681415,
      "grad_norm": 5.975762995547029e-09,
      "learning_rate": 0.13536777604352487,
      "loss": 0.0,
      "num_input_tokens_seen": 12066968,
      "step": 21245
    },
    {
      "epoch": 372.8141592920354,
      "grad_norm": 1.8985764071999256e-08,
      "learning_rate": 0.13530915323974887,
      "loss": 0.0,
      "num_input_tokens_seen": 12070200,
      "step": 21250
    },
    {
      "epoch": 372.9026548672566,
      "grad_norm": 2.3709528562676496e-08,
      "learning_rate": 0.13525053270148596,
      "loss": 0.0,
      "num_input_tokens_seen": 12072728,
      "step": 21255
    },
    {
      "epoch": 372.9911504424779,
      "grad_norm": 2.2514997866096564e-08,
      "learning_rate": 0.13519191443777628,
      "loss": 0.0,
      "num_input_tokens_seen": 12075480,
      "step": 21260
    },
    {
      "epoch": 373.070796460177,
      "grad_norm": 1.7400649809928836e-08,
      "learning_rate": 0.13513329845765953,
      "loss": 0.0,
      "num_input_tokens_seen": 12077736,
      "step": 21265
    },
    {
      "epoch": 373.1592920353982,
      "grad_norm": 5.443795814130681e-08,
      "learning_rate": 0.13507468477017495,
      "loss": 0.0,
      "num_input_tokens_seen": 12080472,
      "step": 21270
    },
    {
      "epoch": 373.24778761061947,
      "grad_norm": 1.7439512944861235e-08,
      "learning_rate": 0.13501607338436153,
      "loss": 0.0,
      "num_input_tokens_seen": 12083224,
      "step": 21275
    },
    {
      "epoch": 373.3362831858407,
      "grad_norm": 3.0176924070701716e-08,
      "learning_rate": 0.13495746430925798,
      "loss": 0.0,
      "num_input_tokens_seen": 12086216,
      "step": 21280
    },
    {
      "epoch": 373.42477876106193,
      "grad_norm": 2.4948381138756304e-08,
      "learning_rate": 0.13489885755390238,
      "loss": 0.0,
      "num_input_tokens_seen": 12088696,
      "step": 21285
    },
    {
      "epoch": 373.5132743362832,
      "grad_norm": 1.145363626164908e-08,
      "learning_rate": 0.13484025312733275,
      "loss": 0.0,
      "num_input_tokens_seen": 12091384,
      "step": 21290
    },
    {
      "epoch": 373.60176991150445,
      "grad_norm": 1.3110120988812923e-08,
      "learning_rate": 0.13478165103858658,
      "loss": 0.0,
      "num_input_tokens_seen": 12094376,
      "step": 21295
    },
    {
      "epoch": 373.69026548672565,
      "grad_norm": 1.1916478470652692e-08,
      "learning_rate": 0.13472305129670106,
      "loss": 0.0,
      "num_input_tokens_seen": 12097176,
      "step": 21300
    },
    {
      "epoch": 373.7787610619469,
      "grad_norm": 1.7910361194140023e-08,
      "learning_rate": 0.13466445391071305,
      "loss": 0.0,
      "num_input_tokens_seen": 12100488,
      "step": 21305
    },
    {
      "epoch": 373.86725663716817,
      "grad_norm": 1.3258038222829782e-08,
      "learning_rate": 0.13460585888965895,
      "loss": 0.0,
      "num_input_tokens_seen": 12103288,
      "step": 21310
    },
    {
      "epoch": 373.95575221238937,
      "grad_norm": 8.428695963402788e-09,
      "learning_rate": 0.13454726624257482,
      "loss": 0.0,
      "num_input_tokens_seen": 12106168,
      "step": 21315
    },
    {
      "epoch": 374.0353982300885,
      "grad_norm": 1.939895888369847e-08,
      "learning_rate": 0.1344886759784965,
      "loss": 0.0,
      "num_input_tokens_seen": 12108248,
      "step": 21320
    },
    {
      "epoch": 374.12389380530976,
      "grad_norm": 1.2268477789234566e-08,
      "learning_rate": 0.13443008810645923,
      "loss": 0.0,
      "num_input_tokens_seen": 12111720,
      "step": 21325
    },
    {
      "epoch": 374.21238938053096,
      "grad_norm": 1.4299653017246783e-08,
      "learning_rate": 0.13437150263549807,
      "loss": 0.0,
      "num_input_tokens_seen": 12115000,
      "step": 21330
    },
    {
      "epoch": 374.3008849557522,
      "grad_norm": 2.141700505831068e-08,
      "learning_rate": 0.13431291957464755,
      "loss": 0.0,
      "num_input_tokens_seen": 12117720,
      "step": 21335
    },
    {
      "epoch": 374.3893805309734,
      "grad_norm": 9.425385805172937e-09,
      "learning_rate": 0.13425433893294197,
      "loss": 0.0,
      "num_input_tokens_seen": 12120152,
      "step": 21340
    },
    {
      "epoch": 374.4778761061947,
      "grad_norm": 8.487467617612765e-09,
      "learning_rate": 0.13419576071941525,
      "loss": 0.0,
      "num_input_tokens_seen": 12122616,
      "step": 21345
    },
    {
      "epoch": 374.56637168141594,
      "grad_norm": 1.8586508332418816e-08,
      "learning_rate": 0.1341371849431008,
      "loss": 0.0,
      "num_input_tokens_seen": 12125672,
      "step": 21350
    },
    {
      "epoch": 374.65486725663715,
      "grad_norm": 2.0426911717663643e-08,
      "learning_rate": 0.13407861161303178,
      "loss": 0.0,
      "num_input_tokens_seen": 12128456,
      "step": 21355
    },
    {
      "epoch": 374.7433628318584,
      "grad_norm": 1.650841596756436e-08,
      "learning_rate": 0.13402004073824098,
      "loss": 0.0,
      "num_input_tokens_seen": 12131288,
      "step": 21360
    },
    {
      "epoch": 374.83185840707966,
      "grad_norm": 9.593998484547228e-09,
      "learning_rate": 0.13396147232776062,
      "loss": 0.0,
      "num_input_tokens_seen": 12133960,
      "step": 21365
    },
    {
      "epoch": 374.92035398230087,
      "grad_norm": 2.178178171163836e-08,
      "learning_rate": 0.13390290639062288,
      "loss": 0.0,
      "num_input_tokens_seen": 12136904,
      "step": 21370
    },
    {
      "epoch": 375.0,
      "grad_norm": 5.793886259652936e-09,
      "learning_rate": 0.13384434293585917,
      "loss": 0.0,
      "num_input_tokens_seen": 12139256,
      "step": 21375
    },
    {
      "epoch": 375.08849557522126,
      "grad_norm": 1.2354564482564001e-08,
      "learning_rate": 0.13378578197250088,
      "loss": 0.0,
      "num_input_tokens_seen": 12142568,
      "step": 21380
    },
    {
      "epoch": 375.17699115044246,
      "grad_norm": 2.7288256987390014e-08,
      "learning_rate": 0.13372722350957872,
      "loss": 0.0,
      "num_input_tokens_seen": 12145368,
      "step": 21385
    },
    {
      "epoch": 375.2654867256637,
      "grad_norm": 6.562236087859219e-09,
      "learning_rate": 0.13366866755612322,
      "loss": 0.0,
      "num_input_tokens_seen": 12148312,
      "step": 21390
    },
    {
      "epoch": 375.353982300885,
      "grad_norm": 1.0230097657881743e-08,
      "learning_rate": 0.13361011412116436,
      "loss": 0.0,
      "num_input_tokens_seen": 12150776,
      "step": 21395
    },
    {
      "epoch": 375.4424778761062,
      "grad_norm": 1.4556003513632731e-08,
      "learning_rate": 0.13355156321373196,
      "loss": 0.0,
      "num_input_tokens_seen": 12153176,
      "step": 21400
    },
    {
      "epoch": 375.4424778761062,
      "eval_loss": 0.6363808512687683,
      "eval_runtime": 0.9402,
      "eval_samples_per_second": 26.59,
      "eval_steps_per_second": 13.827,
      "num_input_tokens_seen": 12153176,
      "step": 21400
    },
    {
      "epoch": 375.53097345132744,
      "grad_norm": 1.7307209887462705e-08,
      "learning_rate": 0.13349301484285514,
      "loss": 0.0,
      "num_input_tokens_seen": 12155576,
      "step": 21405
    },
    {
      "epoch": 375.6194690265487,
      "grad_norm": 2.8373651872470873e-08,
      "learning_rate": 0.13343446901756295,
      "loss": 0.0,
      "num_input_tokens_seen": 12158952,
      "step": 21410
    },
    {
      "epoch": 375.7079646017699,
      "grad_norm": 1.715056896500755e-08,
      "learning_rate": 0.13337592574688376,
      "loss": 0.0,
      "num_input_tokens_seen": 12162232,
      "step": 21415
    },
    {
      "epoch": 375.79646017699116,
      "grad_norm": 1.413116113013757e-08,
      "learning_rate": 0.13331738503984572,
      "loss": 0.0,
      "num_input_tokens_seen": 12165160,
      "step": 21420
    },
    {
      "epoch": 375.88495575221236,
      "grad_norm": 1.3669526843784752e-08,
      "learning_rate": 0.1332588469054766,
      "loss": 0.0,
      "num_input_tokens_seen": 12167704,
      "step": 21425
    },
    {
      "epoch": 375.9734513274336,
      "grad_norm": 1.7933281526438805e-08,
      "learning_rate": 0.1332003113528036,
      "loss": 0.0,
      "num_input_tokens_seen": 12170808,
      "step": 21430
    },
    {
      "epoch": 376.05309734513276,
      "grad_norm": 1.0285207352467296e-08,
      "learning_rate": 0.13314177839085373,
      "loss": 0.0,
      "num_input_tokens_seen": 12173112,
      "step": 21435
    },
    {
      "epoch": 376.14159292035396,
      "grad_norm": 1.235726365678147e-08,
      "learning_rate": 0.13308324802865354,
      "loss": 0.0,
      "num_input_tokens_seen": 12175672,
      "step": 21440
    },
    {
      "epoch": 376.2300884955752,
      "grad_norm": 1.0793367088979267e-08,
      "learning_rate": 0.13302472027522905,
      "loss": 0.0,
      "num_input_tokens_seen": 12178584,
      "step": 21445
    },
    {
      "epoch": 376.3185840707965,
      "grad_norm": 2.4013241173292954e-08,
      "learning_rate": 0.13296619513960606,
      "loss": 0.0,
      "num_input_tokens_seen": 12181048,
      "step": 21450
    },
    {
      "epoch": 376.4070796460177,
      "grad_norm": 1.1807909316985388e-08,
      "learning_rate": 0.1329076726308098,
      "loss": 0.0,
      "num_input_tokens_seen": 12183640,
      "step": 21455
    },
    {
      "epoch": 376.49557522123894,
      "grad_norm": 1.4625924471545204e-08,
      "learning_rate": 0.13284915275786519,
      "loss": 0.0,
      "num_input_tokens_seen": 12187160,
      "step": 21460
    },
    {
      "epoch": 376.5840707964602,
      "grad_norm": 1.1211493067264655e-08,
      "learning_rate": 0.1327906355297968,
      "loss": 0.0,
      "num_input_tokens_seen": 12190152,
      "step": 21465
    },
    {
      "epoch": 376.6725663716814,
      "grad_norm": 2.3921648661939798e-08,
      "learning_rate": 0.13273212095562867,
      "loss": 0.0,
      "num_input_tokens_seen": 12193528,
      "step": 21470
    },
    {
      "epoch": 376.76106194690266,
      "grad_norm": 2.7123686408003778e-08,
      "learning_rate": 0.13267360904438444,
      "loss": 0.0,
      "num_input_tokens_seen": 12196360,
      "step": 21475
    },
    {
      "epoch": 376.8495575221239,
      "grad_norm": 2.3246665037390812e-08,
      "learning_rate": 0.1326150998050875,
      "loss": 0.0,
      "num_input_tokens_seen": 12199016,
      "step": 21480
    },
    {
      "epoch": 376.9380530973451,
      "grad_norm": 1.3751527916383566e-08,
      "learning_rate": 0.1325565932467606,
      "loss": 0.0,
      "num_input_tokens_seen": 12202232,
      "step": 21485
    },
    {
      "epoch": 377.01769911504425,
      "grad_norm": 1.3110107666136628e-08,
      "learning_rate": 0.13249808937842628,
      "loss": 0.0,
      "num_input_tokens_seen": 12204616,
      "step": 21490
    },
    {
      "epoch": 377.1061946902655,
      "grad_norm": 1.4684764515493498e-08,
      "learning_rate": 0.1324395882091065,
      "loss": 0.0,
      "num_input_tokens_seen": 12207736,
      "step": 21495
    },
    {
      "epoch": 377.1946902654867,
      "grad_norm": 1.403149596512776e-08,
      "learning_rate": 0.13238108974782284,
      "loss": 0.0,
      "num_input_tokens_seen": 12210952,
      "step": 21500
    },
    {
      "epoch": 377.283185840708,
      "grad_norm": 1.7238258820384544e-08,
      "learning_rate": 0.13232259400359664,
      "loss": 0.0,
      "num_input_tokens_seen": 12214168,
      "step": 21505
    },
    {
      "epoch": 377.37168141592923,
      "grad_norm": 1.4966405004201988e-08,
      "learning_rate": 0.13226410098544852,
      "loss": 0.0,
      "num_input_tokens_seen": 12216920,
      "step": 21510
    },
    {
      "epoch": 377.46017699115043,
      "grad_norm": 1.6609872588446706e-08,
      "learning_rate": 0.13220561070239892,
      "loss": 0.0,
      "num_input_tokens_seen": 12219864,
      "step": 21515
    },
    {
      "epoch": 377.5486725663717,
      "grad_norm": 8.190641942462662e-09,
      "learning_rate": 0.13214712316346783,
      "loss": 0.0,
      "num_input_tokens_seen": 12223000,
      "step": 21520
    },
    {
      "epoch": 377.6371681415929,
      "grad_norm": 3.6088383836840876e-08,
      "learning_rate": 0.13208863837767465,
      "loss": 0.0,
      "num_input_tokens_seen": 12225368,
      "step": 21525
    },
    {
      "epoch": 377.72566371681415,
      "grad_norm": 3.074510956935228e-08,
      "learning_rate": 0.13203015635403856,
      "loss": 0.0,
      "num_input_tokens_seen": 12228056,
      "step": 21530
    },
    {
      "epoch": 377.8141592920354,
      "grad_norm": 1.28132358057087e-08,
      "learning_rate": 0.13197167710157817,
      "loss": 0.0,
      "num_input_tokens_seen": 12230952,
      "step": 21535
    },
    {
      "epoch": 377.9026548672566,
      "grad_norm": 2.937786724999114e-08,
      "learning_rate": 0.13191320062931167,
      "loss": 0.0,
      "num_input_tokens_seen": 12233944,
      "step": 21540
    },
    {
      "epoch": 377.9911504424779,
      "grad_norm": 1.872537680469577e-08,
      "learning_rate": 0.13185472694625702,
      "loss": 0.0,
      "num_input_tokens_seen": 12236552,
      "step": 21545
    },
    {
      "epoch": 378.070796460177,
      "grad_norm": 1.3543206556221321e-08,
      "learning_rate": 0.13179625606143142,
      "loss": 0.0,
      "num_input_tokens_seen": 12238784,
      "step": 21550
    },
    {
      "epoch": 378.1592920353982,
      "grad_norm": 2.1033050856544833e-08,
      "learning_rate": 0.13173778798385188,
      "loss": 0.0,
      "num_input_tokens_seen": 12241248,
      "step": 21555
    },
    {
      "epoch": 378.24778761061947,
      "grad_norm": 7.65906005284478e-09,
      "learning_rate": 0.13167932272253505,
      "loss": 0.0,
      "num_input_tokens_seen": 12244688,
      "step": 21560
    },
    {
      "epoch": 378.3362831858407,
      "grad_norm": 2.2610937122635733e-08,
      "learning_rate": 0.1316208602864968,
      "loss": 0.0,
      "num_input_tokens_seen": 12247472,
      "step": 21565
    },
    {
      "epoch": 378.42477876106193,
      "grad_norm": 1.4434825779119365e-08,
      "learning_rate": 0.13156240068475292,
      "loss": 0.0,
      "num_input_tokens_seen": 12250176,
      "step": 21570
    },
    {
      "epoch": 378.5132743362832,
      "grad_norm": 2.5421051930152316e-08,
      "learning_rate": 0.1315039439263185,
      "loss": 0.0,
      "num_input_tokens_seen": 12253104,
      "step": 21575
    },
    {
      "epoch": 378.60176991150445,
      "grad_norm": 2.615980676523577e-08,
      "learning_rate": 0.13144549002020833,
      "loss": 0.0,
      "num_input_tokens_seen": 12256016,
      "step": 21580
    },
    {
      "epoch": 378.69026548672565,
      "grad_norm": 4.778605600108676e-08,
      "learning_rate": 0.13138703897543688,
      "loss": 0.0,
      "num_input_tokens_seen": 12259088,
      "step": 21585
    },
    {
      "epoch": 378.7787610619469,
      "grad_norm": 1.6752553122501013e-08,
      "learning_rate": 0.1313285908010178,
      "loss": 0.0,
      "num_input_tokens_seen": 12261824,
      "step": 21590
    },
    {
      "epoch": 378.86725663716817,
      "grad_norm": 2.9572781556908012e-08,
      "learning_rate": 0.13127014550596475,
      "loss": 0.0,
      "num_input_tokens_seen": 12265008,
      "step": 21595
    },
    {
      "epoch": 378.95575221238937,
      "grad_norm": 1.7024936127540968e-08,
      "learning_rate": 0.1312117030992906,
      "loss": 0.0,
      "num_input_tokens_seen": 12267984,
      "step": 21600
    },
    {
      "epoch": 378.95575221238937,
      "eval_loss": 0.6360908150672913,
      "eval_runtime": 0.9394,
      "eval_samples_per_second": 26.613,
      "eval_steps_per_second": 13.839,
      "num_input_tokens_seen": 12267984,
      "step": 21600
    },
    {
      "epoch": 379.0353982300885,
      "grad_norm": 2.4077191795868202e-08,
      "learning_rate": 0.13115326359000795,
      "loss": 0.0,
      "num_input_tokens_seen": 12270560,
      "step": 21605
    },
    {
      "epoch": 379.12389380530976,
      "grad_norm": 2.1546009421058443e-08,
      "learning_rate": 0.13109482698712896,
      "loss": 0.0,
      "num_input_tokens_seen": 12273808,
      "step": 21610
    },
    {
      "epoch": 379.21238938053096,
      "grad_norm": 2.1392050797430784e-08,
      "learning_rate": 0.1310363932996651,
      "loss": 0.0,
      "num_input_tokens_seen": 12276736,
      "step": 21615
    },
    {
      "epoch": 379.3008849557522,
      "grad_norm": 8.689410080364723e-09,
      "learning_rate": 0.13097796253662775,
      "loss": 0.0,
      "num_input_tokens_seen": 12279600,
      "step": 21620
    },
    {
      "epoch": 379.3893805309734,
      "grad_norm": 3.103157908412868e-08,
      "learning_rate": 0.1309195347070277,
      "loss": 0.0,
      "num_input_tokens_seen": 12282304,
      "step": 21625
    },
    {
      "epoch": 379.4778761061947,
      "grad_norm": 2.063122650497462e-08,
      "learning_rate": 0.13086110981987506,
      "loss": 0.0,
      "num_input_tokens_seen": 12285040,
      "step": 21630
    },
    {
      "epoch": 379.56637168141594,
      "grad_norm": 1.059758059085425e-08,
      "learning_rate": 0.13080268788417987,
      "loss": 0.0,
      "num_input_tokens_seen": 12287648,
      "step": 21635
    },
    {
      "epoch": 379.65486725663715,
      "grad_norm": 1.220715528660321e-08,
      "learning_rate": 0.1307442689089515,
      "loss": 0.0,
      "num_input_tokens_seen": 12290288,
      "step": 21640
    },
    {
      "epoch": 379.7433628318584,
      "grad_norm": 2.5657818980562297e-08,
      "learning_rate": 0.13068585290319873,
      "loss": 0.0,
      "num_input_tokens_seen": 12293376,
      "step": 21645
    },
    {
      "epoch": 379.83185840707966,
      "grad_norm": 4.2601719485446665e-08,
      "learning_rate": 0.13062743987593026,
      "loss": 0.0,
      "num_input_tokens_seen": 12295840,
      "step": 21650
    },
    {
      "epoch": 379.92035398230087,
      "grad_norm": 1.530250592907123e-08,
      "learning_rate": 0.13056902983615395,
      "loss": 0.0,
      "num_input_tokens_seen": 12298528,
      "step": 21655
    },
    {
      "epoch": 380.0,
      "grad_norm": 5.174288730813714e-08,
      "learning_rate": 0.13051062279287742,
      "loss": 0.0,
      "num_input_tokens_seen": 12301360,
      "step": 21660
    },
    {
      "epoch": 380.08849557522126,
      "grad_norm": 1.237156688205232e-08,
      "learning_rate": 0.13045221875510782,
      "loss": 0.0,
      "num_input_tokens_seen": 12304800,
      "step": 21665
    },
    {
      "epoch": 380.17699115044246,
      "grad_norm": 4.118275498399271e-08,
      "learning_rate": 0.13039381773185174,
      "loss": 0.0,
      "num_input_tokens_seen": 12307664,
      "step": 21670
    },
    {
      "epoch": 380.2654867256637,
      "grad_norm": 2.3269141280479744e-08,
      "learning_rate": 0.1303354197321153,
      "loss": 0.0,
      "num_input_tokens_seen": 12310624,
      "step": 21675
    },
    {
      "epoch": 380.353982300885,
      "grad_norm": 6.980287015068143e-09,
      "learning_rate": 0.13027702476490433,
      "loss": 0.0,
      "num_input_tokens_seen": 12313376,
      "step": 21680
    },
    {
      "epoch": 380.4424778761062,
      "grad_norm": 1.6939646130254005e-08,
      "learning_rate": 0.1302186328392239,
      "loss": 0.0,
      "num_input_tokens_seen": 12316112,
      "step": 21685
    },
    {
      "epoch": 380.53097345132744,
      "grad_norm": 2.1661577420672984e-08,
      "learning_rate": 0.130160243964079,
      "loss": 0.0,
      "num_input_tokens_seen": 12318848,
      "step": 21690
    },
    {
      "epoch": 380.6194690265487,
      "grad_norm": 1.3422196687429278e-08,
      "learning_rate": 0.13010185814847372,
      "loss": 0.0,
      "num_input_tokens_seen": 12321536,
      "step": 21695
    },
    {
      "epoch": 380.7079646017699,
      "grad_norm": 4.966408173601167e-09,
      "learning_rate": 0.13004347540141192,
      "loss": 0.0,
      "num_input_tokens_seen": 12324336,
      "step": 21700
    },
    {
      "epoch": 380.79646017699116,
      "grad_norm": 5.923156720655243e-08,
      "learning_rate": 0.12998509573189712,
      "loss": 0.0,
      "num_input_tokens_seen": 12327280,
      "step": 21705
    },
    {
      "epoch": 380.88495575221236,
      "grad_norm": 1.3902946349730883e-08,
      "learning_rate": 0.12992671914893203,
      "loss": 0.0,
      "num_input_tokens_seen": 12330336,
      "step": 21710
    },
    {
      "epoch": 380.9734513274336,
      "grad_norm": 1.6472800012934385e-08,
      "learning_rate": 0.12986834566151909,
      "loss": 0.0,
      "num_input_tokens_seen": 12333232,
      "step": 21715
    },
    {
      "epoch": 381.05309734513276,
      "grad_norm": 8.530534501005604e-09,
      "learning_rate": 0.12980997527866028,
      "loss": 0.0,
      "num_input_tokens_seen": 12335352,
      "step": 21720
    },
    {
      "epoch": 381.14159292035396,
      "grad_norm": 2.3939721316423856e-08,
      "learning_rate": 0.12975160800935692,
      "loss": 0.0,
      "num_input_tokens_seen": 12338376,
      "step": 21725
    },
    {
      "epoch": 381.2300884955752,
      "grad_norm": 3.075766130677948e-08,
      "learning_rate": 0.12969324386261016,
      "loss": 0.0,
      "num_input_tokens_seen": 12341416,
      "step": 21730
    },
    {
      "epoch": 381.3185840707965,
      "grad_norm": 1.764943213800052e-08,
      "learning_rate": 0.12963488284742034,
      "loss": 0.0,
      "num_input_tokens_seen": 12344568,
      "step": 21735
    },
    {
      "epoch": 381.4070796460177,
      "grad_norm": 2.8743913915718622e-08,
      "learning_rate": 0.12957652497278752,
      "loss": 0.0,
      "num_input_tokens_seen": 12347592,
      "step": 21740
    },
    {
      "epoch": 381.49557522123894,
      "grad_norm": 1.731848264796554e-08,
      "learning_rate": 0.12951817024771117,
      "loss": 0.0,
      "num_input_tokens_seen": 12350808,
      "step": 21745
    },
    {
      "epoch": 381.5840707964602,
      "grad_norm": 1.4398705339147e-08,
      "learning_rate": 0.12945981868119041,
      "loss": 0.0,
      "num_input_tokens_seen": 12353304,
      "step": 21750
    },
    {
      "epoch": 381.6725663716814,
      "grad_norm": 1.8289671999127677e-08,
      "learning_rate": 0.12940147028222376,
      "loss": 0.0,
      "num_input_tokens_seen": 12356184,
      "step": 21755
    },
    {
      "epoch": 381.76106194690266,
      "grad_norm": 1.3997897951867344e-08,
      "learning_rate": 0.12934312505980916,
      "loss": 0.0,
      "num_input_tokens_seen": 12359080,
      "step": 21760
    },
    {
      "epoch": 381.8495575221239,
      "grad_norm": 2.4316523905554277e-08,
      "learning_rate": 0.1292847830229443,
      "loss": 0.0,
      "num_input_tokens_seen": 12361832,
      "step": 21765
    },
    {
      "epoch": 381.9380530973451,
      "grad_norm": 1.1131783494988667e-08,
      "learning_rate": 0.12922644418062626,
      "loss": 0.0,
      "num_input_tokens_seen": 12364424,
      "step": 21770
    },
    {
      "epoch": 382.01769911504425,
      "grad_norm": 4.7455838370069614e-08,
      "learning_rate": 0.1291681085418515,
      "loss": 0.0,
      "num_input_tokens_seen": 12366928,
      "step": 21775
    },
    {
      "epoch": 382.1061946902655,
      "grad_norm": 2.7103070010525698e-08,
      "learning_rate": 0.12910977611561628,
      "loss": 0.0,
      "num_input_tokens_seen": 12369616,
      "step": 21780
    },
    {
      "epoch": 382.1946902654867,
      "grad_norm": 1.4873815068483509e-08,
      "learning_rate": 0.1290514469109161,
      "loss": 0.0,
      "num_input_tokens_seen": 12372720,
      "step": 21785
    },
    {
      "epoch": 382.283185840708,
      "grad_norm": 1.6079168219107487e-08,
      "learning_rate": 0.128993120936746,
      "loss": 0.0,
      "num_input_tokens_seen": 12375616,
      "step": 21790
    },
    {
      "epoch": 382.37168141592923,
      "grad_norm": 7.752157138440907e-09,
      "learning_rate": 0.12893479820210071,
      "loss": 0.0,
      "num_input_tokens_seen": 12378528,
      "step": 21795
    },
    {
      "epoch": 382.46017699115043,
      "grad_norm": 5.334298780468316e-09,
      "learning_rate": 0.1288764787159742,
      "loss": 0.0,
      "num_input_tokens_seen": 12381424,
      "step": 21800
    },
    {
      "epoch": 382.46017699115043,
      "eval_loss": 0.6419970393180847,
      "eval_runtime": 0.9242,
      "eval_samples_per_second": 27.051,
      "eval_steps_per_second": 14.066,
      "num_input_tokens_seen": 12381424,
      "step": 21800
    },
    {
      "epoch": 382.5486725663717,
      "grad_norm": 1.9997443700958684e-08,
      "learning_rate": 0.1288181624873601,
      "loss": 0.0,
      "num_input_tokens_seen": 12384224,
      "step": 21805
    },
    {
      "epoch": 382.6371681415929,
      "grad_norm": 1.7583657196951208e-08,
      "learning_rate": 0.12875984952525163,
      "loss": 0.0,
      "num_input_tokens_seen": 12387072,
      "step": 21810
    },
    {
      "epoch": 382.72566371681415,
      "grad_norm": 6.896881288298573e-09,
      "learning_rate": 0.12870153983864122,
      "loss": 0.0,
      "num_input_tokens_seen": 12389856,
      "step": 21815
    },
    {
      "epoch": 382.8141592920354,
      "grad_norm": 1.2451110364963824e-08,
      "learning_rate": 0.12864323343652104,
      "loss": 0.0,
      "num_input_tokens_seen": 12392464,
      "step": 21820
    },
    {
      "epoch": 382.9026548672566,
      "grad_norm": 2.6352012127972557e-08,
      "learning_rate": 0.12858493032788268,
      "loss": 0.0,
      "num_input_tokens_seen": 12395520,
      "step": 21825
    },
    {
      "epoch": 382.9911504424779,
      "grad_norm": 2.041293889476492e-08,
      "learning_rate": 0.12852663052171714,
      "loss": 0.0,
      "num_input_tokens_seen": 12398432,
      "step": 21830
    },
    {
      "epoch": 383.070796460177,
      "grad_norm": 6.577066447022162e-09,
      "learning_rate": 0.12846833402701507,
      "loss": 0.0,
      "num_input_tokens_seen": 12400704,
      "step": 21835
    },
    {
      "epoch": 383.1592920353982,
      "grad_norm": 4.9643680277711155e-09,
      "learning_rate": 0.12841004085276642,
      "loss": 0.0,
      "num_input_tokens_seen": 12403776,
      "step": 21840
    },
    {
      "epoch": 383.24778761061947,
      "grad_norm": 9.945113177423082e-09,
      "learning_rate": 0.12835175100796076,
      "loss": 0.0,
      "num_input_tokens_seen": 12406576,
      "step": 21845
    },
    {
      "epoch": 383.3362831858407,
      "grad_norm": 1.7421944775719567e-08,
      "learning_rate": 0.12829346450158724,
      "loss": 0.0,
      "num_input_tokens_seen": 12409088,
      "step": 21850
    },
    {
      "epoch": 383.42477876106193,
      "grad_norm": 1.527706494641734e-08,
      "learning_rate": 0.12823518134263423,
      "loss": 0.0,
      "num_input_tokens_seen": 12411952,
      "step": 21855
    },
    {
      "epoch": 383.5132743362832,
      "grad_norm": 2.5963837302356296e-08,
      "learning_rate": 0.12817690154008973,
      "loss": 0.0,
      "num_input_tokens_seen": 12415440,
      "step": 21860
    },
    {
      "epoch": 383.60176991150445,
      "grad_norm": 1.0052848331554287e-08,
      "learning_rate": 0.12811862510294134,
      "loss": 0.0,
      "num_input_tokens_seen": 12418272,
      "step": 21865
    },
    {
      "epoch": 383.69026548672565,
      "grad_norm": 2.0349252949358743e-08,
      "learning_rate": 0.12806035204017585,
      "loss": 0.0,
      "num_input_tokens_seen": 12421120,
      "step": 21870
    },
    {
      "epoch": 383.7787610619469,
      "grad_norm": 2.2320968184885714e-08,
      "learning_rate": 0.12800208236077987,
      "loss": 0.0,
      "num_input_tokens_seen": 12424272,
      "step": 21875
    },
    {
      "epoch": 383.86725663716817,
      "grad_norm": 1.3297536405332266e-08,
      "learning_rate": 0.12794381607373917,
      "loss": 0.0,
      "num_input_tokens_seen": 12426848,
      "step": 21880
    },
    {
      "epoch": 383.95575221238937,
      "grad_norm": 6.117480300105171e-09,
      "learning_rate": 0.12788555318803924,
      "loss": 0.0,
      "num_input_tokens_seen": 12429424,
      "step": 21885
    },
    {
      "epoch": 384.0353982300885,
      "grad_norm": 1.9726492439531285e-08,
      "learning_rate": 0.1278272937126649,
      "loss": 0.0,
      "num_input_tokens_seen": 12431960,
      "step": 21890
    },
    {
      "epoch": 384.12389380530976,
      "grad_norm": 1.440385766215968e-08,
      "learning_rate": 0.1277690376566005,
      "loss": 0.0,
      "num_input_tokens_seen": 12434840,
      "step": 21895
    },
    {
      "epoch": 384.21238938053096,
      "grad_norm": 1.9955331609367022e-08,
      "learning_rate": 0.12771078502882985,
      "loss": 0.0,
      "num_input_tokens_seen": 12437944,
      "step": 21900
    },
    {
      "epoch": 384.3008849557522,
      "grad_norm": 1.2626526491033019e-08,
      "learning_rate": 0.12765253583833633,
      "loss": 0.0,
      "num_input_tokens_seen": 12440568,
      "step": 21905
    },
    {
      "epoch": 384.3893805309734,
      "grad_norm": 2.2999259385869664e-08,
      "learning_rate": 0.12759429009410256,
      "loss": 0.0,
      "num_input_tokens_seen": 12443912,
      "step": 21910
    },
    {
      "epoch": 384.4778761061947,
      "grad_norm": 1.8385636124662597e-08,
      "learning_rate": 0.12753604780511085,
      "loss": 0.0,
      "num_input_tokens_seen": 12447032,
      "step": 21915
    },
    {
      "epoch": 384.56637168141594,
      "grad_norm": 1.3263515619144073e-08,
      "learning_rate": 0.12747780898034283,
      "loss": 0.0,
      "num_input_tokens_seen": 12449544,
      "step": 21920
    },
    {
      "epoch": 384.65486725663715,
      "grad_norm": 2.5865070085728803e-08,
      "learning_rate": 0.12741957362877973,
      "loss": 0.0,
      "num_input_tokens_seen": 12452504,
      "step": 21925
    },
    {
      "epoch": 384.7433628318584,
      "grad_norm": 1.4113616053634814e-08,
      "learning_rate": 0.12736134175940214,
      "loss": 0.0,
      "num_input_tokens_seen": 12455560,
      "step": 21930
    },
    {
      "epoch": 384.83185840707966,
      "grad_norm": 1.2776934177338717e-08,
      "learning_rate": 0.12730311338119016,
      "loss": 0.0,
      "num_input_tokens_seen": 12458104,
      "step": 21935
    },
    {
      "epoch": 384.92035398230087,
      "grad_norm": 1.2596658827135343e-08,
      "learning_rate": 0.12724488850312327,
      "loss": 0.0,
      "num_input_tokens_seen": 12460680,
      "step": 21940
    },
    {
      "epoch": 385.0,
      "grad_norm": 1.0522375859522981e-08,
      "learning_rate": 0.1271866671341806,
      "loss": 0.0,
      "num_input_tokens_seen": 12462904,
      "step": 21945
    },
    {
      "epoch": 385.08849557522126,
      "grad_norm": 1.3311026059170672e-08,
      "learning_rate": 0.12712844928334047,
      "loss": 0.0,
      "num_input_tokens_seen": 12466184,
      "step": 21950
    },
    {
      "epoch": 385.17699115044246,
      "grad_norm": 1.5029657518539352e-08,
      "learning_rate": 0.12707023495958095,
      "loss": 0.0,
      "num_input_tokens_seen": 12468536,
      "step": 21955
    },
    {
      "epoch": 385.2654867256637,
      "grad_norm": 7.962212222878406e-09,
      "learning_rate": 0.12701202417187932,
      "loss": 0.0,
      "num_input_tokens_seen": 12471016,
      "step": 21960
    },
    {
      "epoch": 385.353982300885,
      "grad_norm": 1.0392604998799015e-08,
      "learning_rate": 0.12695381692921243,
      "loss": 0.0,
      "num_input_tokens_seen": 12474040,
      "step": 21965
    },
    {
      "epoch": 385.4424778761062,
      "grad_norm": 1.0190799315523691e-08,
      "learning_rate": 0.12689561324055665,
      "loss": 0.0,
      "num_input_tokens_seen": 12476904,
      "step": 21970
    },
    {
      "epoch": 385.53097345132744,
      "grad_norm": 1.842513341898666e-08,
      "learning_rate": 0.12683741311488758,
      "loss": 0.0,
      "num_input_tokens_seen": 12479768,
      "step": 21975
    },
    {
      "epoch": 385.6194690265487,
      "grad_norm": 2.778828545046963e-08,
      "learning_rate": 0.1267792165611805,
      "loss": 0.0,
      "num_input_tokens_seen": 12482552,
      "step": 21980
    },
    {
      "epoch": 385.7079646017699,
      "grad_norm": 2.5782131984897205e-08,
      "learning_rate": 0.1267210235884101,
      "loss": 0.0,
      "num_input_tokens_seen": 12485864,
      "step": 21985
    },
    {
      "epoch": 385.79646017699116,
      "grad_norm": 1.5259576713333445e-08,
      "learning_rate": 0.12666283420555033,
      "loss": 0.0,
      "num_input_tokens_seen": 12488568,
      "step": 21990
    },
    {
      "epoch": 385.88495575221236,
      "grad_norm": 1.3593719927484926e-08,
      "learning_rate": 0.12660464842157487,
      "loss": 0.0,
      "num_input_tokens_seen": 12491368,
      "step": 21995
    },
    {
      "epoch": 385.9734513274336,
      "grad_norm": 3.690900740593861e-08,
      "learning_rate": 0.1265464662454566,
      "loss": 0.0,
      "num_input_tokens_seen": 12494280,
      "step": 22000
    },
    {
      "epoch": 385.9734513274336,
      "eval_loss": 0.6455292105674744,
      "eval_runtime": 0.9384,
      "eval_samples_per_second": 26.64,
      "eval_steps_per_second": 13.853,
      "num_input_tokens_seen": 12494280,
      "step": 22000
    },
    {
      "epoch": 386.05309734513276,
      "grad_norm": 2.3234814960915173e-08,
      "learning_rate": 0.12648828768616793,
      "loss": 0.0,
      "num_input_tokens_seen": 12496928,
      "step": 22005
    },
    {
      "epoch": 386.14159292035396,
      "grad_norm": 1.5138942544012934e-08,
      "learning_rate": 0.12643011275268085,
      "loss": 0.0,
      "num_input_tokens_seen": 12499904,
      "step": 22010
    },
    {
      "epoch": 386.2300884955752,
      "grad_norm": 1.3861062519993084e-08,
      "learning_rate": 0.1263719414539665,
      "loss": 0.0,
      "num_input_tokens_seen": 12502752,
      "step": 22015
    },
    {
      "epoch": 386.3185840707965,
      "grad_norm": 1.421829942671593e-08,
      "learning_rate": 0.1263137737989957,
      "loss": 0.0,
      "num_input_tokens_seen": 12505552,
      "step": 22020
    },
    {
      "epoch": 386.4070796460177,
      "grad_norm": 4.025379851668731e-08,
      "learning_rate": 0.1262556097967387,
      "loss": 0.0,
      "num_input_tokens_seen": 12508000,
      "step": 22025
    },
    {
      "epoch": 386.49557522123894,
      "grad_norm": 1.812128225253673e-08,
      "learning_rate": 0.126197449456165,
      "loss": 0.0,
      "num_input_tokens_seen": 12510784,
      "step": 22030
    },
    {
      "epoch": 386.5840707964602,
      "grad_norm": 7.0404904128906765e-09,
      "learning_rate": 0.12613929278624378,
      "loss": 0.0,
      "num_input_tokens_seen": 12513920,
      "step": 22035
    },
    {
      "epoch": 386.6725663716814,
      "grad_norm": 1.991753961760878e-08,
      "learning_rate": 0.12608113979594343,
      "loss": 0.0,
      "num_input_tokens_seen": 12516448,
      "step": 22040
    },
    {
      "epoch": 386.76106194690266,
      "grad_norm": 2.2610192829120024e-08,
      "learning_rate": 0.1260229904942319,
      "loss": 0.0,
      "num_input_tokens_seen": 12519440,
      "step": 22045
    },
    {
      "epoch": 386.8495575221239,
      "grad_norm": 8.463902467781281e-09,
      "learning_rate": 0.12596484489007662,
      "loss": 0.0,
      "num_input_tokens_seen": 12522544,
      "step": 22050
    },
    {
      "epoch": 386.9380530973451,
      "grad_norm": 7.386858236202443e-09,
      "learning_rate": 0.1259067029924442,
      "loss": 0.0,
      "num_input_tokens_seen": 12525408,
      "step": 22055
    },
    {
      "epoch": 387.01769911504425,
      "grad_norm": 2.43399735921912e-08,
      "learning_rate": 0.12584856481030096,
      "loss": 0.0,
      "num_input_tokens_seen": 12527936,
      "step": 22060
    },
    {
      "epoch": 387.1061946902655,
      "grad_norm": 1.6605758546006655e-08,
      "learning_rate": 0.12579043035261261,
      "loss": 0.0,
      "num_input_tokens_seen": 12530976,
      "step": 22065
    },
    {
      "epoch": 387.1946902654867,
      "grad_norm": 2.0735178907216323e-08,
      "learning_rate": 0.1257322996283441,
      "loss": 0.0,
      "num_input_tokens_seen": 12533952,
      "step": 22070
    },
    {
      "epoch": 387.283185840708,
      "grad_norm": 1.9213350910263216e-08,
      "learning_rate": 0.12567417264645994,
      "loss": 0.0,
      "num_input_tokens_seen": 12536656,
      "step": 22075
    },
    {
      "epoch": 387.37168141592923,
      "grad_norm": 2.1179825893113957e-08,
      "learning_rate": 0.12561604941592408,
      "loss": 0.0,
      "num_input_tokens_seen": 12539104,
      "step": 22080
    },
    {
      "epoch": 387.46017699115043,
      "grad_norm": 1.5930703867184093e-08,
      "learning_rate": 0.12555792994569978,
      "loss": 0.0,
      "num_input_tokens_seen": 12541920,
      "step": 22085
    },
    {
      "epoch": 387.5486725663717,
      "grad_norm": 9.965130054467863e-09,
      "learning_rate": 0.1254998142447499,
      "loss": 0.0,
      "num_input_tokens_seen": 12545024,
      "step": 22090
    },
    {
      "epoch": 387.6371681415929,
      "grad_norm": 1.900011525890477e-08,
      "learning_rate": 0.1254417023220365,
      "loss": 0.0,
      "num_input_tokens_seen": 12547920,
      "step": 22095
    },
    {
      "epoch": 387.72566371681415,
      "grad_norm": 9.225914254784584e-09,
      "learning_rate": 0.12538359418652126,
      "loss": 0.0,
      "num_input_tokens_seen": 12550848,
      "step": 22100
    },
    {
      "epoch": 387.8141592920354,
      "grad_norm": 9.254468302799523e-09,
      "learning_rate": 0.12532548984716513,
      "loss": 0.0,
      "num_input_tokens_seen": 12553952,
      "step": 22105
    },
    {
      "epoch": 387.9026548672566,
      "grad_norm": 2.4100847539898496e-08,
      "learning_rate": 0.12526738931292855,
      "loss": 0.0,
      "num_input_tokens_seen": 12556880,
      "step": 22110
    },
    {
      "epoch": 387.9911504424779,
      "grad_norm": 1.742125732562272e-08,
      "learning_rate": 0.1252092925927714,
      "loss": 0.0,
      "num_input_tokens_seen": 12559952,
      "step": 22115
    },
    {
      "epoch": 388.070796460177,
      "grad_norm": 1.3089803019283863e-08,
      "learning_rate": 0.12515119969565278,
      "loss": 0.0,
      "num_input_tokens_seen": 12562832,
      "step": 22120
    },
    {
      "epoch": 388.1592920353982,
      "grad_norm": 1.2578012409392159e-08,
      "learning_rate": 0.12509311063053144,
      "loss": 0.0,
      "num_input_tokens_seen": 12566016,
      "step": 22125
    },
    {
      "epoch": 388.24778761061947,
      "grad_norm": 2.5503510414637276e-08,
      "learning_rate": 0.1250350254063655,
      "loss": 0.0,
      "num_input_tokens_seen": 12568544,
      "step": 22130
    },
    {
      "epoch": 388.3362831858407,
      "grad_norm": 6.92286583614532e-09,
      "learning_rate": 0.1249769440321123,
      "loss": 0.0,
      "num_input_tokens_seen": 12571184,
      "step": 22135
    },
    {
      "epoch": 388.42477876106193,
      "grad_norm": 1.763817536470924e-08,
      "learning_rate": 0.12491886651672884,
      "loss": 0.0,
      "num_input_tokens_seen": 12573904,
      "step": 22140
    },
    {
      "epoch": 388.5132743362832,
      "grad_norm": 1.7815311892377395e-08,
      "learning_rate": 0.12486079286917139,
      "loss": 0.0,
      "num_input_tokens_seen": 12576800,
      "step": 22145
    },
    {
      "epoch": 388.60176991150445,
      "grad_norm": 1.657811310451507e-08,
      "learning_rate": 0.12480272309839553,
      "loss": 0.0,
      "num_input_tokens_seen": 12579664,
      "step": 22150
    },
    {
      "epoch": 388.69026548672565,
      "grad_norm": 1.2017911998896125e-08,
      "learning_rate": 0.12474465721335648,
      "loss": 0.0,
      "num_input_tokens_seen": 12582288,
      "step": 22155
    },
    {
      "epoch": 388.7787610619469,
      "grad_norm": 2.3396788506602206e-08,
      "learning_rate": 0.12468659522300861,
      "loss": 0.0,
      "num_input_tokens_seen": 12585184,
      "step": 22160
    },
    {
      "epoch": 388.86725663716817,
      "grad_norm": 1.4254811553371383e-08,
      "learning_rate": 0.12462853713630584,
      "loss": 0.0,
      "num_input_tokens_seen": 12587952,
      "step": 22165
    },
    {
      "epoch": 388.95575221238937,
      "grad_norm": 7.74776331979865e-09,
      "learning_rate": 0.12457048296220156,
      "loss": 0.0,
      "num_input_tokens_seen": 12590960,
      "step": 22170
    },
    {
      "epoch": 389.0353982300885,
      "grad_norm": 1.2084876210849416e-08,
      "learning_rate": 0.12451243270964832,
      "loss": 0.0,
      "num_input_tokens_seen": 12593288,
      "step": 22175
    },
    {
      "epoch": 389.12389380530976,
      "grad_norm": 1.1579198933020507e-08,
      "learning_rate": 0.12445438638759827,
      "loss": 0.0,
      "num_input_tokens_seen": 12596200,
      "step": 22180
    },
    {
      "epoch": 389.21238938053096,
      "grad_norm": 1.934990478957843e-08,
      "learning_rate": 0.1243963440050029,
      "loss": 0.0,
      "num_input_tokens_seen": 12599256,
      "step": 22185
    },
    {
      "epoch": 389.3008849557522,
      "grad_norm": 1.4611671872444276e-08,
      "learning_rate": 0.12433830557081298,
      "loss": 0.0,
      "num_input_tokens_seen": 12601800,
      "step": 22190
    },
    {
      "epoch": 389.3893805309734,
      "grad_norm": 2.3437731755393543e-08,
      "learning_rate": 0.12428027109397889,
      "loss": 0.0,
      "num_input_tokens_seen": 12604984,
      "step": 22195
    },
    {
      "epoch": 389.4778761061947,
      "grad_norm": 8.249762650791581e-09,
      "learning_rate": 0.12422224058345015,
      "loss": 0.0,
      "num_input_tokens_seen": 12608008,
      "step": 22200
    },
    {
      "epoch": 389.4778761061947,
      "eval_loss": 0.6497951149940491,
      "eval_runtime": 0.9367,
      "eval_samples_per_second": 26.689,
      "eval_steps_per_second": 13.878,
      "num_input_tokens_seen": 12608008,
      "step": 22200
    },
    {
      "epoch": 389.56637168141594,
      "grad_norm": 2.2578566571951342e-08,
      "learning_rate": 0.12416421404817583,
      "loss": 0.0,
      "num_input_tokens_seen": 12610984,
      "step": 22205
    },
    {
      "epoch": 389.65486725663715,
      "grad_norm": 1.8115187572220748e-08,
      "learning_rate": 0.12410619149710447,
      "loss": 0.0,
      "num_input_tokens_seen": 12613400,
      "step": 22210
    },
    {
      "epoch": 389.7433628318584,
      "grad_norm": 1.897621437763064e-08,
      "learning_rate": 0.12404817293918374,
      "loss": 0.0,
      "num_input_tokens_seen": 12615992,
      "step": 22215
    },
    {
      "epoch": 389.83185840707966,
      "grad_norm": 2.4747663474045112e-08,
      "learning_rate": 0.12399015838336086,
      "loss": 0.0,
      "num_input_tokens_seen": 12619112,
      "step": 22220
    },
    {
      "epoch": 389.92035398230087,
      "grad_norm": 1.6079084730336035e-08,
      "learning_rate": 0.12393214783858246,
      "loss": 0.0,
      "num_input_tokens_seen": 12622136,
      "step": 22225
    },
    {
      "epoch": 390.0,
      "grad_norm": 2.5466855291256252e-08,
      "learning_rate": 0.1238741413137944,
      "loss": 0.0,
      "num_input_tokens_seen": 12624536,
      "step": 22230
    },
    {
      "epoch": 390.08849557522126,
      "grad_norm": 1.863835663584723e-08,
      "learning_rate": 0.12381613881794212,
      "loss": 0.0,
      "num_input_tokens_seen": 12627944,
      "step": 22235
    },
    {
      "epoch": 390.17699115044246,
      "grad_norm": 1.3968273648856666e-08,
      "learning_rate": 0.12375814035997022,
      "loss": 0.0,
      "num_input_tokens_seen": 12631048,
      "step": 22240
    },
    {
      "epoch": 390.2654867256637,
      "grad_norm": 1.1178486580831759e-08,
      "learning_rate": 0.12370014594882285,
      "loss": 0.0,
      "num_input_tokens_seen": 12633560,
      "step": 22245
    },
    {
      "epoch": 390.353982300885,
      "grad_norm": 1.7459278467413242e-08,
      "learning_rate": 0.12364215559344356,
      "loss": 0.0,
      "num_input_tokens_seen": 12636600,
      "step": 22250
    },
    {
      "epoch": 390.4424778761062,
      "grad_norm": 1.5260058106036922e-08,
      "learning_rate": 0.12358416930277506,
      "loss": 0.0,
      "num_input_tokens_seen": 12639320,
      "step": 22255
    },
    {
      "epoch": 390.53097345132744,
      "grad_norm": 1.3597761139294562e-08,
      "learning_rate": 0.1235261870857596,
      "loss": 0.0,
      "num_input_tokens_seen": 12642328,
      "step": 22260
    },
    {
      "epoch": 390.6194690265487,
      "grad_norm": 2.0572903380866592e-08,
      "learning_rate": 0.12346820895133884,
      "loss": 0.0,
      "num_input_tokens_seen": 12644936,
      "step": 22265
    },
    {
      "epoch": 390.7079646017699,
      "grad_norm": 7.003775781555532e-09,
      "learning_rate": 0.12341023490845361,
      "loss": 0.0,
      "num_input_tokens_seen": 12647432,
      "step": 22270
    },
    {
      "epoch": 390.79646017699116,
      "grad_norm": 2.5497255862205748e-08,
      "learning_rate": 0.12335226496604437,
      "loss": 0.0,
      "num_input_tokens_seen": 12650072,
      "step": 22275
    },
    {
      "epoch": 390.88495575221236,
      "grad_norm": 2.455562153613755e-08,
      "learning_rate": 0.12329429913305069,
      "loss": 0.0,
      "num_input_tokens_seen": 12652728,
      "step": 22280
    },
    {
      "epoch": 390.9734513274336,
      "grad_norm": 9.907348719195852e-09,
      "learning_rate": 0.12323633741841171,
      "loss": 0.0,
      "num_input_tokens_seen": 12656328,
      "step": 22285
    },
    {
      "epoch": 391.05309734513276,
      "grad_norm": 5.421924686999091e-09,
      "learning_rate": 0.12317837983106583,
      "loss": 0.0,
      "num_input_tokens_seen": 12658760,
      "step": 22290
    },
    {
      "epoch": 391.14159292035396,
      "grad_norm": 2.3390636982867363e-08,
      "learning_rate": 0.12312042637995087,
      "loss": 0.0,
      "num_input_tokens_seen": 12661448,
      "step": 22295
    },
    {
      "epoch": 391.2300884955752,
      "grad_norm": 2.296941126189722e-08,
      "learning_rate": 0.12306247707400389,
      "loss": 0.0,
      "num_input_tokens_seen": 12664920,
      "step": 22300
    },
    {
      "epoch": 391.3185840707965,
      "grad_norm": 1.3966984901969681e-08,
      "learning_rate": 0.12300453192216154,
      "loss": 0.0,
      "num_input_tokens_seen": 12667864,
      "step": 22305
    },
    {
      "epoch": 391.4070796460177,
      "grad_norm": 2.3648436098255843e-08,
      "learning_rate": 0.12294659093335956,
      "loss": 0.0,
      "num_input_tokens_seen": 12671160,
      "step": 22310
    },
    {
      "epoch": 391.49557522123894,
      "grad_norm": 7.777452459833967e-09,
      "learning_rate": 0.12288865411653327,
      "loss": 0.0,
      "num_input_tokens_seen": 12673672,
      "step": 22315
    },
    {
      "epoch": 391.5840707964602,
      "grad_norm": 2.966413070737417e-08,
      "learning_rate": 0.12283072148061717,
      "loss": 0.0,
      "num_input_tokens_seen": 12676552,
      "step": 22320
    },
    {
      "epoch": 391.6725663716814,
      "grad_norm": 1.828832729700025e-08,
      "learning_rate": 0.12277279303454529,
      "loss": 0.0,
      "num_input_tokens_seen": 12679160,
      "step": 22325
    },
    {
      "epoch": 391.76106194690266,
      "grad_norm": 3.6674300929462333e-08,
      "learning_rate": 0.12271486878725091,
      "loss": 0.0,
      "num_input_tokens_seen": 12681640,
      "step": 22330
    },
    {
      "epoch": 391.8495575221239,
      "grad_norm": 1.7966572229966005e-08,
      "learning_rate": 0.12265694874766658,
      "loss": 0.0,
      "num_input_tokens_seen": 12684328,
      "step": 22335
    },
    {
      "epoch": 391.9380530973451,
      "grad_norm": 2.835227874697921e-08,
      "learning_rate": 0.12259903292472435,
      "loss": 0.0,
      "num_input_tokens_seen": 12687224,
      "step": 22340
    },
    {
      "epoch": 392.01769911504425,
      "grad_norm": 1.3514561025829153e-08,
      "learning_rate": 0.12254112132735567,
      "loss": 0.0,
      "num_input_tokens_seen": 12689680,
      "step": 22345
    },
    {
      "epoch": 392.1061946902655,
      "grad_norm": 1.0831961105850496e-08,
      "learning_rate": 0.12248321396449108,
      "loss": 0.0,
      "num_input_tokens_seen": 12692400,
      "step": 22350
    },
    {
      "epoch": 392.1946902654867,
      "grad_norm": 1.1851597037093597e-08,
      "learning_rate": 0.12242531084506075,
      "loss": 0.0,
      "num_input_tokens_seen": 12695712,
      "step": 22355
    },
    {
      "epoch": 392.283185840708,
      "grad_norm": 1.562199969384892e-08,
      "learning_rate": 0.122367411977994,
      "loss": 0.0,
      "num_input_tokens_seen": 12698448,
      "step": 22360
    },
    {
      "epoch": 392.37168141592923,
      "grad_norm": 1.570048446808414e-08,
      "learning_rate": 0.12230951737221954,
      "loss": 0.0,
      "num_input_tokens_seen": 12701216,
      "step": 22365
    },
    {
      "epoch": 392.46017699115043,
      "grad_norm": 1.9046989763182864e-08,
      "learning_rate": 0.12225162703666555,
      "loss": 0.0,
      "num_input_tokens_seen": 12703984,
      "step": 22370
    },
    {
      "epoch": 392.5486725663717,
      "grad_norm": 9.888730723162098e-09,
      "learning_rate": 0.1221937409802593,
      "loss": 0.0,
      "num_input_tokens_seen": 12707440,
      "step": 22375
    },
    {
      "epoch": 392.6371681415929,
      "grad_norm": 1.0944146922042819e-08,
      "learning_rate": 0.12213585921192768,
      "loss": 0.0,
      "num_input_tokens_seen": 12710336,
      "step": 22380
    },
    {
      "epoch": 392.72566371681415,
      "grad_norm": 6.787971074118104e-09,
      "learning_rate": 0.1220779817405967,
      "loss": 0.0,
      "num_input_tokens_seen": 12713024,
      "step": 22385
    },
    {
      "epoch": 392.8141592920354,
      "grad_norm": 1.3164699552703496e-08,
      "learning_rate": 0.12202010857519181,
      "loss": 0.0,
      "num_input_tokens_seen": 12715952,
      "step": 22390
    },
    {
      "epoch": 392.9026548672566,
      "grad_norm": 5.5493267758777165e-09,
      "learning_rate": 0.12196223972463785,
      "loss": 0.0,
      "num_input_tokens_seen": 12718672,
      "step": 22395
    },
    {
      "epoch": 392.9911504424779,
      "grad_norm": 1.971590002369794e-08,
      "learning_rate": 0.12190437519785885,
      "loss": 0.0,
      "num_input_tokens_seen": 12721456,
      "step": 22400
    },
    {
      "epoch": 392.9911504424779,
      "eval_loss": 0.6347877979278564,
      "eval_runtime": 0.9435,
      "eval_samples_per_second": 26.497,
      "eval_steps_per_second": 13.778,
      "num_input_tokens_seen": 12721456,
      "step": 22400
    },
    {
      "epoch": 393.070796460177,
      "grad_norm": 6.805064067805233e-09,
      "learning_rate": 0.12184651500377823,
      "loss": 0.0,
      "num_input_tokens_seen": 12724032,
      "step": 22405
    },
    {
      "epoch": 393.1592920353982,
      "grad_norm": 1.1738399585681236e-08,
      "learning_rate": 0.12178865915131885,
      "loss": 0.0,
      "num_input_tokens_seen": 12727008,
      "step": 22410
    },
    {
      "epoch": 393.24778761061947,
      "grad_norm": 2.0762097818760594e-08,
      "learning_rate": 0.1217308076494027,
      "loss": 0.0,
      "num_input_tokens_seen": 12730048,
      "step": 22415
    },
    {
      "epoch": 393.3362831858407,
      "grad_norm": 3.953227079023236e-08,
      "learning_rate": 0.12167296050695134,
      "loss": 0.0,
      "num_input_tokens_seen": 12732960,
      "step": 22420
    },
    {
      "epoch": 393.42477876106193,
      "grad_norm": 5.054475504806533e-09,
      "learning_rate": 0.12161511773288536,
      "loss": 0.0,
      "num_input_tokens_seen": 12735632,
      "step": 22425
    },
    {
      "epoch": 393.5132743362832,
      "grad_norm": 1.0298231600813779e-08,
      "learning_rate": 0.121557279336125,
      "loss": 0.0,
      "num_input_tokens_seen": 12738240,
      "step": 22430
    },
    {
      "epoch": 393.60176991150445,
      "grad_norm": 2.23158593826156e-08,
      "learning_rate": 0.12149944532558957,
      "loss": 0.0,
      "num_input_tokens_seen": 12740544,
      "step": 22435
    },
    {
      "epoch": 393.69026548672565,
      "grad_norm": 1.9754546443095933e-08,
      "learning_rate": 0.12144161571019785,
      "loss": 0.0,
      "num_input_tokens_seen": 12743872,
      "step": 22440
    },
    {
      "epoch": 393.7787610619469,
      "grad_norm": 1.5650918783194356e-08,
      "learning_rate": 0.12138379049886781,
      "loss": 0.0,
      "num_input_tokens_seen": 12746496,
      "step": 22445
    },
    {
      "epoch": 393.86725663716817,
      "grad_norm": 2.7703826788183505e-08,
      "learning_rate": 0.12132596970051697,
      "loss": 0.0,
      "num_input_tokens_seen": 12749568,
      "step": 22450
    },
    {
      "epoch": 393.95575221238937,
      "grad_norm": 2.0582669790769614e-08,
      "learning_rate": 0.12126815332406189,
      "loss": 0.0,
      "num_input_tokens_seen": 12753008,
      "step": 22455
    },
    {
      "epoch": 394.0353982300885,
      "grad_norm": 1.161786933323583e-08,
      "learning_rate": 0.12121034137841868,
      "loss": 0.0,
      "num_input_tokens_seen": 12755224,
      "step": 22460
    },
    {
      "epoch": 394.12389380530976,
      "grad_norm": 2.8027114851170154e-08,
      "learning_rate": 0.12115253387250258,
      "loss": 0.0,
      "num_input_tokens_seen": 12757896,
      "step": 22465
    },
    {
      "epoch": 394.21238938053096,
      "grad_norm": 2.590366676713529e-08,
      "learning_rate": 0.12109473081522831,
      "loss": 0.0,
      "num_input_tokens_seen": 12760840,
      "step": 22470
    },
    {
      "epoch": 394.3008849557522,
      "grad_norm": 1.5732085856257072e-08,
      "learning_rate": 0.12103693221550982,
      "loss": 0.0,
      "num_input_tokens_seen": 12763656,
      "step": 22475
    },
    {
      "epoch": 394.3893805309734,
      "grad_norm": 2.304725477131342e-08,
      "learning_rate": 0.12097913808226027,
      "loss": 0.0,
      "num_input_tokens_seen": 12766712,
      "step": 22480
    },
    {
      "epoch": 394.4778761061947,
      "grad_norm": 3.9600998036348756e-08,
      "learning_rate": 0.12092134842439234,
      "loss": 0.0,
      "num_input_tokens_seen": 12769368,
      "step": 22485
    },
    {
      "epoch": 394.56637168141594,
      "grad_norm": 1.4522219871082598e-08,
      "learning_rate": 0.12086356325081798,
      "loss": 0.0,
      "num_input_tokens_seen": 12772568,
      "step": 22490
    },
    {
      "epoch": 394.65486725663715,
      "grad_norm": 1.1444687864070602e-08,
      "learning_rate": 0.12080578257044824,
      "loss": 0.0,
      "num_input_tokens_seen": 12775304,
      "step": 22495
    },
    {
      "epoch": 394.7433628318584,
      "grad_norm": 1.0336536071520186e-08,
      "learning_rate": 0.12074800639219378,
      "loss": 0.0,
      "num_input_tokens_seen": 12778184,
      "step": 22500
    },
    {
      "epoch": 394.83185840707966,
      "grad_norm": 2.841681379095462e-08,
      "learning_rate": 0.12069023472496428,
      "loss": 0.0,
      "num_input_tokens_seen": 12781112,
      "step": 22505
    },
    {
      "epoch": 394.92035398230087,
      "grad_norm": 1.45842928844786e-08,
      "learning_rate": 0.12063246757766893,
      "loss": 0.0,
      "num_input_tokens_seen": 12783848,
      "step": 22510
    },
    {
      "epoch": 395.0,
      "grad_norm": 3.797373082647937e-08,
      "learning_rate": 0.12057470495921618,
      "loss": 0.0,
      "num_input_tokens_seen": 12786624,
      "step": 22515
    },
    {
      "epoch": 395.08849557522126,
      "grad_norm": 1.4754335531108609e-08,
      "learning_rate": 0.12051694687851364,
      "loss": 0.0,
      "num_input_tokens_seen": 12789248,
      "step": 22520
    },
    {
      "epoch": 395.17699115044246,
      "grad_norm": 2.7719547546212198e-08,
      "learning_rate": 0.12045919334446839,
      "loss": 0.0,
      "num_input_tokens_seen": 12792016,
      "step": 22525
    },
    {
      "epoch": 395.2654867256637,
      "grad_norm": 2.43986786330197e-08,
      "learning_rate": 0.12040144436598683,
      "loss": 0.0,
      "num_input_tokens_seen": 12795568,
      "step": 22530
    },
    {
      "epoch": 395.353982300885,
      "grad_norm": 7.483545338970998e-09,
      "learning_rate": 0.12034369995197444,
      "loss": 0.0,
      "num_input_tokens_seen": 12798624,
      "step": 22535
    },
    {
      "epoch": 395.4424778761062,
      "grad_norm": 2.376306440510234e-08,
      "learning_rate": 0.12028596011133627,
      "loss": 0.0,
      "num_input_tokens_seen": 12801184,
      "step": 22540
    },
    {
      "epoch": 395.53097345132744,
      "grad_norm": 9.38770039482506e-09,
      "learning_rate": 0.12022822485297643,
      "loss": 0.0,
      "num_input_tokens_seen": 12803872,
      "step": 22545
    },
    {
      "epoch": 395.6194690265487,
      "grad_norm": 9.240691767331555e-09,
      "learning_rate": 0.12017049418579843,
      "loss": 0.0,
      "num_input_tokens_seen": 12806544,
      "step": 22550
    },
    {
      "epoch": 395.7079646017699,
      "grad_norm": 2.2414560874040035e-08,
      "learning_rate": 0.12011276811870514,
      "loss": 0.0,
      "num_input_tokens_seen": 12809376,
      "step": 22555
    },
    {
      "epoch": 395.79646017699116,
      "grad_norm": 1.5335395175952726e-08,
      "learning_rate": 0.12005504666059852,
      "loss": 0.0,
      "num_input_tokens_seen": 12812368,
      "step": 22560
    },
    {
      "epoch": 395.88495575221236,
      "grad_norm": 5.622640131264234e-09,
      "learning_rate": 0.11999732982038003,
      "loss": 0.0,
      "num_input_tokens_seen": 12815424,
      "step": 22565
    },
    {
      "epoch": 395.9734513274336,
      "grad_norm": 9.676749179732269e-09,
      "learning_rate": 0.11993961760695038,
      "loss": 0.0,
      "num_input_tokens_seen": 12818208,
      "step": 22570
    },
    {
      "epoch": 396.05309734513276,
      "grad_norm": 1.4318051633210871e-08,
      "learning_rate": 0.11988191002920942,
      "loss": 0.0,
      "num_input_tokens_seen": 12820568,
      "step": 22575
    },
    {
      "epoch": 396.14159292035396,
      "grad_norm": 1.4125025593614282e-08,
      "learning_rate": 0.11982420709605641,
      "loss": 0.0,
      "num_input_tokens_seen": 12823480,
      "step": 22580
    },
    {
      "epoch": 396.2300884955752,
      "grad_norm": 1.4863754671523566e-08,
      "learning_rate": 0.11976650881638991,
      "loss": 0.0,
      "num_input_tokens_seen": 12826072,
      "step": 22585
    },
    {
      "epoch": 396.3185840707965,
      "grad_norm": 2.248228447854217e-08,
      "learning_rate": 0.11970881519910764,
      "loss": 0.0,
      "num_input_tokens_seen": 12829112,
      "step": 22590
    },
    {
      "epoch": 396.4070796460177,
      "grad_norm": 1.6384424483817384e-08,
      "learning_rate": 0.1196511262531068,
      "loss": 0.0,
      "num_input_tokens_seen": 12831832,
      "step": 22595
    },
    {
      "epoch": 396.49557522123894,
      "grad_norm": 1.7098988891461886e-08,
      "learning_rate": 0.11959344198728361,
      "loss": 0.0,
      "num_input_tokens_seen": 12835240,
      "step": 22600
    },
    {
      "epoch": 396.49557522123894,
      "eval_loss": 0.6525802612304688,
      "eval_runtime": 0.9417,
      "eval_samples_per_second": 26.546,
      "eval_steps_per_second": 13.804,
      "num_input_tokens_seen": 12835240,
      "step": 22600
    },
    {
      "epoch": 396.5840707964602,
      "grad_norm": 1.3225540662631374e-08,
      "learning_rate": 0.11953576241053378,
      "loss": 0.0,
      "num_input_tokens_seen": 12838904,
      "step": 22605
    },
    {
      "epoch": 396.6725663716814,
      "grad_norm": 7.3576611470116404e-09,
      "learning_rate": 0.11947808753175228,
      "loss": 0.0,
      "num_input_tokens_seen": 12841800,
      "step": 22610
    },
    {
      "epoch": 396.76106194690266,
      "grad_norm": 1.7066506430296613e-08,
      "learning_rate": 0.1194204173598332,
      "loss": 0.0,
      "num_input_tokens_seen": 12844472,
      "step": 22615
    },
    {
      "epoch": 396.8495575221239,
      "grad_norm": 1.5326394375847485e-08,
      "learning_rate": 0.11936275190367007,
      "loss": 0.0,
      "num_input_tokens_seen": 12847128,
      "step": 22620
    },
    {
      "epoch": 396.9380530973451,
      "grad_norm": 2.2233178853525715e-08,
      "learning_rate": 0.11930509117215563,
      "loss": 0.0,
      "num_input_tokens_seen": 12849640,
      "step": 22625
    },
    {
      "epoch": 397.01769911504425,
      "grad_norm": 1.512944436399266e-08,
      "learning_rate": 0.11924743517418179,
      "loss": 0.0,
      "num_input_tokens_seen": 12851704,
      "step": 22630
    },
    {
      "epoch": 397.1061946902655,
      "grad_norm": 1.7388154915920495e-08,
      "learning_rate": 0.11918978391864,
      "loss": 0.0,
      "num_input_tokens_seen": 12854392,
      "step": 22635
    },
    {
      "epoch": 397.1946902654867,
      "grad_norm": 1.7031128507483118e-08,
      "learning_rate": 0.11913213741442065,
      "loss": 0.0,
      "num_input_tokens_seen": 12857016,
      "step": 22640
    },
    {
      "epoch": 397.283185840708,
      "grad_norm": 2.5244951018521533e-08,
      "learning_rate": 0.11907449567041364,
      "loss": 0.0,
      "num_input_tokens_seen": 12859896,
      "step": 22645
    },
    {
      "epoch": 397.37168141592923,
      "grad_norm": 1.9413663565615025e-08,
      "learning_rate": 0.11901685869550803,
      "loss": 0.0,
      "num_input_tokens_seen": 12862744,
      "step": 22650
    },
    {
      "epoch": 397.46017699115043,
      "grad_norm": 8.936064332942806e-09,
      "learning_rate": 0.1189592264985922,
      "loss": 0.0,
      "num_input_tokens_seen": 12865464,
      "step": 22655
    },
    {
      "epoch": 397.5486725663717,
      "grad_norm": 1.9158358455229063e-08,
      "learning_rate": 0.11890159908855373,
      "loss": 0.0,
      "num_input_tokens_seen": 12868120,
      "step": 22660
    },
    {
      "epoch": 397.6371681415929,
      "grad_norm": 1.7200580515464026e-08,
      "learning_rate": 0.11884397647427941,
      "loss": 0.0,
      "num_input_tokens_seen": 12871112,
      "step": 22665
    },
    {
      "epoch": 397.72566371681415,
      "grad_norm": 2.1743174372090834e-08,
      "learning_rate": 0.11878635866465546,
      "loss": 0.0,
      "num_input_tokens_seen": 12874008,
      "step": 22670
    },
    {
      "epoch": 397.8141592920354,
      "grad_norm": 8.967786513380815e-09,
      "learning_rate": 0.11872874566856734,
      "loss": 0.0,
      "num_input_tokens_seen": 12877112,
      "step": 22675
    },
    {
      "epoch": 397.9026548672566,
      "grad_norm": 3.6348314580436636e-08,
      "learning_rate": 0.11867113749489955,
      "loss": 0.0,
      "num_input_tokens_seen": 12879944,
      "step": 22680
    },
    {
      "epoch": 397.9911504424779,
      "grad_norm": 1.7296160947921635e-08,
      "learning_rate": 0.11861353415253607,
      "loss": 0.0,
      "num_input_tokens_seen": 12883672,
      "step": 22685
    },
    {
      "epoch": 398.070796460177,
      "grad_norm": 2.3365389623108967e-08,
      "learning_rate": 0.11855593565036011,
      "loss": 0.0,
      "num_input_tokens_seen": 12886360,
      "step": 22690
    },
    {
      "epoch": 398.1592920353982,
      "grad_norm": 2.6991326507186386e-08,
      "learning_rate": 0.11849834199725394,
      "loss": 0.0,
      "num_input_tokens_seen": 12888904,
      "step": 22695
    },
    {
      "epoch": 398.24778761061947,
      "grad_norm": 1.8704033877270376e-08,
      "learning_rate": 0.1184407532020994,
      "loss": 0.0,
      "num_input_tokens_seen": 12891864,
      "step": 22700
    },
    {
      "epoch": 398.3362831858407,
      "grad_norm": 3.5521580343811365e-08,
      "learning_rate": 0.11838316927377723,
      "loss": 0.0,
      "num_input_tokens_seen": 12894888,
      "step": 22705
    },
    {
      "epoch": 398.42477876106193,
      "grad_norm": 1.0431282504441697e-08,
      "learning_rate": 0.11832559022116766,
      "loss": 0.0,
      "num_input_tokens_seen": 12897768,
      "step": 22710
    },
    {
      "epoch": 398.5132743362832,
      "grad_norm": 1.550418105011886e-08,
      "learning_rate": 0.11826801605315022,
      "loss": 0.0,
      "num_input_tokens_seen": 12900392,
      "step": 22715
    },
    {
      "epoch": 398.60176991150445,
      "grad_norm": 1.648567149459268e-08,
      "learning_rate": 0.1182104467786034,
      "loss": 0.0,
      "num_input_tokens_seen": 12903304,
      "step": 22720
    },
    {
      "epoch": 398.69026548672565,
      "grad_norm": 1.7989917111549403e-08,
      "learning_rate": 0.1181528824064052,
      "loss": 0.0,
      "num_input_tokens_seen": 12906360,
      "step": 22725
    },
    {
      "epoch": 398.7787610619469,
      "grad_norm": 5.307334305371114e-08,
      "learning_rate": 0.11809532294543279,
      "loss": 0.0,
      "num_input_tokens_seen": 12909448,
      "step": 22730
    },
    {
      "epoch": 398.86725663716817,
      "grad_norm": 6.676551311812773e-09,
      "learning_rate": 0.11803776840456245,
      "loss": 0.0,
      "num_input_tokens_seen": 12912376,
      "step": 22735
    },
    {
      "epoch": 398.95575221238937,
      "grad_norm": 1.3535728093927446e-08,
      "learning_rate": 0.11798021879266997,
      "loss": 0.0,
      "num_input_tokens_seen": 12914872,
      "step": 22740
    },
    {
      "epoch": 399.0353982300885,
      "grad_norm": 1.3234355833446898e-08,
      "learning_rate": 0.11792267411863006,
      "loss": 0.0,
      "num_input_tokens_seen": 12917160,
      "step": 22745
    },
    {
      "epoch": 399.12389380530976,
      "grad_norm": 1.7656676121191595e-08,
      "learning_rate": 0.1178651343913169,
      "loss": 0.0,
      "num_input_tokens_seen": 12919800,
      "step": 22750
    },
    {
      "epoch": 399.21238938053096,
      "grad_norm": 1.9406472873129132e-08,
      "learning_rate": 0.11780759961960392,
      "loss": 0.0,
      "num_input_tokens_seen": 12922920,
      "step": 22755
    },
    {
      "epoch": 399.3008849557522,
      "grad_norm": 2.1896052970760138e-08,
      "learning_rate": 0.1177500698123636,
      "loss": 0.0,
      "num_input_tokens_seen": 12925656,
      "step": 22760
    },
    {
      "epoch": 399.3893805309734,
      "grad_norm": 1.781878644635526e-08,
      "learning_rate": 0.11769254497846778,
      "loss": 0.0,
      "num_input_tokens_seen": 12928568,
      "step": 22765
    },
    {
      "epoch": 399.4778761061947,
      "grad_norm": 2.1404279237913215e-08,
      "learning_rate": 0.11763502512678758,
      "loss": 0.0,
      "num_input_tokens_seen": 12931720,
      "step": 22770
    },
    {
      "epoch": 399.56637168141594,
      "grad_norm": 2.0152281621221846e-08,
      "learning_rate": 0.11757751026619315,
      "loss": 0.0,
      "num_input_tokens_seen": 12934264,
      "step": 22775
    },
    {
      "epoch": 399.65486725663715,
      "grad_norm": 1.4029194694842317e-08,
      "learning_rate": 0.11752000040555416,
      "loss": 0.0,
      "num_input_tokens_seen": 12937848,
      "step": 22780
    },
    {
      "epoch": 399.7433628318584,
      "grad_norm": 2.1692054374966574e-08,
      "learning_rate": 0.11746249555373921,
      "loss": 0.0,
      "num_input_tokens_seen": 12940632,
      "step": 22785
    },
    {
      "epoch": 399.83185840707966,
      "grad_norm": 1.3162543055500464e-08,
      "learning_rate": 0.11740499571961638,
      "loss": 0.0,
      "num_input_tokens_seen": 12943512,
      "step": 22790
    },
    {
      "epoch": 399.92035398230087,
      "grad_norm": 7.184137285065617e-09,
      "learning_rate": 0.11734750091205279,
      "loss": 0.0,
      "num_input_tokens_seen": 12946456,
      "step": 22795
    },
    {
      "epoch": 400.0,
      "grad_norm": 7.554487702066126e-09,
      "learning_rate": 0.11729001113991493,
      "loss": 0.0,
      "num_input_tokens_seen": 12948416,
      "step": 22800
    },
    {
      "epoch": 400.0,
      "eval_loss": 0.6555616855621338,
      "eval_runtime": 0.9104,
      "eval_samples_per_second": 27.461,
      "eval_steps_per_second": 14.28,
      "num_input_tokens_seen": 12948416,
      "step": 22800
    },
    {
      "epoch": 400.08849557522126,
      "grad_norm": 3.993587327499881e-08,
      "learning_rate": 0.11723252641206837,
      "loss": 0.0,
      "num_input_tokens_seen": 12951568,
      "step": 22805
    },
    {
      "epoch": 400.17699115044246,
      "grad_norm": 5.0706567833458394e-08,
      "learning_rate": 0.11717504673737808,
      "loss": 0.0,
      "num_input_tokens_seen": 12954160,
      "step": 22810
    },
    {
      "epoch": 400.2654867256637,
      "grad_norm": 1.3663618680936906e-08,
      "learning_rate": 0.11711757212470802,
      "loss": 0.0,
      "num_input_tokens_seen": 12957136,
      "step": 22815
    },
    {
      "epoch": 400.353982300885,
      "grad_norm": 9.915761545187252e-09,
      "learning_rate": 0.11706010258292165,
      "loss": 0.0,
      "num_input_tokens_seen": 12959712,
      "step": 22820
    },
    {
      "epoch": 400.4424778761062,
      "grad_norm": 6.230533866613541e-09,
      "learning_rate": 0.11700263812088131,
      "loss": 0.0,
      "num_input_tokens_seen": 12962416,
      "step": 22825
    },
    {
      "epoch": 400.53097345132744,
      "grad_norm": 6.045437483948035e-09,
      "learning_rate": 0.11694517874744892,
      "loss": 0.0,
      "num_input_tokens_seen": 12965456,
      "step": 22830
    },
    {
      "epoch": 400.6194690265487,
      "grad_norm": 1.4797997494042647e-08,
      "learning_rate": 0.11688772447148532,
      "loss": 0.0,
      "num_input_tokens_seen": 12968480,
      "step": 22835
    },
    {
      "epoch": 400.7079646017699,
      "grad_norm": 3.485905608613393e-08,
      "learning_rate": 0.11683027530185074,
      "loss": 0.0,
      "num_input_tokens_seen": 12971376,
      "step": 22840
    },
    {
      "epoch": 400.79646017699116,
      "grad_norm": 1.717506847853656e-08,
      "learning_rate": 0.11677283124740451,
      "loss": 0.0,
      "num_input_tokens_seen": 12974288,
      "step": 22845
    },
    {
      "epoch": 400.88495575221236,
      "grad_norm": 2.5411662107899247e-08,
      "learning_rate": 0.11671539231700531,
      "loss": 0.0,
      "num_input_tokens_seen": 12976928,
      "step": 22850
    },
    {
      "epoch": 400.9734513274336,
      "grad_norm": 1.417283357341148e-08,
      "learning_rate": 0.11665795851951084,
      "loss": 0.0,
      "num_input_tokens_seen": 12979712,
      "step": 22855
    },
    {
      "epoch": 401.05309734513276,
      "grad_norm": 1.56341446455599e-08,
      "learning_rate": 0.11660052986377825,
      "loss": 0.0,
      "num_input_tokens_seen": 12982208,
      "step": 22860
    },
    {
      "epoch": 401.14159292035396,
      "grad_norm": 1.665713256215895e-08,
      "learning_rate": 0.1165431063586636,
      "loss": 0.0,
      "num_input_tokens_seen": 12985184,
      "step": 22865
    },
    {
      "epoch": 401.2300884955752,
      "grad_norm": 1.4885169541400955e-08,
      "learning_rate": 0.11648568801302245,
      "loss": 0.0,
      "num_input_tokens_seen": 12987984,
      "step": 22870
    },
    {
      "epoch": 401.3185840707965,
      "grad_norm": 2.28700116622349e-08,
      "learning_rate": 0.11642827483570937,
      "loss": 0.0,
      "num_input_tokens_seen": 12990400,
      "step": 22875
    },
    {
      "epoch": 401.4070796460177,
      "grad_norm": 1.6769041266684326e-08,
      "learning_rate": 0.11637086683557815,
      "loss": 0.0,
      "num_input_tokens_seen": 12993296,
      "step": 22880
    },
    {
      "epoch": 401.49557522123894,
      "grad_norm": 1.4107182977340926e-08,
      "learning_rate": 0.11631346402148188,
      "loss": 0.0,
      "num_input_tokens_seen": 12996224,
      "step": 22885
    },
    {
      "epoch": 401.5840707964602,
      "grad_norm": 3.0523256810965904e-08,
      "learning_rate": 0.11625606640227285,
      "loss": 0.0,
      "num_input_tokens_seen": 12999248,
      "step": 22890
    },
    {
      "epoch": 401.6725663716814,
      "grad_norm": 1.4714084173306219e-08,
      "learning_rate": 0.11619867398680238,
      "loss": 0.0,
      "num_input_tokens_seen": 13002672,
      "step": 22895
    },
    {
      "epoch": 401.76106194690266,
      "grad_norm": 1.1894740303830531e-08,
      "learning_rate": 0.11614128678392119,
      "loss": 0.0,
      "num_input_tokens_seen": 13005984,
      "step": 22900
    },
    {
      "epoch": 401.8495575221239,
      "grad_norm": 6.3458962529239216e-09,
      "learning_rate": 0.11608390480247906,
      "loss": 0.0,
      "num_input_tokens_seen": 13008480,
      "step": 22905
    },
    {
      "epoch": 401.9380530973451,
      "grad_norm": 1.9208753698762848e-08,
      "learning_rate": 0.11602652805132499,
      "loss": 0.0,
      "num_input_tokens_seen": 13011088,
      "step": 22910
    },
    {
      "epoch": 402.01769911504425,
      "grad_norm": 7.036338622867788e-09,
      "learning_rate": 0.11596915653930731,
      "loss": 0.0,
      "num_input_tokens_seen": 13013504,
      "step": 22915
    },
    {
      "epoch": 402.1061946902655,
      "grad_norm": 1.71735869969325e-08,
      "learning_rate": 0.11591179027527328,
      "loss": 0.0,
      "num_input_tokens_seen": 13016416,
      "step": 22920
    },
    {
      "epoch": 402.1946902654867,
      "grad_norm": 5.721998874719247e-09,
      "learning_rate": 0.11585442926806956,
      "loss": 0.0,
      "num_input_tokens_seen": 13018992,
      "step": 22925
    },
    {
      "epoch": 402.283185840708,
      "grad_norm": 2.736266502267881e-08,
      "learning_rate": 0.11579707352654202,
      "loss": 0.0,
      "num_input_tokens_seen": 13022208,
      "step": 22930
    },
    {
      "epoch": 402.37168141592923,
      "grad_norm": 1.415112471647717e-08,
      "learning_rate": 0.11573972305953548,
      "loss": 0.0,
      "num_input_tokens_seen": 13024912,
      "step": 22935
    },
    {
      "epoch": 402.46017699115043,
      "grad_norm": 1.6397095237152826e-08,
      "learning_rate": 0.11568237787589426,
      "loss": 0.0,
      "num_input_tokens_seen": 13027872,
      "step": 22940
    },
    {
      "epoch": 402.5486725663717,
      "grad_norm": 4.694610922229003e-09,
      "learning_rate": 0.11562503798446161,
      "loss": 0.0,
      "num_input_tokens_seen": 13030640,
      "step": 22945
    },
    {
      "epoch": 402.6371681415929,
      "grad_norm": 1.8762506215352914e-08,
      "learning_rate": 0.11556770339408005,
      "loss": 0.0,
      "num_input_tokens_seen": 13033904,
      "step": 22950
    },
    {
      "epoch": 402.72566371681415,
      "grad_norm": 9.63910373741328e-09,
      "learning_rate": 0.1155103741135914,
      "loss": 0.0,
      "num_input_tokens_seen": 13036784,
      "step": 22955
    },
    {
      "epoch": 402.8141592920354,
      "grad_norm": 2.2597976823135468e-08,
      "learning_rate": 0.1154530501518364,
      "loss": 0.0,
      "num_input_tokens_seen": 13039744,
      "step": 22960
    },
    {
      "epoch": 402.9026548672566,
      "grad_norm": 1.3377373875300691e-08,
      "learning_rate": 0.11539573151765523,
      "loss": 0.0,
      "num_input_tokens_seen": 13042384,
      "step": 22965
    },
    {
      "epoch": 402.9911504424779,
      "grad_norm": 2.3121671688386414e-08,
      "learning_rate": 0.11533841821988719,
      "loss": 0.0,
      "num_input_tokens_seen": 13045216,
      "step": 22970
    },
    {
      "epoch": 403.070796460177,
      "grad_norm": 1.1252958564966775e-08,
      "learning_rate": 0.11528111026737059,
      "loss": 0.0,
      "num_input_tokens_seen": 13047600,
      "step": 22975
    },
    {
      "epoch": 403.1592920353982,
      "grad_norm": 6.688443132674138e-09,
      "learning_rate": 0.11522380766894312,
      "loss": 0.0,
      "num_input_tokens_seen": 13049888,
      "step": 22980
    },
    {
      "epoch": 403.24778761061947,
      "grad_norm": 8.139422469355395e-09,
      "learning_rate": 0.11516651043344152,
      "loss": 0.0,
      "num_input_tokens_seen": 13052912,
      "step": 22985
    },
    {
      "epoch": 403.3362831858407,
      "grad_norm": 9.687149749026958e-09,
      "learning_rate": 0.11510921856970172,
      "loss": 0.0,
      "num_input_tokens_seen": 13055584,
      "step": 22990
    },
    {
      "epoch": 403.42477876106193,
      "grad_norm": 1.5801285613292748e-08,
      "learning_rate": 0.11505193208655895,
      "loss": 0.0,
      "num_input_tokens_seen": 13058736,
      "step": 22995
    },
    {
      "epoch": 403.5132743362832,
      "grad_norm": 8.92944651553762e-09,
      "learning_rate": 0.11499465099284738,
      "loss": 0.0,
      "num_input_tokens_seen": 13061472,
      "step": 23000
    },
    {
      "epoch": 403.5132743362832,
      "eval_loss": 0.6734206676483154,
      "eval_runtime": 0.9379,
      "eval_samples_per_second": 26.656,
      "eval_steps_per_second": 13.861,
      "num_input_tokens_seen": 13061472,
      "step": 23000
    },
    {
      "epoch": 403.60176991150445,
      "grad_norm": 1.771767443869976e-08,
      "learning_rate": 0.1149373752974006,
      "loss": 0.0,
      "num_input_tokens_seen": 13064528,
      "step": 23005
    },
    {
      "epoch": 403.69026548672565,
      "grad_norm": 1.0293218721812991e-08,
      "learning_rate": 0.11488010500905109,
      "loss": 0.0,
      "num_input_tokens_seen": 13067712,
      "step": 23010
    },
    {
      "epoch": 403.7787610619469,
      "grad_norm": 1.4144703186502738e-08,
      "learning_rate": 0.11482284013663077,
      "loss": 0.0,
      "num_input_tokens_seen": 13070528,
      "step": 23015
    },
    {
      "epoch": 403.86725663716817,
      "grad_norm": 1.1841510882959483e-08,
      "learning_rate": 0.11476558068897061,
      "loss": 0.0,
      "num_input_tokens_seen": 13073328,
      "step": 23020
    },
    {
      "epoch": 403.95575221238937,
      "grad_norm": 3.120117852972726e-08,
      "learning_rate": 0.11470832667490061,
      "loss": 0.0,
      "num_input_tokens_seen": 13076688,
      "step": 23025
    },
    {
      "epoch": 404.0353982300885,
      "grad_norm": 6.324176293759365e-09,
      "learning_rate": 0.11465107810325013,
      "loss": 0.0,
      "num_input_tokens_seen": 13079304,
      "step": 23030
    },
    {
      "epoch": 404.12389380530976,
      "grad_norm": 1.684606232288388e-08,
      "learning_rate": 0.11459383498284771,
      "loss": 0.0,
      "num_input_tokens_seen": 13082024,
      "step": 23035
    },
    {
      "epoch": 404.21238938053096,
      "grad_norm": 1.900956725364722e-08,
      "learning_rate": 0.11453659732252082,
      "loss": 0.0,
      "num_input_tokens_seen": 13084680,
      "step": 23040
    },
    {
      "epoch": 404.3008849557522,
      "grad_norm": 7.702327664560471e-09,
      "learning_rate": 0.11447936513109633,
      "loss": 0.0,
      "num_input_tokens_seen": 13087848,
      "step": 23045
    },
    {
      "epoch": 404.3893805309734,
      "grad_norm": 2.7196165319764987e-08,
      "learning_rate": 0.11442213841740011,
      "loss": 0.0,
      "num_input_tokens_seen": 13090840,
      "step": 23050
    },
    {
      "epoch": 404.4778761061947,
      "grad_norm": 1.5479223236525286e-08,
      "learning_rate": 0.1143649171902572,
      "loss": 0.0,
      "num_input_tokens_seen": 13093688,
      "step": 23055
    },
    {
      "epoch": 404.56637168141594,
      "grad_norm": 6.005317576551761e-09,
      "learning_rate": 0.11430770145849194,
      "loss": 0.0,
      "num_input_tokens_seen": 13096616,
      "step": 23060
    },
    {
      "epoch": 404.65486725663715,
      "grad_norm": 1.1879356165422905e-08,
      "learning_rate": 0.11425049123092756,
      "loss": 0.0,
      "num_input_tokens_seen": 13099816,
      "step": 23065
    },
    {
      "epoch": 404.7433628318584,
      "grad_norm": 1.3877059501510303e-08,
      "learning_rate": 0.11419328651638674,
      "loss": 0.0,
      "num_input_tokens_seen": 13102680,
      "step": 23070
    },
    {
      "epoch": 404.83185840707966,
      "grad_norm": 2.6002698660931856e-08,
      "learning_rate": 0.11413608732369115,
      "loss": 0.0,
      "num_input_tokens_seen": 13105320,
      "step": 23075
    },
    {
      "epoch": 404.92035398230087,
      "grad_norm": 2.0933553557256346e-08,
      "learning_rate": 0.11407889366166153,
      "loss": 0.0,
      "num_input_tokens_seen": 13107912,
      "step": 23080
    },
    {
      "epoch": 405.0,
      "grad_norm": 1.389931103545905e-08,
      "learning_rate": 0.11402170553911797,
      "loss": 0.0,
      "num_input_tokens_seen": 13110120,
      "step": 23085
    },
    {
      "epoch": 405.08849557522126,
      "grad_norm": 2.5882272325361555e-08,
      "learning_rate": 0.11396452296487955,
      "loss": 0.0,
      "num_input_tokens_seen": 13112888,
      "step": 23090
    },
    {
      "epoch": 405.17699115044246,
      "grad_norm": 2.8278396513314874e-08,
      "learning_rate": 0.11390734594776449,
      "loss": 0.0,
      "num_input_tokens_seen": 13115752,
      "step": 23095
    },
    {
      "epoch": 405.2654867256637,
      "grad_norm": 2.7061229701530465e-08,
      "learning_rate": 0.11385017449659031,
      "loss": 0.0,
      "num_input_tokens_seen": 13118424,
      "step": 23100
    },
    {
      "epoch": 405.353982300885,
      "grad_norm": 1.2547345384916753e-08,
      "learning_rate": 0.11379300862017344,
      "loss": 0.0,
      "num_input_tokens_seen": 13121432,
      "step": 23105
    },
    {
      "epoch": 405.4424778761062,
      "grad_norm": 3.2688351581100505e-08,
      "learning_rate": 0.11373584832732966,
      "loss": 0.0,
      "num_input_tokens_seen": 13124024,
      "step": 23110
    },
    {
      "epoch": 405.53097345132744,
      "grad_norm": 1.5208270198741047e-08,
      "learning_rate": 0.11367869362687386,
      "loss": 0.0,
      "num_input_tokens_seen": 13127096,
      "step": 23115
    },
    {
      "epoch": 405.6194690265487,
      "grad_norm": 1.1899675023130385e-08,
      "learning_rate": 0.11362154452761988,
      "loss": 0.0,
      "num_input_tokens_seen": 13130136,
      "step": 23120
    },
    {
      "epoch": 405.7079646017699,
      "grad_norm": 1.6653078915851438e-08,
      "learning_rate": 0.11356440103838095,
      "loss": 0.0,
      "num_input_tokens_seen": 13133576,
      "step": 23125
    },
    {
      "epoch": 405.79646017699116,
      "grad_norm": 3.01733571461682e-08,
      "learning_rate": 0.11350726316796922,
      "loss": 0.0,
      "num_input_tokens_seen": 13136472,
      "step": 23130
    },
    {
      "epoch": 405.88495575221236,
      "grad_norm": 1.4410870718961633e-08,
      "learning_rate": 0.11345013092519607,
      "loss": 0.0,
      "num_input_tokens_seen": 13139368,
      "step": 23135
    },
    {
      "epoch": 405.9734513274336,
      "grad_norm": 1.582699304947255e-08,
      "learning_rate": 0.11339300431887213,
      "loss": 0.0,
      "num_input_tokens_seen": 13142104,
      "step": 23140
    },
    {
      "epoch": 406.05309734513276,
      "grad_norm": 2.502267903992106e-08,
      "learning_rate": 0.11333588335780687,
      "loss": 0.0,
      "num_input_tokens_seen": 13144904,
      "step": 23145
    },
    {
      "epoch": 406.14159292035396,
      "grad_norm": 1.275447480963976e-08,
      "learning_rate": 0.11327876805080916,
      "loss": 0.0,
      "num_input_tokens_seen": 13148008,
      "step": 23150
    },
    {
      "epoch": 406.2300884955752,
      "grad_norm": 1.990478359914505e-08,
      "learning_rate": 0.11322165840668696,
      "loss": 0.0,
      "num_input_tokens_seen": 13151032,
      "step": 23155
    },
    {
      "epoch": 406.3185840707965,
      "grad_norm": 3.045416008262691e-08,
      "learning_rate": 0.11316455443424717,
      "loss": 0.0,
      "num_input_tokens_seen": 13154152,
      "step": 23160
    },
    {
      "epoch": 406.4070796460177,
      "grad_norm": 1.721425135770005e-08,
      "learning_rate": 0.11310745614229603,
      "loss": 0.0,
      "num_input_tokens_seen": 13156744,
      "step": 23165
    },
    {
      "epoch": 406.49557522123894,
      "grad_norm": 2.7009129155430855e-08,
      "learning_rate": 0.1130503635396387,
      "loss": 0.0,
      "num_input_tokens_seen": 13159512,
      "step": 23170
    },
    {
      "epoch": 406.5840707964602,
      "grad_norm": 1.8396399070752523e-08,
      "learning_rate": 0.11299327663507966,
      "loss": 0.0,
      "num_input_tokens_seen": 13162248,
      "step": 23175
    },
    {
      "epoch": 406.6725663716814,
      "grad_norm": 7.2525634386977345e-09,
      "learning_rate": 0.11293619543742246,
      "loss": 0.0,
      "num_input_tokens_seen": 13165464,
      "step": 23180
    },
    {
      "epoch": 406.76106194690266,
      "grad_norm": 1.1231595209437728e-08,
      "learning_rate": 0.11287911995546965,
      "loss": 0.0,
      "num_input_tokens_seen": 13167848,
      "step": 23185
    },
    {
      "epoch": 406.8495575221239,
      "grad_norm": 2.610593341501044e-08,
      "learning_rate": 0.11282205019802308,
      "loss": 0.0,
      "num_input_tokens_seen": 13170648,
      "step": 23190
    },
    {
      "epoch": 406.9380530973451,
      "grad_norm": 1.7198015456187932e-08,
      "learning_rate": 0.11276498617388354,
      "loss": 0.0,
      "num_input_tokens_seen": 13173528,
      "step": 23195
    },
    {
      "epoch": 407.01769911504425,
      "grad_norm": 2.9755598873748568e-08,
      "learning_rate": 0.11270792789185109,
      "loss": 0.0,
      "num_input_tokens_seen": 13175888,
      "step": 23200
    },
    {
      "epoch": 407.01769911504425,
      "eval_loss": 0.6624284982681274,
      "eval_runtime": 0.9383,
      "eval_samples_per_second": 26.645,
      "eval_steps_per_second": 13.855,
      "num_input_tokens_seen": 13175888,
      "step": 23200
    },
    {
      "epoch": 407.1061946902655,
      "grad_norm": 5.904866640094042e-08,
      "learning_rate": 0.11265087536072482,
      "loss": 0.0,
      "num_input_tokens_seen": 13178864,
      "step": 23205
    },
    {
      "epoch": 407.1946902654867,
      "grad_norm": 1.0544271233925429e-08,
      "learning_rate": 0.11259382858930288,
      "loss": 0.0,
      "num_input_tokens_seen": 13181568,
      "step": 23210
    },
    {
      "epoch": 407.283185840708,
      "grad_norm": 3.1689801005541085e-08,
      "learning_rate": 0.11253678758638262,
      "loss": 0.0,
      "num_input_tokens_seen": 13184592,
      "step": 23215
    },
    {
      "epoch": 407.37168141592923,
      "grad_norm": 4.387031182773171e-09,
      "learning_rate": 0.11247975236076059,
      "loss": 0.0,
      "num_input_tokens_seen": 13187648,
      "step": 23220
    },
    {
      "epoch": 407.46017699115043,
      "grad_norm": 6.863122070654981e-09,
      "learning_rate": 0.11242272292123218,
      "loss": 0.0,
      "num_input_tokens_seen": 13190576,
      "step": 23225
    },
    {
      "epoch": 407.5486725663717,
      "grad_norm": 9.499886211017383e-09,
      "learning_rate": 0.11236569927659217,
      "loss": 0.0,
      "num_input_tokens_seen": 13193664,
      "step": 23230
    },
    {
      "epoch": 407.6371681415929,
      "grad_norm": 1.5419667320770714e-08,
      "learning_rate": 0.11230868143563429,
      "loss": 0.0,
      "num_input_tokens_seen": 13196592,
      "step": 23235
    },
    {
      "epoch": 407.72566371681415,
      "grad_norm": 1.3003311316595045e-08,
      "learning_rate": 0.11225166940715131,
      "loss": 0.0,
      "num_input_tokens_seen": 13199264,
      "step": 23240
    },
    {
      "epoch": 407.8141592920354,
      "grad_norm": 7.246398592286596e-09,
      "learning_rate": 0.11219466319993537,
      "loss": 0.0,
      "num_input_tokens_seen": 13202096,
      "step": 23245
    },
    {
      "epoch": 407.9026548672566,
      "grad_norm": 1.9832606668046537e-08,
      "learning_rate": 0.11213766282277739,
      "loss": 0.0,
      "num_input_tokens_seen": 13205088,
      "step": 23250
    },
    {
      "epoch": 407.9911504424779,
      "grad_norm": 4.228149563800798e-08,
      "learning_rate": 0.11208066828446761,
      "loss": 0.0,
      "num_input_tokens_seen": 13207872,
      "step": 23255
    },
    {
      "epoch": 408.070796460177,
      "grad_norm": 1.3120314612535822e-08,
      "learning_rate": 0.11202367959379537,
      "loss": 0.0,
      "num_input_tokens_seen": 13210088,
      "step": 23260
    },
    {
      "epoch": 408.1592920353982,
      "grad_norm": 1.497315160747803e-08,
      "learning_rate": 0.11196669675954894,
      "loss": 0.0,
      "num_input_tokens_seen": 13213192,
      "step": 23265
    },
    {
      "epoch": 408.24778761061947,
      "grad_norm": 1.1162573976264412e-08,
      "learning_rate": 0.1119097197905158,
      "loss": 0.0,
      "num_input_tokens_seen": 13215784,
      "step": 23270
    },
    {
      "epoch": 408.3362831858407,
      "grad_norm": 1.4044536200685798e-08,
      "learning_rate": 0.11185274869548259,
      "loss": 0.0,
      "num_input_tokens_seen": 13219048,
      "step": 23275
    },
    {
      "epoch": 408.42477876106193,
      "grad_norm": 1.5450698498398197e-08,
      "learning_rate": 0.11179578348323486,
      "loss": 0.0,
      "num_input_tokens_seen": 13221960,
      "step": 23280
    },
    {
      "epoch": 408.5132743362832,
      "grad_norm": 2.975899882073918e-08,
      "learning_rate": 0.1117388241625575,
      "loss": 0.0,
      "num_input_tokens_seen": 13224824,
      "step": 23285
    },
    {
      "epoch": 408.60176991150445,
      "grad_norm": 1.9304582821177974e-08,
      "learning_rate": 0.11168187074223421,
      "loss": 0.0,
      "num_input_tokens_seen": 13227688,
      "step": 23290
    },
    {
      "epoch": 408.69026548672565,
      "grad_norm": 9.234964792881328e-09,
      "learning_rate": 0.11162492323104796,
      "loss": 0.0,
      "num_input_tokens_seen": 13230552,
      "step": 23295
    },
    {
      "epoch": 408.7787610619469,
      "grad_norm": 8.634789772088425e-09,
      "learning_rate": 0.11156798163778091,
      "loss": 0.0,
      "num_input_tokens_seen": 13233128,
      "step": 23300
    },
    {
      "epoch": 408.86725663716817,
      "grad_norm": 1.545158134774738e-08,
      "learning_rate": 0.11151104597121399,
      "loss": 0.0,
      "num_input_tokens_seen": 13236200,
      "step": 23305
    },
    {
      "epoch": 408.95575221238937,
      "grad_norm": 1.3841445323237167e-08,
      "learning_rate": 0.11145411624012742,
      "loss": 0.0,
      "num_input_tokens_seen": 13239048,
      "step": 23310
    },
    {
      "epoch": 409.0353982300885,
      "grad_norm": 4.715678869615658e-08,
      "learning_rate": 0.11139719245330063,
      "loss": 0.0,
      "num_input_tokens_seen": 13241352,
      "step": 23315
    },
    {
      "epoch": 409.12389380530976,
      "grad_norm": 1.3138969023884783e-08,
      "learning_rate": 0.11134027461951179,
      "loss": 0.0,
      "num_input_tokens_seen": 13244296,
      "step": 23320
    },
    {
      "epoch": 409.21238938053096,
      "grad_norm": 1.742924915504318e-08,
      "learning_rate": 0.11128336274753849,
      "loss": 0.0,
      "num_input_tokens_seen": 13247288,
      "step": 23325
    },
    {
      "epoch": 409.3008849557522,
      "grad_norm": 1.8698862902510882e-08,
      "learning_rate": 0.11122645684615715,
      "loss": 0.0,
      "num_input_tokens_seen": 13250040,
      "step": 23330
    },
    {
      "epoch": 409.3893805309734,
      "grad_norm": 5.7857607593803095e-09,
      "learning_rate": 0.11116955692414345,
      "loss": 0.0,
      "num_input_tokens_seen": 13252664,
      "step": 23335
    },
    {
      "epoch": 409.4778761061947,
      "grad_norm": 1.854470177420353e-08,
      "learning_rate": 0.11111266299027203,
      "loss": 0.0,
      "num_input_tokens_seen": 13255512,
      "step": 23340
    },
    {
      "epoch": 409.56637168141594,
      "grad_norm": 3.695391015412497e-08,
      "learning_rate": 0.11105577505331668,
      "loss": 0.0,
      "num_input_tokens_seen": 13258952,
      "step": 23345
    },
    {
      "epoch": 409.65486725663715,
      "grad_norm": 1.4777440604518688e-08,
      "learning_rate": 0.11099889312205018,
      "loss": 0.0,
      "num_input_tokens_seen": 13262264,
      "step": 23350
    },
    {
      "epoch": 409.7433628318584,
      "grad_norm": 1.2590103182219536e-08,
      "learning_rate": 0.11094201720524455,
      "loss": 0.0,
      "num_input_tokens_seen": 13264920,
      "step": 23355
    },
    {
      "epoch": 409.83185840707966,
      "grad_norm": 2.3628235368278183e-08,
      "learning_rate": 0.11088514731167064,
      "loss": 0.0,
      "num_input_tokens_seen": 13267336,
      "step": 23360
    },
    {
      "epoch": 409.92035398230087,
      "grad_norm": 1.117708947617757e-08,
      "learning_rate": 0.11082828345009862,
      "loss": 0.0,
      "num_input_tokens_seen": 13270216,
      "step": 23365
    },
    {
      "epoch": 410.0,
      "grad_norm": 1.4269455839155398e-08,
      "learning_rate": 0.11077142562929748,
      "loss": 0.0,
      "num_input_tokens_seen": 13272776,
      "step": 23370
    },
    {
      "epoch": 410.08849557522126,
      "grad_norm": 1.4635761047543383e-08,
      "learning_rate": 0.11071457385803554,
      "loss": 0.0,
      "num_input_tokens_seen": 13275736,
      "step": 23375
    },
    {
      "epoch": 410.17699115044246,
      "grad_norm": 3.40902666096099e-08,
      "learning_rate": 0.11065772814508001,
      "loss": 0.0,
      "num_input_tokens_seen": 13278712,
      "step": 23380
    },
    {
      "epoch": 410.2654867256637,
      "grad_norm": 2.284498812343827e-08,
      "learning_rate": 0.11060088849919715,
      "loss": 0.0,
      "num_input_tokens_seen": 13281368,
      "step": 23385
    },
    {
      "epoch": 410.353982300885,
      "grad_norm": 1.7419941045204723e-08,
      "learning_rate": 0.11054405492915244,
      "loss": 0.0,
      "num_input_tokens_seen": 13284088,
      "step": 23390
    },
    {
      "epoch": 410.4424778761062,
      "grad_norm": 6.968247312499898e-09,
      "learning_rate": 0.11048722744371031,
      "loss": 0.0,
      "num_input_tokens_seen": 13287048,
      "step": 23395
    },
    {
      "epoch": 410.53097345132744,
      "grad_norm": 3.521601854572509e-08,
      "learning_rate": 0.1104304060516342,
      "loss": 0.0,
      "num_input_tokens_seen": 13289752,
      "step": 23400
    },
    {
      "epoch": 410.53097345132744,
      "eval_loss": 0.6691233515739441,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.587,
      "eval_steps_per_second": 13.825,
      "num_input_tokens_seen": 13289752,
      "step": 23400
    },
    {
      "epoch": 410.6194690265487,
      "grad_norm": 1.3993168401782441e-08,
      "learning_rate": 0.11037359076168682,
      "loss": 0.0,
      "num_input_tokens_seen": 13292584,
      "step": 23405
    },
    {
      "epoch": 410.7079646017699,
      "grad_norm": 1.5348335935527757e-08,
      "learning_rate": 0.11031678158262966,
      "loss": 0.0,
      "num_input_tokens_seen": 13295704,
      "step": 23410
    },
    {
      "epoch": 410.79646017699116,
      "grad_norm": 2.4803979314924618e-08,
      "learning_rate": 0.11025997852322349,
      "loss": 0.0,
      "num_input_tokens_seen": 13298600,
      "step": 23415
    },
    {
      "epoch": 410.88495575221236,
      "grad_norm": 1.5970096356454633e-08,
      "learning_rate": 0.11020318159222807,
      "loss": 0.0,
      "num_input_tokens_seen": 13301912,
      "step": 23420
    },
    {
      "epoch": 410.9734513274336,
      "grad_norm": 8.863681344450924e-09,
      "learning_rate": 0.1101463907984021,
      "loss": 0.0,
      "num_input_tokens_seen": 13304472,
      "step": 23425
    },
    {
      "epoch": 411.05309734513276,
      "grad_norm": 8.989633926148599e-09,
      "learning_rate": 0.11008960615050352,
      "loss": 0.0,
      "num_input_tokens_seen": 13307080,
      "step": 23430
    },
    {
      "epoch": 411.14159292035396,
      "grad_norm": 1.6773583411122672e-08,
      "learning_rate": 0.11003282765728925,
      "loss": 0.0,
      "num_input_tokens_seen": 13310152,
      "step": 23435
    },
    {
      "epoch": 411.2300884955752,
      "grad_norm": 1.326870702200722e-08,
      "learning_rate": 0.10997605532751518,
      "loss": 0.0,
      "num_input_tokens_seen": 13312840,
      "step": 23440
    },
    {
      "epoch": 411.3185840707965,
      "grad_norm": 1.2515298131177133e-08,
      "learning_rate": 0.1099192891699364,
      "loss": 0.0,
      "num_input_tokens_seen": 13316168,
      "step": 23445
    },
    {
      "epoch": 411.4070796460177,
      "grad_norm": 2.2525105336512752e-08,
      "learning_rate": 0.10986252919330687,
      "loss": 0.0,
      "num_input_tokens_seen": 13319304,
      "step": 23450
    },
    {
      "epoch": 411.49557522123894,
      "grad_norm": 9.780164234030053e-09,
      "learning_rate": 0.10980577540637973,
      "loss": 0.0,
      "num_input_tokens_seen": 13321912,
      "step": 23455
    },
    {
      "epoch": 411.5840707964602,
      "grad_norm": 1.3979986945855671e-08,
      "learning_rate": 0.10974902781790719,
      "loss": 0.0,
      "num_input_tokens_seen": 13324232,
      "step": 23460
    },
    {
      "epoch": 411.6725663716814,
      "grad_norm": 2.0850441373454487e-08,
      "learning_rate": 0.10969228643664032,
      "loss": 0.0,
      "num_input_tokens_seen": 13327016,
      "step": 23465
    },
    {
      "epoch": 411.76106194690266,
      "grad_norm": 2.302249413332902e-08,
      "learning_rate": 0.10963555127132942,
      "loss": 0.0,
      "num_input_tokens_seen": 13329784,
      "step": 23470
    },
    {
      "epoch": 411.8495575221239,
      "grad_norm": 2.3937388959893724e-08,
      "learning_rate": 0.10957882233072382,
      "loss": 0.0,
      "num_input_tokens_seen": 13332616,
      "step": 23475
    },
    {
      "epoch": 411.9380530973451,
      "grad_norm": 2.9374922050351415e-08,
      "learning_rate": 0.10952209962357176,
      "loss": 0.0,
      "num_input_tokens_seen": 13335736,
      "step": 23480
    },
    {
      "epoch": 412.01769911504425,
      "grad_norm": 2.4371290763269826e-08,
      "learning_rate": 0.10946538315862062,
      "loss": 0.0,
      "num_input_tokens_seen": 13338224,
      "step": 23485
    },
    {
      "epoch": 412.1061946902655,
      "grad_norm": 1.6097486010835382e-08,
      "learning_rate": 0.10940867294461679,
      "loss": 0.0,
      "num_input_tokens_seen": 13341264,
      "step": 23490
    },
    {
      "epoch": 412.1946902654867,
      "grad_norm": 1.0998035371301285e-08,
      "learning_rate": 0.10935196899030565,
      "loss": 0.0,
      "num_input_tokens_seen": 13344048,
      "step": 23495
    },
    {
      "epoch": 412.283185840708,
      "grad_norm": 5.5595510417560945e-09,
      "learning_rate": 0.10929527130443177,
      "loss": 0.0,
      "num_input_tokens_seen": 13346416,
      "step": 23500
    },
    {
      "epoch": 412.37168141592923,
      "grad_norm": 2.810078392201376e-08,
      "learning_rate": 0.1092385798957385,
      "loss": 0.0,
      "num_input_tokens_seen": 13349984,
      "step": 23505
    },
    {
      "epoch": 412.46017699115043,
      "grad_norm": 1.6443516770436872e-08,
      "learning_rate": 0.10918189477296848,
      "loss": 0.0,
      "num_input_tokens_seen": 13352736,
      "step": 23510
    },
    {
      "epoch": 412.5486725663717,
      "grad_norm": 1.9833244380151882e-08,
      "learning_rate": 0.1091252159448633,
      "loss": 0.0,
      "num_input_tokens_seen": 13355136,
      "step": 23515
    },
    {
      "epoch": 412.6371681415929,
      "grad_norm": 1.0777743142398322e-08,
      "learning_rate": 0.10906854342016345,
      "loss": 0.0,
      "num_input_tokens_seen": 13358032,
      "step": 23520
    },
    {
      "epoch": 412.72566371681415,
      "grad_norm": 1.5804111797024234e-08,
      "learning_rate": 0.10901187720760858,
      "loss": 0.0,
      "num_input_tokens_seen": 13360528,
      "step": 23525
    },
    {
      "epoch": 412.8141592920354,
      "grad_norm": 2.0875198458725208e-08,
      "learning_rate": 0.10895521731593734,
      "loss": 0.0,
      "num_input_tokens_seen": 13363904,
      "step": 23530
    },
    {
      "epoch": 412.9026548672566,
      "grad_norm": 2.4561208178397465e-08,
      "learning_rate": 0.10889856375388733,
      "loss": 0.0,
      "num_input_tokens_seen": 13366576,
      "step": 23535
    },
    {
      "epoch": 412.9911504424779,
      "grad_norm": 1.8343975227708142e-08,
      "learning_rate": 0.1088419165301954,
      "loss": 0.0,
      "num_input_tokens_seen": 13369584,
      "step": 23540
    },
    {
      "epoch": 413.070796460177,
      "grad_norm": 2.4901494199980334e-08,
      "learning_rate": 0.1087852756535971,
      "loss": 0.0,
      "num_input_tokens_seen": 13371928,
      "step": 23545
    },
    {
      "epoch": 413.1592920353982,
      "grad_norm": 1.4291774874664043e-08,
      "learning_rate": 0.10872864113282725,
      "loss": 0.0,
      "num_input_tokens_seen": 13374888,
      "step": 23550
    },
    {
      "epoch": 413.24778761061947,
      "grad_norm": 1.9592768296661234e-08,
      "learning_rate": 0.10867201297661958,
      "loss": 0.0,
      "num_input_tokens_seen": 13377848,
      "step": 23555
    },
    {
      "epoch": 413.3362831858407,
      "grad_norm": 1.2940991389598366e-08,
      "learning_rate": 0.10861539119370689,
      "loss": 0.0,
      "num_input_tokens_seen": 13380408,
      "step": 23560
    },
    {
      "epoch": 413.42477876106193,
      "grad_norm": 1.4583505070220326e-08,
      "learning_rate": 0.10855877579282096,
      "loss": 0.0,
      "num_input_tokens_seen": 13383128,
      "step": 23565
    },
    {
      "epoch": 413.5132743362832,
      "grad_norm": 1.986278341803427e-08,
      "learning_rate": 0.10850216678269252,
      "loss": 0.0,
      "num_input_tokens_seen": 13386104,
      "step": 23570
    },
    {
      "epoch": 413.60176991150445,
      "grad_norm": 2.2927629572677688e-08,
      "learning_rate": 0.10844556417205146,
      "loss": 0.0,
      "num_input_tokens_seen": 13389144,
      "step": 23575
    },
    {
      "epoch": 413.69026548672565,
      "grad_norm": 5.493663479683164e-08,
      "learning_rate": 0.10838896796962669,
      "loss": 0.0,
      "num_input_tokens_seen": 13392312,
      "step": 23580
    },
    {
      "epoch": 413.7787610619469,
      "grad_norm": 2.254039443982947e-08,
      "learning_rate": 0.1083323781841459,
      "loss": 0.0,
      "num_input_tokens_seen": 13395096,
      "step": 23585
    },
    {
      "epoch": 413.86725663716817,
      "grad_norm": 1.772010094214238e-08,
      "learning_rate": 0.10827579482433607,
      "loss": 0.0,
      "num_input_tokens_seen": 13398008,
      "step": 23590
    },
    {
      "epoch": 413.95575221238937,
      "grad_norm": 8.396543904609644e-09,
      "learning_rate": 0.10821921789892304,
      "loss": 0.0,
      "num_input_tokens_seen": 13401112,
      "step": 23595
    },
    {
      "epoch": 414.0353982300885,
      "grad_norm": 8.434754228403563e-09,
      "learning_rate": 0.10816264741663158,
      "loss": 0.0,
      "num_input_tokens_seen": 13403848,
      "step": 23600
    },
    {
      "epoch": 414.0353982300885,
      "eval_loss": 0.6630703806877136,
      "eval_runtime": 0.9183,
      "eval_samples_per_second": 27.224,
      "eval_steps_per_second": 14.156,
      "num_input_tokens_seen": 13403848,
      "step": 23600
    },
    {
      "epoch": 414.12389380530976,
      "grad_norm": 6.935042762279409e-09,
      "learning_rate": 0.10810608338618573,
      "loss": 0.0,
      "num_input_tokens_seen": 13406680,
      "step": 23605
    },
    {
      "epoch": 414.21238938053096,
      "grad_norm": 2.136934185159589e-08,
      "learning_rate": 0.10804952581630821,
      "loss": 0.0,
      "num_input_tokens_seen": 13409160,
      "step": 23610
    },
    {
      "epoch": 414.3008849557522,
      "grad_norm": 1.868936294613377e-08,
      "learning_rate": 0.10799297471572102,
      "loss": 0.0,
      "num_input_tokens_seen": 13411752,
      "step": 23615
    },
    {
      "epoch": 414.3893805309734,
      "grad_norm": 9.654744559384199e-09,
      "learning_rate": 0.10793643009314507,
      "loss": 0.0,
      "num_input_tokens_seen": 13413896,
      "step": 23620
    },
    {
      "epoch": 414.4778761061947,
      "grad_norm": 2.977247604007971e-08,
      "learning_rate": 0.10787989195730015,
      "loss": 0.0,
      "num_input_tokens_seen": 13417048,
      "step": 23625
    },
    {
      "epoch": 414.56637168141594,
      "grad_norm": 1.2168228202824594e-08,
      "learning_rate": 0.10782336031690525,
      "loss": 0.0,
      "num_input_tokens_seen": 13419896,
      "step": 23630
    },
    {
      "epoch": 414.65486725663715,
      "grad_norm": 1.496701962366842e-08,
      "learning_rate": 0.10776683518067821,
      "loss": 0.0,
      "num_input_tokens_seen": 13422920,
      "step": 23635
    },
    {
      "epoch": 414.7433628318584,
      "grad_norm": 2.9012804603212317e-08,
      "learning_rate": 0.10771031655733587,
      "loss": 0.0,
      "num_input_tokens_seen": 13426424,
      "step": 23640
    },
    {
      "epoch": 414.83185840707966,
      "grad_norm": 1.098116442221908e-08,
      "learning_rate": 0.10765380445559422,
      "loss": 0.0,
      "num_input_tokens_seen": 13429304,
      "step": 23645
    },
    {
      "epoch": 414.92035398230087,
      "grad_norm": 2.2911457620011788e-08,
      "learning_rate": 0.10759729888416801,
      "loss": 0.0,
      "num_input_tokens_seen": 13432536,
      "step": 23650
    },
    {
      "epoch": 415.0,
      "grad_norm": 1.0395956095976544e-08,
      "learning_rate": 0.10754079985177119,
      "loss": 0.0,
      "num_input_tokens_seen": 13435104,
      "step": 23655
    },
    {
      "epoch": 415.08849557522126,
      "grad_norm": 2.1259085158931157e-08,
      "learning_rate": 0.10748430736711667,
      "loss": 0.0,
      "num_input_tokens_seen": 13437776,
      "step": 23660
    },
    {
      "epoch": 415.17699115044246,
      "grad_norm": 1.7276256869536155e-08,
      "learning_rate": 0.10742782143891623,
      "loss": 0.0,
      "num_input_tokens_seen": 13440528,
      "step": 23665
    },
    {
      "epoch": 415.2654867256637,
      "grad_norm": 1.6972878214005505e-08,
      "learning_rate": 0.10737134207588069,
      "loss": 0.0,
      "num_input_tokens_seen": 13442944,
      "step": 23670
    },
    {
      "epoch": 415.353982300885,
      "grad_norm": 1.6328147722788344e-08,
      "learning_rate": 0.10731486928671992,
      "loss": 0.0,
      "num_input_tokens_seen": 13445856,
      "step": 23675
    },
    {
      "epoch": 415.4424778761062,
      "grad_norm": 2.218165917611259e-08,
      "learning_rate": 0.10725840308014269,
      "loss": 0.0,
      "num_input_tokens_seen": 13449424,
      "step": 23680
    },
    {
      "epoch": 415.53097345132744,
      "grad_norm": 1.5456746993436354e-08,
      "learning_rate": 0.10720194346485688,
      "loss": 0.0,
      "num_input_tokens_seen": 13452608,
      "step": 23685
    },
    {
      "epoch": 415.6194690265487,
      "grad_norm": 1.153248341267954e-08,
      "learning_rate": 0.10714549044956918,
      "loss": 0.0,
      "num_input_tokens_seen": 13455776,
      "step": 23690
    },
    {
      "epoch": 415.7079646017699,
      "grad_norm": 1.8261339107539243e-08,
      "learning_rate": 0.10708904404298542,
      "loss": 0.0,
      "num_input_tokens_seen": 13459024,
      "step": 23695
    },
    {
      "epoch": 415.79646017699116,
      "grad_norm": 2.5647835855124868e-08,
      "learning_rate": 0.1070326042538103,
      "loss": 0.0,
      "num_input_tokens_seen": 13461776,
      "step": 23700
    },
    {
      "epoch": 415.88495575221236,
      "grad_norm": 5.042371498120701e-08,
      "learning_rate": 0.10697617109074758,
      "loss": 0.0,
      "num_input_tokens_seen": 13464416,
      "step": 23705
    },
    {
      "epoch": 415.9734513274336,
      "grad_norm": 2.424854095295359e-08,
      "learning_rate": 0.10691974456249999,
      "loss": 0.0,
      "num_input_tokens_seen": 13467072,
      "step": 23710
    },
    {
      "epoch": 416.05309734513276,
      "grad_norm": 2.609938754005725e-08,
      "learning_rate": 0.10686332467776909,
      "loss": 0.0,
      "num_input_tokens_seen": 13469936,
      "step": 23715
    },
    {
      "epoch": 416.14159292035396,
      "grad_norm": 1.153541528964297e-08,
      "learning_rate": 0.10680691144525563,
      "loss": 0.0,
      "num_input_tokens_seen": 13472528,
      "step": 23720
    },
    {
      "epoch": 416.2300884955752,
      "grad_norm": 8.795050909782276e-09,
      "learning_rate": 0.10675050487365928,
      "loss": 0.0,
      "num_input_tokens_seen": 13475264,
      "step": 23725
    },
    {
      "epoch": 416.3185840707965,
      "grad_norm": 1.6890737697394798e-08,
      "learning_rate": 0.10669410497167851,
      "loss": 0.0,
      "num_input_tokens_seen": 13478288,
      "step": 23730
    },
    {
      "epoch": 416.4070796460177,
      "grad_norm": 1.765055657187986e-08,
      "learning_rate": 0.10663771174801102,
      "loss": 0.0,
      "num_input_tokens_seen": 13481312,
      "step": 23735
    },
    {
      "epoch": 416.49557522123894,
      "grad_norm": 1.7016301256944644e-08,
      "learning_rate": 0.10658132521135329,
      "loss": 0.0,
      "num_input_tokens_seen": 13484352,
      "step": 23740
    },
    {
      "epoch": 416.5840707964602,
      "grad_norm": 1.950981776133176e-08,
      "learning_rate": 0.10652494537040084,
      "loss": 0.0,
      "num_input_tokens_seen": 13487712,
      "step": 23745
    },
    {
      "epoch": 416.6725663716814,
      "grad_norm": 1.8957804215347096e-08,
      "learning_rate": 0.1064685722338482,
      "loss": 0.0,
      "num_input_tokens_seen": 13490576,
      "step": 23750
    },
    {
      "epoch": 416.76106194690266,
      "grad_norm": 1.077756817124964e-08,
      "learning_rate": 0.10641220581038871,
      "loss": 0.0,
      "num_input_tokens_seen": 13493104,
      "step": 23755
    },
    {
      "epoch": 416.8495575221239,
      "grad_norm": 1.410130678891619e-08,
      "learning_rate": 0.10635584610871483,
      "loss": 0.0,
      "num_input_tokens_seen": 13495792,
      "step": 23760
    },
    {
      "epoch": 416.9380530973451,
      "grad_norm": 1.108687541773179e-08,
      "learning_rate": 0.10629949313751803,
      "loss": 0.0,
      "num_input_tokens_seen": 13498688,
      "step": 23765
    },
    {
      "epoch": 417.01769911504425,
      "grad_norm": 9.97520910317462e-09,
      "learning_rate": 0.10624314690548849,
      "loss": 0.0,
      "num_input_tokens_seen": 13500624,
      "step": 23770
    },
    {
      "epoch": 417.1061946902655,
      "grad_norm": 2.4093974815286856e-08,
      "learning_rate": 0.1061868074213156,
      "loss": 0.0,
      "num_input_tokens_seen": 13503808,
      "step": 23775
    },
    {
      "epoch": 417.1946902654867,
      "grad_norm": 1.6165726535177782e-08,
      "learning_rate": 0.10613047469368765,
      "loss": 0.0,
      "num_input_tokens_seen": 13507136,
      "step": 23780
    },
    {
      "epoch": 417.283185840708,
      "grad_norm": 2.140992627630567e-08,
      "learning_rate": 0.10607414873129171,
      "loss": 0.0,
      "num_input_tokens_seen": 13509728,
      "step": 23785
    },
    {
      "epoch": 417.37168141592923,
      "grad_norm": 1.31434827466137e-08,
      "learning_rate": 0.10601782954281413,
      "loss": 0.0,
      "num_input_tokens_seen": 13512400,
      "step": 23790
    },
    {
      "epoch": 417.46017699115043,
      "grad_norm": 1.7027115717382912e-08,
      "learning_rate": 0.1059615171369399,
      "loss": 0.0,
      "num_input_tokens_seen": 13515904,
      "step": 23795
    },
    {
      "epoch": 417.5486725663717,
      "grad_norm": 1.3262433817828878e-08,
      "learning_rate": 0.10590521152235312,
      "loss": 0.0,
      "num_input_tokens_seen": 13518496,
      "step": 23800
    },
    {
      "epoch": 417.5486725663717,
      "eval_loss": 0.6664594411849976,
      "eval_runtime": 0.9322,
      "eval_samples_per_second": 26.818,
      "eval_steps_per_second": 13.945,
      "num_input_tokens_seen": 13518496,
      "step": 23800
    },
    {
      "epoch": 417.6371681415929,
      "grad_norm": 2.3354372657991007e-08,
      "learning_rate": 0.1058489127077369,
      "loss": 0.0,
      "num_input_tokens_seen": 13521296,
      "step": 23805
    },
    {
      "epoch": 417.72566371681415,
      "grad_norm": 8.658715522358307e-09,
      "learning_rate": 0.1057926207017732,
      "loss": 0.0,
      "num_input_tokens_seen": 13523728,
      "step": 23810
    },
    {
      "epoch": 417.8141592920354,
      "grad_norm": 1.5148568621725644e-08,
      "learning_rate": 0.10573633551314285,
      "loss": 0.0,
      "num_input_tokens_seen": 13526624,
      "step": 23815
    },
    {
      "epoch": 417.9026548672566,
      "grad_norm": 9.920798405005371e-09,
      "learning_rate": 0.1056800571505259,
      "loss": 0.0,
      "num_input_tokens_seen": 13529888,
      "step": 23820
    },
    {
      "epoch": 417.9911504424779,
      "grad_norm": 1.0793336002734577e-08,
      "learning_rate": 0.10562378562260105,
      "loss": 0.0,
      "num_input_tokens_seen": 13532272,
      "step": 23825
    },
    {
      "epoch": 418.070796460177,
      "grad_norm": 1.7681996311580406e-08,
      "learning_rate": 0.10556752093804615,
      "loss": 0.0,
      "num_input_tokens_seen": 13535136,
      "step": 23830
    },
    {
      "epoch": 418.1592920353982,
      "grad_norm": 1.22331327290226e-08,
      "learning_rate": 0.10551126310553786,
      "loss": 0.0,
      "num_input_tokens_seen": 13538032,
      "step": 23835
    },
    {
      "epoch": 418.24778761061947,
      "grad_norm": 9.0353635684437e-09,
      "learning_rate": 0.10545501213375187,
      "loss": 0.0,
      "num_input_tokens_seen": 13540816,
      "step": 23840
    },
    {
      "epoch": 418.3362831858407,
      "grad_norm": 1.5939530584319073e-08,
      "learning_rate": 0.10539876803136287,
      "loss": 0.0,
      "num_input_tokens_seen": 13543776,
      "step": 23845
    },
    {
      "epoch": 418.42477876106193,
      "grad_norm": 1.3554539712856695e-08,
      "learning_rate": 0.10534253080704428,
      "loss": 0.0,
      "num_input_tokens_seen": 13546464,
      "step": 23850
    },
    {
      "epoch": 418.5132743362832,
      "grad_norm": 1.1078339134940052e-08,
      "learning_rate": 0.10528630046946862,
      "loss": 0.0,
      "num_input_tokens_seen": 13549136,
      "step": 23855
    },
    {
      "epoch": 418.60176991150445,
      "grad_norm": 9.127957056875857e-09,
      "learning_rate": 0.1052300770273074,
      "loss": 0.0,
      "num_input_tokens_seen": 13551552,
      "step": 23860
    },
    {
      "epoch": 418.69026548672565,
      "grad_norm": 1.996546927784948e-08,
      "learning_rate": 0.10517386048923086,
      "loss": 0.0,
      "num_input_tokens_seen": 13554240,
      "step": 23865
    },
    {
      "epoch": 418.7787610619469,
      "grad_norm": 2.7619559972436036e-08,
      "learning_rate": 0.10511765086390841,
      "loss": 0.0,
      "num_input_tokens_seen": 13557056,
      "step": 23870
    },
    {
      "epoch": 418.86725663716817,
      "grad_norm": 9.347899343481458e-09,
      "learning_rate": 0.10506144816000816,
      "loss": 0.0,
      "num_input_tokens_seen": 13560880,
      "step": 23875
    },
    {
      "epoch": 418.95575221238937,
      "grad_norm": 1.8250934985530876e-08,
      "learning_rate": 0.10500525238619736,
      "loss": 0.0,
      "num_input_tokens_seen": 13563792,
      "step": 23880
    },
    {
      "epoch": 419.0353982300885,
      "grad_norm": 2.1587649001730824e-08,
      "learning_rate": 0.10494906355114209,
      "loss": 0.0,
      "num_input_tokens_seen": 13566536,
      "step": 23885
    },
    {
      "epoch": 419.12389380530976,
      "grad_norm": 7.017772585271587e-09,
      "learning_rate": 0.10489288166350737,
      "loss": 0.0,
      "num_input_tokens_seen": 13569256,
      "step": 23890
    },
    {
      "epoch": 419.21238938053096,
      "grad_norm": 1.6259392054962518e-08,
      "learning_rate": 0.10483670673195711,
      "loss": 0.0,
      "num_input_tokens_seen": 13572216,
      "step": 23895
    },
    {
      "epoch": 419.3008849557522,
      "grad_norm": 1.9554819985501126e-08,
      "learning_rate": 0.10478053876515431,
      "loss": 0.0,
      "num_input_tokens_seen": 13574984,
      "step": 23900
    },
    {
      "epoch": 419.3893805309734,
      "grad_norm": 1.6344701592174715e-08,
      "learning_rate": 0.10472437777176061,
      "loss": 0.0,
      "num_input_tokens_seen": 13578040,
      "step": 23905
    },
    {
      "epoch": 419.4778761061947,
      "grad_norm": 2.0534988820486433e-08,
      "learning_rate": 0.1046682237604369,
      "loss": 0.0,
      "num_input_tokens_seen": 13581080,
      "step": 23910
    },
    {
      "epoch": 419.56637168141594,
      "grad_norm": 1.0028012198404213e-08,
      "learning_rate": 0.1046120767398427,
      "loss": 0.0,
      "num_input_tokens_seen": 13584104,
      "step": 23915
    },
    {
      "epoch": 419.65486725663715,
      "grad_norm": 1.1960072932026833e-08,
      "learning_rate": 0.10455593671863667,
      "loss": 0.0,
      "num_input_tokens_seen": 13586840,
      "step": 23920
    },
    {
      "epoch": 419.7433628318584,
      "grad_norm": 7.756448816564898e-09,
      "learning_rate": 0.1044998037054763,
      "loss": 0.0,
      "num_input_tokens_seen": 13589448,
      "step": 23925
    },
    {
      "epoch": 419.83185840707966,
      "grad_norm": 1.4699782724392207e-08,
      "learning_rate": 0.10444367770901794,
      "loss": 0.0,
      "num_input_tokens_seen": 13592264,
      "step": 23930
    },
    {
      "epoch": 419.92035398230087,
      "grad_norm": 7.233229837311228e-08,
      "learning_rate": 0.10438755873791698,
      "loss": 0.0,
      "num_input_tokens_seen": 13595080,
      "step": 23935
    },
    {
      "epoch": 420.0,
      "grad_norm": 1.72510112861346e-08,
      "learning_rate": 0.10433144680082775,
      "loss": 0.0,
      "num_input_tokens_seen": 13597592,
      "step": 23940
    },
    {
      "epoch": 420.08849557522126,
      "grad_norm": 1.908260038874232e-08,
      "learning_rate": 0.10427534190640322,
      "loss": 0.0,
      "num_input_tokens_seen": 13600248,
      "step": 23945
    },
    {
      "epoch": 420.17699115044246,
      "grad_norm": 1.2434132834471257e-08,
      "learning_rate": 0.10421924406329568,
      "loss": 0.0,
      "num_input_tokens_seen": 13603144,
      "step": 23950
    },
    {
      "epoch": 420.2654867256637,
      "grad_norm": 2.1936871874572716e-08,
      "learning_rate": 0.10416315328015598,
      "loss": 0.0,
      "num_input_tokens_seen": 13606296,
      "step": 23955
    },
    {
      "epoch": 420.353982300885,
      "grad_norm": 1.3908443285970407e-08,
      "learning_rate": 0.10410706956563402,
      "loss": 0.0,
      "num_input_tokens_seen": 13609592,
      "step": 23960
    },
    {
      "epoch": 420.4424778761062,
      "grad_norm": 9.548700496964102e-09,
      "learning_rate": 0.10405099292837874,
      "loss": 0.0,
      "num_input_tokens_seen": 13612024,
      "step": 23965
    },
    {
      "epoch": 420.53097345132744,
      "grad_norm": 1.9694622821475605e-08,
      "learning_rate": 0.10399492337703771,
      "loss": 0.0,
      "num_input_tokens_seen": 13615128,
      "step": 23970
    },
    {
      "epoch": 420.6194690265487,
      "grad_norm": 1.770651181232097e-08,
      "learning_rate": 0.10393886092025764,
      "loss": 0.0,
      "num_input_tokens_seen": 13618024,
      "step": 23975
    },
    {
      "epoch": 420.7079646017699,
      "grad_norm": 1.7677342256661177e-08,
      "learning_rate": 0.10388280556668412,
      "loss": 0.0,
      "num_input_tokens_seen": 13621128,
      "step": 23980
    },
    {
      "epoch": 420.79646017699116,
      "grad_norm": 2.0323303928648784e-08,
      "learning_rate": 0.10382675732496145,
      "loss": 0.0,
      "num_input_tokens_seen": 13624168,
      "step": 23985
    },
    {
      "epoch": 420.88495575221236,
      "grad_norm": 1.8389787470596275e-08,
      "learning_rate": 0.10377071620373311,
      "loss": 0.0,
      "num_input_tokens_seen": 13626616,
      "step": 23990
    },
    {
      "epoch": 420.9734513274336,
      "grad_norm": 2.4333775883178532e-08,
      "learning_rate": 0.10371468221164128,
      "loss": 0.0,
      "num_input_tokens_seen": 13629272,
      "step": 23995
    },
    {
      "epoch": 421.05309734513276,
      "grad_norm": 1.5674833875323202e-08,
      "learning_rate": 0.10365865535732706,
      "loss": 0.0,
      "num_input_tokens_seen": 13631704,
      "step": 24000
    },
    {
      "epoch": 421.05309734513276,
      "eval_loss": 0.6828293204307556,
      "eval_runtime": 0.9413,
      "eval_samples_per_second": 26.56,
      "eval_steps_per_second": 13.811,
      "num_input_tokens_seen": 13631704,
      "step": 24000
    },
    {
      "epoch": 421.14159292035396,
      "grad_norm": 2.2386631215454145e-08,
      "learning_rate": 0.10360263564943062,
      "loss": 0.0,
      "num_input_tokens_seen": 13634808,
      "step": 24005
    },
    {
      "epoch": 421.2300884955752,
      "grad_norm": 3.4005953608584605e-08,
      "learning_rate": 0.10354662309659075,
      "loss": 0.0,
      "num_input_tokens_seen": 13637448,
      "step": 24010
    },
    {
      "epoch": 421.3185840707965,
      "grad_norm": 1.2718325947957965e-08,
      "learning_rate": 0.10349061770744537,
      "loss": 0.0,
      "num_input_tokens_seen": 13640520,
      "step": 24015
    },
    {
      "epoch": 421.4070796460177,
      "grad_norm": 2.510599017568893e-08,
      "learning_rate": 0.10343461949063128,
      "loss": 0.0,
      "num_input_tokens_seen": 13643064,
      "step": 24020
    },
    {
      "epoch": 421.49557522123894,
      "grad_norm": 4.1802589834105675e-08,
      "learning_rate": 0.103378628454784,
      "loss": 0.0,
      "num_input_tokens_seen": 13646296,
      "step": 24025
    },
    {
      "epoch": 421.5840707964602,
      "grad_norm": 2.321601222377012e-08,
      "learning_rate": 0.10332264460853811,
      "loss": 0.0,
      "num_input_tokens_seen": 13649128,
      "step": 24030
    },
    {
      "epoch": 421.6725663716814,
      "grad_norm": 2.8648244665419043e-08,
      "learning_rate": 0.10326666796052701,
      "loss": 0.0,
      "num_input_tokens_seen": 13651688,
      "step": 24035
    },
    {
      "epoch": 421.76106194690266,
      "grad_norm": 4.0814377655351564e-08,
      "learning_rate": 0.10321069851938296,
      "loss": 0.0,
      "num_input_tokens_seen": 13654744,
      "step": 24040
    },
    {
      "epoch": 421.8495575221239,
      "grad_norm": 2.784245545228714e-08,
      "learning_rate": 0.10315473629373724,
      "loss": 0.0,
      "num_input_tokens_seen": 13657544,
      "step": 24045
    },
    {
      "epoch": 421.9380530973451,
      "grad_norm": 9.91120518989419e-09,
      "learning_rate": 0.10309878129221982,
      "loss": 0.0,
      "num_input_tokens_seen": 13660504,
      "step": 24050
    },
    {
      "epoch": 422.01769911504425,
      "grad_norm": 2.085777595084437e-08,
      "learning_rate": 0.10304283352345973,
      "loss": 0.0,
      "num_input_tokens_seen": 13662744,
      "step": 24055
    },
    {
      "epoch": 422.1061946902655,
      "grad_norm": 2.5363235067743517e-08,
      "learning_rate": 0.10298689299608486,
      "loss": 0.0,
      "num_input_tokens_seen": 13665576,
      "step": 24060
    },
    {
      "epoch": 422.1946902654867,
      "grad_norm": 2.9616124663789378e-08,
      "learning_rate": 0.10293095971872188,
      "loss": 0.0,
      "num_input_tokens_seen": 13668056,
      "step": 24065
    },
    {
      "epoch": 422.283185840708,
      "grad_norm": 2.2449189174267303e-08,
      "learning_rate": 0.10287503369999645,
      "loss": 0.0,
      "num_input_tokens_seen": 13670872,
      "step": 24070
    },
    {
      "epoch": 422.37168141592923,
      "grad_norm": 8.507095472509718e-09,
      "learning_rate": 0.10281911494853295,
      "loss": 0.0,
      "num_input_tokens_seen": 13674632,
      "step": 24075
    },
    {
      "epoch": 422.46017699115043,
      "grad_norm": 7.160232406988598e-09,
      "learning_rate": 0.10276320347295485,
      "loss": 0.0,
      "num_input_tokens_seen": 13677272,
      "step": 24080
    },
    {
      "epoch": 422.5486725663717,
      "grad_norm": 2.52450487181477e-08,
      "learning_rate": 0.10270729928188446,
      "loss": 0.0,
      "num_input_tokens_seen": 13680264,
      "step": 24085
    },
    {
      "epoch": 422.6371681415929,
      "grad_norm": 1.808686711513019e-08,
      "learning_rate": 0.10265140238394276,
      "loss": 0.0,
      "num_input_tokens_seen": 13683432,
      "step": 24090
    },
    {
      "epoch": 422.72566371681415,
      "grad_norm": 5.898430632811369e-09,
      "learning_rate": 0.10259551278774988,
      "loss": 0.0,
      "num_input_tokens_seen": 13686168,
      "step": 24095
    },
    {
      "epoch": 422.8141592920354,
      "grad_norm": 1.644991698412923e-08,
      "learning_rate": 0.10253963050192462,
      "loss": 0.0,
      "num_input_tokens_seen": 13688984,
      "step": 24100
    },
    {
      "epoch": 422.9026548672566,
      "grad_norm": 1.63677302822407e-08,
      "learning_rate": 0.10248375553508478,
      "loss": 0.0,
      "num_input_tokens_seen": 13691880,
      "step": 24105
    },
    {
      "epoch": 422.9911504424779,
      "grad_norm": 8.617729641002825e-09,
      "learning_rate": 0.102427887895847,
      "loss": 0.0,
      "num_input_tokens_seen": 13694440,
      "step": 24110
    },
    {
      "epoch": 423.070796460177,
      "grad_norm": 9.974002068702248e-09,
      "learning_rate": 0.10237202759282668,
      "loss": 0.0,
      "num_input_tokens_seen": 13696576,
      "step": 24115
    },
    {
      "epoch": 423.1592920353982,
      "grad_norm": 2.6756598714428037e-08,
      "learning_rate": 0.10231617463463821,
      "loss": 0.0,
      "num_input_tokens_seen": 13699392,
      "step": 24120
    },
    {
      "epoch": 423.24778761061947,
      "grad_norm": 1.6031043159614455e-08,
      "learning_rate": 0.10226032902989492,
      "loss": 0.0,
      "num_input_tokens_seen": 13702160,
      "step": 24125
    },
    {
      "epoch": 423.3362831858407,
      "grad_norm": 1.575125452291104e-08,
      "learning_rate": 0.10220449078720877,
      "loss": 0.0,
      "num_input_tokens_seen": 13705200,
      "step": 24130
    },
    {
      "epoch": 423.42477876106193,
      "grad_norm": 1.8440307059108818e-08,
      "learning_rate": 0.1021486599151908,
      "loss": 0.0,
      "num_input_tokens_seen": 13707872,
      "step": 24135
    },
    {
      "epoch": 423.5132743362832,
      "grad_norm": 2.0640598563659296e-08,
      "learning_rate": 0.10209283642245084,
      "loss": 0.0,
      "num_input_tokens_seen": 13710896,
      "step": 24140
    },
    {
      "epoch": 423.60176991150445,
      "grad_norm": 1.3931455100646417e-08,
      "learning_rate": 0.10203702031759748,
      "loss": 0.0,
      "num_input_tokens_seen": 13713632,
      "step": 24145
    },
    {
      "epoch": 423.69026548672565,
      "grad_norm": 2.8616893743560468e-08,
      "learning_rate": 0.1019812116092384,
      "loss": 0.0,
      "num_input_tokens_seen": 13716848,
      "step": 24150
    },
    {
      "epoch": 423.7787610619469,
      "grad_norm": 1.528298909647674e-08,
      "learning_rate": 0.10192541030597986,
      "loss": 0.0,
      "num_input_tokens_seen": 13719632,
      "step": 24155
    },
    {
      "epoch": 423.86725663716817,
      "grad_norm": 1.8383865096893715e-08,
      "learning_rate": 0.1018696164164272,
      "loss": 0.0,
      "num_input_tokens_seen": 13722576,
      "step": 24160
    },
    {
      "epoch": 423.95575221238937,
      "grad_norm": 8.85377460235759e-09,
      "learning_rate": 0.10181382994918459,
      "loss": 0.0,
      "num_input_tokens_seen": 13725440,
      "step": 24165
    },
    {
      "epoch": 424.0353982300885,
      "grad_norm": 8.7520906077998e-09,
      "learning_rate": 0.10175805091285492,
      "loss": 0.0,
      "num_input_tokens_seen": 13728208,
      "step": 24170
    },
    {
      "epoch": 424.12389380530976,
      "grad_norm": 1.6841095629160918e-08,
      "learning_rate": 0.10170227931603999,
      "loss": 0.0,
      "num_input_tokens_seen": 13731456,
      "step": 24175
    },
    {
      "epoch": 424.21238938053096,
      "grad_norm": 7.774830557139012e-09,
      "learning_rate": 0.10164651516734062,
      "loss": 0.0,
      "num_input_tokens_seen": 13734144,
      "step": 24180
    },
    {
      "epoch": 424.3008849557522,
      "grad_norm": 1.4597838493557447e-08,
      "learning_rate": 0.1015907584753562,
      "loss": 0.0,
      "num_input_tokens_seen": 13736816,
      "step": 24185
    },
    {
      "epoch": 424.3893805309734,
      "grad_norm": 8.156400888026383e-09,
      "learning_rate": 0.10153500924868523,
      "loss": 0.0,
      "num_input_tokens_seen": 13739264,
      "step": 24190
    },
    {
      "epoch": 424.4778761061947,
      "grad_norm": 1.9641605675246865e-08,
      "learning_rate": 0.10147926749592483,
      "loss": 0.0,
      "num_input_tokens_seen": 13742176,
      "step": 24195
    },
    {
      "epoch": 424.56637168141594,
      "grad_norm": 2.3396735215897024e-08,
      "learning_rate": 0.10142353322567112,
      "loss": 0.0,
      "num_input_tokens_seen": 13745200,
      "step": 24200
    },
    {
      "epoch": 424.56637168141594,
      "eval_loss": 0.6715885400772095,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 26.717,
      "eval_steps_per_second": 13.893,
      "num_input_tokens_seen": 13745200,
      "step": 24200
    },
    {
      "epoch": 424.65486725663715,
      "grad_norm": 2.9654023236957983e-08,
      "learning_rate": 0.1013678064465191,
      "loss": 0.0,
      "num_input_tokens_seen": 13748144,
      "step": 24205
    },
    {
      "epoch": 424.7433628318584,
      "grad_norm": 1.979486263792296e-08,
      "learning_rate": 0.10131208716706244,
      "loss": 0.0,
      "num_input_tokens_seen": 13751504,
      "step": 24210
    },
    {
      "epoch": 424.83185840707966,
      "grad_norm": 2.0616676366103093e-08,
      "learning_rate": 0.10125637539589379,
      "loss": 0.0,
      "num_input_tokens_seen": 13754336,
      "step": 24215
    },
    {
      "epoch": 424.92035398230087,
      "grad_norm": 2.2947043376575493e-08,
      "learning_rate": 0.10120067114160464,
      "loss": 0.0,
      "num_input_tokens_seen": 13757184,
      "step": 24220
    },
    {
      "epoch": 425.0,
      "grad_norm": 5.1930246769416044e-08,
      "learning_rate": 0.10114497441278517,
      "loss": 0.0,
      "num_input_tokens_seen": 13759352,
      "step": 24225
    },
    {
      "epoch": 425.08849557522126,
      "grad_norm": 1.0561343799508904e-08,
      "learning_rate": 0.10108928521802468,
      "loss": 0.0,
      "num_input_tokens_seen": 13762296,
      "step": 24230
    },
    {
      "epoch": 425.17699115044246,
      "grad_norm": 1.9751928093114657e-08,
      "learning_rate": 0.101033603565911,
      "loss": 0.0,
      "num_input_tokens_seen": 13765320,
      "step": 24235
    },
    {
      "epoch": 425.2654867256637,
      "grad_norm": 2.025234913105578e-08,
      "learning_rate": 0.10097792946503102,
      "loss": 0.0,
      "num_input_tokens_seen": 13768184,
      "step": 24240
    },
    {
      "epoch": 425.353982300885,
      "grad_norm": 1.0818581586136133e-08,
      "learning_rate": 0.10092226292397039,
      "loss": 0.0,
      "num_input_tokens_seen": 13770888,
      "step": 24245
    },
    {
      "epoch": 425.4424778761062,
      "grad_norm": 1.6381303424850557e-08,
      "learning_rate": 0.10086660395131354,
      "loss": 0.0,
      "num_input_tokens_seen": 13773432,
      "step": 24250
    },
    {
      "epoch": 425.53097345132744,
      "grad_norm": 9.701216718838168e-09,
      "learning_rate": 0.10081095255564385,
      "loss": 0.0,
      "num_input_tokens_seen": 13776440,
      "step": 24255
    },
    {
      "epoch": 425.6194690265487,
      "grad_norm": 1.1961954093919758e-08,
      "learning_rate": 0.10075530874554335,
      "loss": 0.0,
      "num_input_tokens_seen": 13779320,
      "step": 24260
    },
    {
      "epoch": 425.7079646017699,
      "grad_norm": 1.575720531832303e-08,
      "learning_rate": 0.10069967252959311,
      "loss": 0.0,
      "num_input_tokens_seen": 13782168,
      "step": 24265
    },
    {
      "epoch": 425.79646017699116,
      "grad_norm": 3.116558033866568e-08,
      "learning_rate": 0.10064404391637297,
      "loss": 0.0,
      "num_input_tokens_seen": 13785288,
      "step": 24270
    },
    {
      "epoch": 425.88495575221236,
      "grad_norm": 2.3190386500004934e-08,
      "learning_rate": 0.10058842291446145,
      "loss": 0.0,
      "num_input_tokens_seen": 13788056,
      "step": 24275
    },
    {
      "epoch": 425.9734513274336,
      "grad_norm": 1.1791805754057805e-08,
      "learning_rate": 0.10053280953243608,
      "loss": 0.0,
      "num_input_tokens_seen": 13790920,
      "step": 24280
    },
    {
      "epoch": 426.05309734513276,
      "grad_norm": 9.672472600641413e-09,
      "learning_rate": 0.10047720377887315,
      "loss": 0.0,
      "num_input_tokens_seen": 13793464,
      "step": 24285
    },
    {
      "epoch": 426.14159292035396,
      "grad_norm": 1.589108045152443e-08,
      "learning_rate": 0.10042160566234767,
      "loss": 0.0,
      "num_input_tokens_seen": 13796424,
      "step": 24290
    },
    {
      "epoch": 426.2300884955752,
      "grad_norm": 2.7987011819163854e-08,
      "learning_rate": 0.10036601519143372,
      "loss": 0.0,
      "num_input_tokens_seen": 13799064,
      "step": 24295
    },
    {
      "epoch": 426.3185840707965,
      "grad_norm": 9.196863715033032e-09,
      "learning_rate": 0.1003104323747039,
      "loss": 0.0,
      "num_input_tokens_seen": 13802408,
      "step": 24300
    },
    {
      "epoch": 426.4070796460177,
      "grad_norm": 1.524469439573295e-08,
      "learning_rate": 0.10025485722072984,
      "loss": 0.0,
      "num_input_tokens_seen": 13804920,
      "step": 24305
    },
    {
      "epoch": 426.49557522123894,
      "grad_norm": 1.164058893721176e-08,
      "learning_rate": 0.10019928973808201,
      "loss": 0.0,
      "num_input_tokens_seen": 13808008,
      "step": 24310
    },
    {
      "epoch": 426.5840707964602,
      "grad_norm": 1.3659242625863044e-08,
      "learning_rate": 0.10014372993532945,
      "loss": 0.0,
      "num_input_tokens_seen": 13811160,
      "step": 24315
    },
    {
      "epoch": 426.6725663716814,
      "grad_norm": 9.726346839045164e-09,
      "learning_rate": 0.1000881778210403,
      "loss": 0.0,
      "num_input_tokens_seen": 13814472,
      "step": 24320
    },
    {
      "epoch": 426.76106194690266,
      "grad_norm": 1.7420003217694102e-08,
      "learning_rate": 0.10003263340378142,
      "loss": 0.0,
      "num_input_tokens_seen": 13817528,
      "step": 24325
    },
    {
      "epoch": 426.8495575221239,
      "grad_norm": 1.7506126326338745e-08,
      "learning_rate": 0.09997709669211834,
      "loss": 0.0,
      "num_input_tokens_seen": 13819960,
      "step": 24330
    },
    {
      "epoch": 426.9380530973451,
      "grad_norm": 1.804729699017571e-08,
      "learning_rate": 0.0999215676946156,
      "loss": 0.0,
      "num_input_tokens_seen": 13822712,
      "step": 24335
    },
    {
      "epoch": 427.01769911504425,
      "grad_norm": 1.9975336940092348e-08,
      "learning_rate": 0.0998660464198364,
      "loss": 0.0,
      "num_input_tokens_seen": 13825240,
      "step": 24340
    },
    {
      "epoch": 427.1061946902655,
      "grad_norm": 9.838304393383623e-09,
      "learning_rate": 0.09981053287634288,
      "loss": 0.0,
      "num_input_tokens_seen": 13827688,
      "step": 24345
    },
    {
      "epoch": 427.1946902654867,
      "grad_norm": 2.9348646179983007e-08,
      "learning_rate": 0.09975502707269596,
      "loss": 0.0,
      "num_input_tokens_seen": 13830840,
      "step": 24350
    },
    {
      "epoch": 427.283185840708,
      "grad_norm": 2.9994154715495824e-08,
      "learning_rate": 0.09969952901745524,
      "loss": 0.0,
      "num_input_tokens_seen": 13834264,
      "step": 24355
    },
    {
      "epoch": 427.37168141592923,
      "grad_norm": 1.6671299007953166e-08,
      "learning_rate": 0.09964403871917925,
      "loss": 0.0,
      "num_input_tokens_seen": 13837336,
      "step": 24360
    },
    {
      "epoch": 427.46017699115043,
      "grad_norm": 1.6614343678611476e-08,
      "learning_rate": 0.09958855618642536,
      "loss": 0.0,
      "num_input_tokens_seen": 13840168,
      "step": 24365
    },
    {
      "epoch": 427.5486725663717,
      "grad_norm": 5.596457963719104e-09,
      "learning_rate": 0.09953308142774955,
      "loss": 0.0,
      "num_input_tokens_seen": 13842760,
      "step": 24370
    },
    {
      "epoch": 427.6371681415929,
      "grad_norm": 1.4678498416742514e-08,
      "learning_rate": 0.09947761445170686,
      "loss": 0.0,
      "num_input_tokens_seen": 13845592,
      "step": 24375
    },
    {
      "epoch": 427.72566371681415,
      "grad_norm": 8.726074973708364e-09,
      "learning_rate": 0.09942215526685086,
      "loss": 0.0,
      "num_input_tokens_seen": 13848568,
      "step": 24380
    },
    {
      "epoch": 427.8141592920354,
      "grad_norm": 1.5947787090908605e-08,
      "learning_rate": 0.09936670388173414,
      "loss": 0.0,
      "num_input_tokens_seen": 13851464,
      "step": 24385
    },
    {
      "epoch": 427.9026548672566,
      "grad_norm": 5.098774291667496e-09,
      "learning_rate": 0.09931126030490799,
      "loss": 0.0,
      "num_input_tokens_seen": 13854744,
      "step": 24390
    },
    {
      "epoch": 427.9911504424779,
      "grad_norm": 6.868448032548713e-09,
      "learning_rate": 0.0992558245449225,
      "loss": 0.0,
      "num_input_tokens_seen": 13857256,
      "step": 24395
    },
    {
      "epoch": 428.070796460177,
      "grad_norm": 1.3744265281445678e-08,
      "learning_rate": 0.09920039661032651,
      "loss": 0.0,
      "num_input_tokens_seen": 13859752,
      "step": 24400
    },
    {
      "epoch": 428.070796460177,
      "eval_loss": 0.6813579201698303,
      "eval_runtime": 0.9258,
      "eval_samples_per_second": 27.004,
      "eval_steps_per_second": 14.042,
      "num_input_tokens_seen": 13859752,
      "step": 24400
    },
    {
      "epoch": 428.1592920353982,
      "grad_norm": 2.3601018028784893e-08,
      "learning_rate": 0.09914497650966782,
      "loss": 0.0,
      "num_input_tokens_seen": 13863672,
      "step": 24405
    },
    {
      "epoch": 428.24778761061947,
      "grad_norm": 1.54743862168516e-08,
      "learning_rate": 0.09908956425149276,
      "loss": 0.0,
      "num_input_tokens_seen": 13866632,
      "step": 24410
    },
    {
      "epoch": 428.3362831858407,
      "grad_norm": 2.54750087691491e-08,
      "learning_rate": 0.09903415984434677,
      "loss": 0.0,
      "num_input_tokens_seen": 13869592,
      "step": 24415
    },
    {
      "epoch": 428.42477876106193,
      "grad_norm": 2.4714006841008995e-08,
      "learning_rate": 0.09897876329677373,
      "loss": 0.0,
      "num_input_tokens_seen": 13872472,
      "step": 24420
    },
    {
      "epoch": 428.5132743362832,
      "grad_norm": 2.1826659590828967e-08,
      "learning_rate": 0.09892337461731658,
      "loss": 0.0,
      "num_input_tokens_seen": 13875032,
      "step": 24425
    },
    {
      "epoch": 428.60176991150445,
      "grad_norm": 1.160288309876023e-08,
      "learning_rate": 0.09886799381451693,
      "loss": 0.0,
      "num_input_tokens_seen": 13877624,
      "step": 24430
    },
    {
      "epoch": 428.69026548672565,
      "grad_norm": 5.234942257459352e-09,
      "learning_rate": 0.09881262089691521,
      "loss": 0.0,
      "num_input_tokens_seen": 13880056,
      "step": 24435
    },
    {
      "epoch": 428.7787610619469,
      "grad_norm": 1.8167332527241342e-08,
      "learning_rate": 0.09875725587305059,
      "loss": 0.0,
      "num_input_tokens_seen": 13883192,
      "step": 24440
    },
    {
      "epoch": 428.86725663716817,
      "grad_norm": 1.7706662802652318e-08,
      "learning_rate": 0.09870189875146111,
      "loss": 0.0,
      "num_input_tokens_seen": 13885720,
      "step": 24445
    },
    {
      "epoch": 428.95575221238937,
      "grad_norm": 7.57153539865385e-09,
      "learning_rate": 0.09864654954068346,
      "loss": 0.0,
      "num_input_tokens_seen": 13888776,
      "step": 24450
    },
    {
      "epoch": 429.0353982300885,
      "grad_norm": 4.4220303863085064e-08,
      "learning_rate": 0.09859120824925326,
      "loss": 0.0,
      "num_input_tokens_seen": 13891224,
      "step": 24455
    },
    {
      "epoch": 429.12389380530976,
      "grad_norm": 1.4896378353057571e-08,
      "learning_rate": 0.09853587488570474,
      "loss": 0.0,
      "num_input_tokens_seen": 13894136,
      "step": 24460
    },
    {
      "epoch": 429.21238938053096,
      "grad_norm": 1.0092270130712677e-08,
      "learning_rate": 0.09848054945857107,
      "loss": 0.0,
      "num_input_tokens_seen": 13896744,
      "step": 24465
    },
    {
      "epoch": 429.3008849557522,
      "grad_norm": 1.0971993091857257e-08,
      "learning_rate": 0.09842523197638416,
      "loss": 0.0,
      "num_input_tokens_seen": 13899512,
      "step": 24470
    },
    {
      "epoch": 429.3893805309734,
      "grad_norm": 1.292303153377361e-08,
      "learning_rate": 0.09836992244767452,
      "loss": 0.0,
      "num_input_tokens_seen": 13902616,
      "step": 24475
    },
    {
      "epoch": 429.4778761061947,
      "grad_norm": 1.2649548075671646e-08,
      "learning_rate": 0.09831462088097168,
      "loss": 0.0,
      "num_input_tokens_seen": 13905256,
      "step": 24480
    },
    {
      "epoch": 429.56637168141594,
      "grad_norm": 1.4461504882490317e-08,
      "learning_rate": 0.09825932728480385,
      "loss": 0.0,
      "num_input_tokens_seen": 13908536,
      "step": 24485
    },
    {
      "epoch": 429.65486725663715,
      "grad_norm": 1.2747034539017932e-08,
      "learning_rate": 0.09820404166769794,
      "loss": 0.0,
      "num_input_tokens_seen": 13911384,
      "step": 24490
    },
    {
      "epoch": 429.7433628318584,
      "grad_norm": 2.8783643912788648e-08,
      "learning_rate": 0.09814876403817978,
      "loss": 0.0,
      "num_input_tokens_seen": 13914520,
      "step": 24495
    },
    {
      "epoch": 429.83185840707966,
      "grad_norm": 2.8067038471135675e-08,
      "learning_rate": 0.09809349440477376,
      "loss": 0.0,
      "num_input_tokens_seen": 13917448,
      "step": 24500
    },
    {
      "epoch": 429.92035398230087,
      "grad_norm": 1.9494818204179865e-08,
      "learning_rate": 0.09803823277600317,
      "loss": 0.0,
      "num_input_tokens_seen": 13920136,
      "step": 24505
    },
    {
      "epoch": 430.0,
      "grad_norm": 2.657998088295699e-08,
      "learning_rate": 0.09798297916039014,
      "loss": 0.0,
      "num_input_tokens_seen": 13922320,
      "step": 24510
    },
    {
      "epoch": 430.08849557522126,
      "grad_norm": 1.0516676418603765e-08,
      "learning_rate": 0.09792773356645534,
      "loss": 0.0,
      "num_input_tokens_seen": 13925120,
      "step": 24515
    },
    {
      "epoch": 430.17699115044246,
      "grad_norm": 1.2195171095186197e-08,
      "learning_rate": 0.09787249600271843,
      "loss": 0.0,
      "num_input_tokens_seen": 13928400,
      "step": 24520
    },
    {
      "epoch": 430.2654867256637,
      "grad_norm": 8.019510389090101e-09,
      "learning_rate": 0.09781726647769776,
      "loss": 0.0,
      "num_input_tokens_seen": 13930816,
      "step": 24525
    },
    {
      "epoch": 430.353982300885,
      "grad_norm": 6.884379732952084e-09,
      "learning_rate": 0.0977620449999103,
      "loss": 0.0,
      "num_input_tokens_seen": 13933712,
      "step": 24530
    },
    {
      "epoch": 430.4424778761062,
      "grad_norm": 4.826856692119463e-09,
      "learning_rate": 0.09770683157787204,
      "loss": 0.0,
      "num_input_tokens_seen": 13936576,
      "step": 24535
    },
    {
      "epoch": 430.53097345132744,
      "grad_norm": 1.5148563292655126e-08,
      "learning_rate": 0.09765162622009745,
      "loss": 0.0,
      "num_input_tokens_seen": 13939824,
      "step": 24540
    },
    {
      "epoch": 430.6194690265487,
      "grad_norm": 2.2606663208080136e-08,
      "learning_rate": 0.09759642893509995,
      "loss": 0.0,
      "num_input_tokens_seen": 13942752,
      "step": 24545
    },
    {
      "epoch": 430.7079646017699,
      "grad_norm": 2.077787009113763e-08,
      "learning_rate": 0.09754123973139169,
      "loss": 0.0,
      "num_input_tokens_seen": 13945504,
      "step": 24550
    },
    {
      "epoch": 430.79646017699116,
      "grad_norm": 1.6844516892433603e-08,
      "learning_rate": 0.09748605861748345,
      "loss": 0.0,
      "num_input_tokens_seen": 13948240,
      "step": 24555
    },
    {
      "epoch": 430.88495575221236,
      "grad_norm": 1.5362864758117212e-08,
      "learning_rate": 0.0974308856018849,
      "loss": 0.0,
      "num_input_tokens_seen": 13951040,
      "step": 24560
    },
    {
      "epoch": 430.9734513274336,
      "grad_norm": 7.404143076428227e-09,
      "learning_rate": 0.09737572069310449,
      "loss": 0.0,
      "num_input_tokens_seen": 13953808,
      "step": 24565
    },
    {
      "epoch": 431.05309734513276,
      "grad_norm": 1.840582442014238e-08,
      "learning_rate": 0.09732056389964922,
      "loss": 0.0,
      "num_input_tokens_seen": 13956216,
      "step": 24570
    },
    {
      "epoch": 431.14159292035396,
      "grad_norm": 1.0124649563181265e-08,
      "learning_rate": 0.097265415230025,
      "loss": 0.0,
      "num_input_tokens_seen": 13958872,
      "step": 24575
    },
    {
      "epoch": 431.2300884955752,
      "grad_norm": 1.3716680236086631e-08,
      "learning_rate": 0.09721027469273648,
      "loss": 0.0,
      "num_input_tokens_seen": 13961848,
      "step": 24580
    },
    {
      "epoch": 431.3185840707965,
      "grad_norm": 1.2224049328324327e-08,
      "learning_rate": 0.09715514229628695,
      "loss": 0.0,
      "num_input_tokens_seen": 13964632,
      "step": 24585
    },
    {
      "epoch": 431.4070796460177,
      "grad_norm": 1.6347108555692103e-08,
      "learning_rate": 0.09710001804917864,
      "loss": 0.0,
      "num_input_tokens_seen": 13967048,
      "step": 24590
    },
    {
      "epoch": 431.49557522123894,
      "grad_norm": 1.797405069225988e-08,
      "learning_rate": 0.09704490195991226,
      "loss": 0.0,
      "num_input_tokens_seen": 13969480,
      "step": 24595
    },
    {
      "epoch": 431.5840707964602,
      "grad_norm": 1.2947486638381633e-08,
      "learning_rate": 0.09698979403698753,
      "loss": 0.0,
      "num_input_tokens_seen": 13972648,
      "step": 24600
    },
    {
      "epoch": 431.5840707964602,
      "eval_loss": 0.7037013173103333,
      "eval_runtime": 0.9251,
      "eval_samples_per_second": 27.025,
      "eval_steps_per_second": 14.053,
      "num_input_tokens_seen": 13972648,
      "step": 24600
    },
    {
      "epoch": 431.6725663716814,
      "grad_norm": 1.2729199916350353e-08,
      "learning_rate": 0.0969346942889027,
      "loss": 0.0,
      "num_input_tokens_seen": 13975832,
      "step": 24605
    },
    {
      "epoch": 431.76106194690266,
      "grad_norm": 2.8263142937134944e-08,
      "learning_rate": 0.09687960272415487,
      "loss": 0.0,
      "num_input_tokens_seen": 13979112,
      "step": 24610
    },
    {
      "epoch": 431.8495575221239,
      "grad_norm": 2.130036058645146e-08,
      "learning_rate": 0.0968245193512399,
      "loss": 0.0,
      "num_input_tokens_seen": 13981960,
      "step": 24615
    },
    {
      "epoch": 431.9380530973451,
      "grad_norm": 3.26237987735567e-08,
      "learning_rate": 0.09676944417865221,
      "loss": 0.0,
      "num_input_tokens_seen": 13984536,
      "step": 24620
    },
    {
      "epoch": 432.01769911504425,
      "grad_norm": 1.6969975646929925e-08,
      "learning_rate": 0.09671437721488517,
      "loss": 0.0,
      "num_input_tokens_seen": 13986872,
      "step": 24625
    },
    {
      "epoch": 432.1061946902655,
      "grad_norm": 1.1209075445606231e-08,
      "learning_rate": 0.09665931846843086,
      "loss": 0.0,
      "num_input_tokens_seen": 13989352,
      "step": 24630
    },
    {
      "epoch": 432.1946902654867,
      "grad_norm": 2.604737581179961e-08,
      "learning_rate": 0.0966042679477799,
      "loss": 0.0,
      "num_input_tokens_seen": 13992280,
      "step": 24635
    },
    {
      "epoch": 432.283185840708,
      "grad_norm": 9.309838233662049e-09,
      "learning_rate": 0.09654922566142186,
      "loss": 0.0,
      "num_input_tokens_seen": 13995256,
      "step": 24640
    },
    {
      "epoch": 432.37168141592923,
      "grad_norm": 2.3717360519981412e-08,
      "learning_rate": 0.09649419161784498,
      "loss": 0.0,
      "num_input_tokens_seen": 13997992,
      "step": 24645
    },
    {
      "epoch": 432.46017699115043,
      "grad_norm": 2.1024375129741202e-08,
      "learning_rate": 0.09643916582553606,
      "loss": 0.0,
      "num_input_tokens_seen": 14000792,
      "step": 24650
    },
    {
      "epoch": 432.5486725663717,
      "grad_norm": 5.286521798097965e-08,
      "learning_rate": 0.09638414829298093,
      "loss": 0.0,
      "num_input_tokens_seen": 14003352,
      "step": 24655
    },
    {
      "epoch": 432.6371681415929,
      "grad_norm": 1.8994064987509773e-08,
      "learning_rate": 0.09632913902866386,
      "loss": 0.0,
      "num_input_tokens_seen": 14007016,
      "step": 24660
    },
    {
      "epoch": 432.72566371681415,
      "grad_norm": 1.5377258577586872e-08,
      "learning_rate": 0.096274138041068,
      "loss": 0.0,
      "num_input_tokens_seen": 14009928,
      "step": 24665
    },
    {
      "epoch": 432.8141592920354,
      "grad_norm": 1.2973567109497708e-08,
      "learning_rate": 0.09621914533867527,
      "loss": 0.0,
      "num_input_tokens_seen": 14012792,
      "step": 24670
    },
    {
      "epoch": 432.9026548672566,
      "grad_norm": 1.608490229898507e-08,
      "learning_rate": 0.09616416092996616,
      "loss": 0.0,
      "num_input_tokens_seen": 14015400,
      "step": 24675
    },
    {
      "epoch": 432.9911504424779,
      "grad_norm": 2.2943297040001198e-08,
      "learning_rate": 0.09610918482342,
      "loss": 0.0,
      "num_input_tokens_seen": 14018920,
      "step": 24680
    },
    {
      "epoch": 433.070796460177,
      "grad_norm": 6.555514797668138e-09,
      "learning_rate": 0.09605421702751478,
      "loss": 0.0,
      "num_input_tokens_seen": 14021720,
      "step": 24685
    },
    {
      "epoch": 433.1592920353982,
      "grad_norm": 1.609571320670966e-08,
      "learning_rate": 0.09599925755072718,
      "loss": 0.0,
      "num_input_tokens_seen": 14025016,
      "step": 24690
    },
    {
      "epoch": 433.24778761061947,
      "grad_norm": 2.2465760807222068e-08,
      "learning_rate": 0.09594430640153273,
      "loss": 0.0,
      "num_input_tokens_seen": 14028056,
      "step": 24695
    },
    {
      "epoch": 433.3362831858407,
      "grad_norm": 1.955738326842038e-08,
      "learning_rate": 0.09588936358840547,
      "loss": 0.0,
      "num_input_tokens_seen": 14030888,
      "step": 24700
    },
    {
      "epoch": 433.42477876106193,
      "grad_norm": 9.37789934596367e-09,
      "learning_rate": 0.09583442911981836,
      "loss": 0.0,
      "num_input_tokens_seen": 14033480,
      "step": 24705
    },
    {
      "epoch": 433.5132743362832,
      "grad_norm": 1.684567862980657e-08,
      "learning_rate": 0.09577950300424302,
      "loss": 0.0,
      "num_input_tokens_seen": 14036648,
      "step": 24710
    },
    {
      "epoch": 433.60176991150445,
      "grad_norm": 2.5509919510113832e-08,
      "learning_rate": 0.09572458525014967,
      "loss": 0.0,
      "num_input_tokens_seen": 14039096,
      "step": 24715
    },
    {
      "epoch": 433.69026548672565,
      "grad_norm": 2.5883581500352193e-08,
      "learning_rate": 0.0956696758660073,
      "loss": 0.0,
      "num_input_tokens_seen": 14042024,
      "step": 24720
    },
    {
      "epoch": 433.7787610619469,
      "grad_norm": 2.49113707440074e-08,
      "learning_rate": 0.09561477486028373,
      "loss": 0.0,
      "num_input_tokens_seen": 14045032,
      "step": 24725
    },
    {
      "epoch": 433.86725663716817,
      "grad_norm": 4.982338541736908e-09,
      "learning_rate": 0.09555988224144528,
      "loss": 0.0,
      "num_input_tokens_seen": 14047384,
      "step": 24730
    },
    {
      "epoch": 433.95575221238937,
      "grad_norm": 1.3694565481614518e-08,
      "learning_rate": 0.09550499801795717,
      "loss": 0.0,
      "num_input_tokens_seen": 14050136,
      "step": 24735
    },
    {
      "epoch": 434.0353982300885,
      "grad_norm": 1.4087083499703112e-08,
      "learning_rate": 0.09545012219828314,
      "loss": 0.0,
      "num_input_tokens_seen": 14052824,
      "step": 24740
    },
    {
      "epoch": 434.12389380530976,
      "grad_norm": 1.2442194829986875e-08,
      "learning_rate": 0.09539525479088577,
      "loss": 0.0,
      "num_input_tokens_seen": 14055816,
      "step": 24745
    },
    {
      "epoch": 434.21238938053096,
      "grad_norm": 2.2323366266618905e-08,
      "learning_rate": 0.0953403958042264,
      "loss": 0.0,
      "num_input_tokens_seen": 14058712,
      "step": 24750
    },
    {
      "epoch": 434.3008849557522,
      "grad_norm": 1.1496656959764096e-08,
      "learning_rate": 0.09528554524676484,
      "loss": 0.0,
      "num_input_tokens_seen": 14061176,
      "step": 24755
    },
    {
      "epoch": 434.3893805309734,
      "grad_norm": 1.7933215801235747e-08,
      "learning_rate": 0.09523070312695978,
      "loss": 0.0,
      "num_input_tokens_seen": 14064056,
      "step": 24760
    },
    {
      "epoch": 434.4778761061947,
      "grad_norm": 2.0176425863382974e-08,
      "learning_rate": 0.09517586945326863,
      "loss": 0.0,
      "num_input_tokens_seen": 14067096,
      "step": 24765
    },
    {
      "epoch": 434.56637168141594,
      "grad_norm": 6.98669344600944e-09,
      "learning_rate": 0.0951210442341473,
      "loss": 0.0,
      "num_input_tokens_seen": 14070264,
      "step": 24770
    },
    {
      "epoch": 434.65486725663715,
      "grad_norm": 2.0521886412439017e-08,
      "learning_rate": 0.09506622747805066,
      "loss": 0.0,
      "num_input_tokens_seen": 14072728,
      "step": 24775
    },
    {
      "epoch": 434.7433628318584,
      "grad_norm": 1.3090804884541285e-08,
      "learning_rate": 0.09501141919343203,
      "loss": 0.0,
      "num_input_tokens_seen": 14075464,
      "step": 24780
    },
    {
      "epoch": 434.83185840707966,
      "grad_norm": 1.0282419360407857e-08,
      "learning_rate": 0.09495661938874361,
      "loss": 0.0,
      "num_input_tokens_seen": 14078776,
      "step": 24785
    },
    {
      "epoch": 434.92035398230087,
      "grad_norm": 2.406540922095246e-08,
      "learning_rate": 0.0949018280724362,
      "loss": 0.0,
      "num_input_tokens_seen": 14081576,
      "step": 24790
    },
    {
      "epoch": 435.0,
      "grad_norm": 3.7556864285193114e-09,
      "learning_rate": 0.09484704525295934,
      "loss": 0.0,
      "num_input_tokens_seen": 14083896,
      "step": 24795
    },
    {
      "epoch": 435.08849557522126,
      "grad_norm": 1.189509646337683e-08,
      "learning_rate": 0.09479227093876112,
      "loss": 0.0,
      "num_input_tokens_seen": 14086360,
      "step": 24800
    },
    {
      "epoch": 435.08849557522126,
      "eval_loss": 0.7048231363296509,
      "eval_runtime": 0.9385,
      "eval_samples_per_second": 26.639,
      "eval_steps_per_second": 13.852,
      "num_input_tokens_seen": 14086360,
      "step": 24800
    },
    {
      "epoch": 435.17699115044246,
      "grad_norm": 1.5620324589349366e-08,
      "learning_rate": 0.0947375051382886,
      "loss": 0.0,
      "num_input_tokens_seen": 14089400,
      "step": 24805
    },
    {
      "epoch": 435.2654867256637,
      "grad_norm": 4.3884018197104524e-08,
      "learning_rate": 0.09468274785998718,
      "loss": 0.0,
      "num_input_tokens_seen": 14092600,
      "step": 24810
    },
    {
      "epoch": 435.353982300885,
      "grad_norm": 1.5795084351566402e-08,
      "learning_rate": 0.09462799911230127,
      "loss": 0.0,
      "num_input_tokens_seen": 14095240,
      "step": 24815
    },
    {
      "epoch": 435.4424778761062,
      "grad_norm": 1.5666474340036984e-08,
      "learning_rate": 0.0945732589036737,
      "loss": 0.0,
      "num_input_tokens_seen": 14098200,
      "step": 24820
    },
    {
      "epoch": 435.53097345132744,
      "grad_norm": 8.971531961776691e-09,
      "learning_rate": 0.09451852724254614,
      "loss": 0.0,
      "num_input_tokens_seen": 14101480,
      "step": 24825
    },
    {
      "epoch": 435.6194690265487,
      "grad_norm": 7.4403372352094266e-09,
      "learning_rate": 0.09446380413735894,
      "loss": 0.0,
      "num_input_tokens_seen": 14103960,
      "step": 24830
    },
    {
      "epoch": 435.7079646017699,
      "grad_norm": 1.036785945984775e-08,
      "learning_rate": 0.09440908959655099,
      "loss": 0.0,
      "num_input_tokens_seen": 14107048,
      "step": 24835
    },
    {
      "epoch": 435.79646017699116,
      "grad_norm": 1.9933104056235607e-08,
      "learning_rate": 0.09435438362856004,
      "loss": 0.0,
      "num_input_tokens_seen": 14110008,
      "step": 24840
    },
    {
      "epoch": 435.88495575221236,
      "grad_norm": 8.240979454399167e-09,
      "learning_rate": 0.0942996862418225,
      "loss": 0.0,
      "num_input_tokens_seen": 14113000,
      "step": 24845
    },
    {
      "epoch": 435.9734513274336,
      "grad_norm": 1.6430469429451477e-08,
      "learning_rate": 0.09424499744477322,
      "loss": 0.0,
      "num_input_tokens_seen": 14115752,
      "step": 24850
    },
    {
      "epoch": 436.05309734513276,
      "grad_norm": 1.2888797584764689e-08,
      "learning_rate": 0.09419031724584608,
      "loss": 0.0,
      "num_input_tokens_seen": 14118128,
      "step": 24855
    },
    {
      "epoch": 436.14159292035396,
      "grad_norm": 3.079478361200927e-08,
      "learning_rate": 0.09413564565347331,
      "loss": 0.0,
      "num_input_tokens_seen": 14121072,
      "step": 24860
    },
    {
      "epoch": 436.2300884955752,
      "grad_norm": 1.0627303481669514e-08,
      "learning_rate": 0.094080982676086,
      "loss": 0.0,
      "num_input_tokens_seen": 14124304,
      "step": 24865
    },
    {
      "epoch": 436.3185840707965,
      "grad_norm": 2.1888443058060147e-08,
      "learning_rate": 0.09402632832211395,
      "loss": 0.0,
      "num_input_tokens_seen": 14127056,
      "step": 24870
    },
    {
      "epoch": 436.4070796460177,
      "grad_norm": 1.6213340003901067e-08,
      "learning_rate": 0.09397168259998541,
      "loss": 0.0,
      "num_input_tokens_seen": 14129776,
      "step": 24875
    },
    {
      "epoch": 436.49557522123894,
      "grad_norm": 2.2197815141566934e-08,
      "learning_rate": 0.09391704551812759,
      "loss": 0.0,
      "num_input_tokens_seen": 14132400,
      "step": 24880
    },
    {
      "epoch": 436.5840707964602,
      "grad_norm": 1.6259660284845268e-08,
      "learning_rate": 0.09386241708496605,
      "loss": 0.0,
      "num_input_tokens_seen": 14135280,
      "step": 24885
    },
    {
      "epoch": 436.6725663716814,
      "grad_norm": 3.5416672261590065e-08,
      "learning_rate": 0.09380779730892527,
      "loss": 0.0,
      "num_input_tokens_seen": 14137776,
      "step": 24890
    },
    {
      "epoch": 436.76106194690266,
      "grad_norm": 8.90942697395758e-09,
      "learning_rate": 0.09375318619842836,
      "loss": 0.0,
      "num_input_tokens_seen": 14140768,
      "step": 24895
    },
    {
      "epoch": 436.8495575221239,
      "grad_norm": 1.7470728863600016e-08,
      "learning_rate": 0.09369858376189696,
      "loss": 0.0,
      "num_input_tokens_seen": 14143744,
      "step": 24900
    },
    {
      "epoch": 436.9380530973451,
      "grad_norm": 9.069882622725345e-09,
      "learning_rate": 0.09364399000775143,
      "loss": 0.0,
      "num_input_tokens_seen": 14146448,
      "step": 24905
    },
    {
      "epoch": 437.01769911504425,
      "grad_norm": 1.2190141340795435e-08,
      "learning_rate": 0.09358940494441093,
      "loss": 0.0,
      "num_input_tokens_seen": 14149152,
      "step": 24910
    },
    {
      "epoch": 437.1061946902655,
      "grad_norm": 7.442679805791386e-09,
      "learning_rate": 0.09353482858029301,
      "loss": 0.0,
      "num_input_tokens_seen": 14151952,
      "step": 24915
    },
    {
      "epoch": 437.1946902654867,
      "grad_norm": 1.4810519921582e-08,
      "learning_rate": 0.09348026092381419,
      "loss": 0.0,
      "num_input_tokens_seen": 14154784,
      "step": 24920
    },
    {
      "epoch": 437.283185840708,
      "grad_norm": 1.9193564071429137e-08,
      "learning_rate": 0.09342570198338931,
      "loss": 0.0,
      "num_input_tokens_seen": 14157872,
      "step": 24925
    },
    {
      "epoch": 437.37168141592923,
      "grad_norm": 1.7272729024853106e-08,
      "learning_rate": 0.0933711517674322,
      "loss": 0.0,
      "num_input_tokens_seen": 14161040,
      "step": 24930
    },
    {
      "epoch": 437.46017699115043,
      "grad_norm": 1.0443019782258034e-08,
      "learning_rate": 0.09331661028435513,
      "loss": 0.0,
      "num_input_tokens_seen": 14163872,
      "step": 24935
    },
    {
      "epoch": 437.5486725663717,
      "grad_norm": 2.437110246944485e-08,
      "learning_rate": 0.09326207754256909,
      "loss": 0.0,
      "num_input_tokens_seen": 14167152,
      "step": 24940
    },
    {
      "epoch": 437.6371681415929,
      "grad_norm": 1.1533236587979445e-08,
      "learning_rate": 0.09320755355048366,
      "loss": 0.0,
      "num_input_tokens_seen": 14169888,
      "step": 24945
    },
    {
      "epoch": 437.72566371681415,
      "grad_norm": 2.420979683392943e-08,
      "learning_rate": 0.09315303831650722,
      "loss": 0.0,
      "num_input_tokens_seen": 14172992,
      "step": 24950
    },
    {
      "epoch": 437.8141592920354,
      "grad_norm": 2.9012666047378843e-08,
      "learning_rate": 0.09309853184904661,
      "loss": 0.0,
      "num_input_tokens_seen": 14175568,
      "step": 24955
    },
    {
      "epoch": 437.9026548672566,
      "grad_norm": 8.08491140702472e-09,
      "learning_rate": 0.09304403415650753,
      "loss": 0.0,
      "num_input_tokens_seen": 14178144,
      "step": 24960
    },
    {
      "epoch": 437.9911504424779,
      "grad_norm": 3.081795085790873e-08,
      "learning_rate": 0.09298954524729405,
      "loss": 0.0,
      "num_input_tokens_seen": 14181104,
      "step": 24965
    },
    {
      "epoch": 438.070796460177,
      "grad_norm": 9.074105911111019e-09,
      "learning_rate": 0.09293506512980916,
      "loss": 0.0,
      "num_input_tokens_seen": 14183192,
      "step": 24970
    },
    {
      "epoch": 438.1592920353982,
      "grad_norm": 9.462334915610882e-09,
      "learning_rate": 0.0928805938124544,
      "loss": 0.0,
      "num_input_tokens_seen": 14185976,
      "step": 24975
    },
    {
      "epoch": 438.24778761061947,
      "grad_norm": 1.4073005871750865e-08,
      "learning_rate": 0.09282613130362982,
      "loss": 0.0,
      "num_input_tokens_seen": 14189144,
      "step": 24980
    },
    {
      "epoch": 438.3362831858407,
      "grad_norm": 2.7817570469323982e-08,
      "learning_rate": 0.09277167761173427,
      "loss": 0.0,
      "num_input_tokens_seen": 14192472,
      "step": 24985
    },
    {
      "epoch": 438.42477876106193,
      "grad_norm": 1.8608405483178103e-08,
      "learning_rate": 0.0927172327451653,
      "loss": 0.0,
      "num_input_tokens_seen": 14195560,
      "step": 24990
    },
    {
      "epoch": 438.5132743362832,
      "grad_norm": 1.631699042548007e-08,
      "learning_rate": 0.09266279671231882,
      "loss": 0.0,
      "num_input_tokens_seen": 14198760,
      "step": 24995
    },
    {
      "epoch": 438.60176991150445,
      "grad_norm": 1.515447500821665e-08,
      "learning_rate": 0.09260836952158967,
      "loss": 0.0,
      "num_input_tokens_seen": 14201656,
      "step": 25000
    },
    {
      "epoch": 438.60176991150445,
      "eval_loss": 0.7057361602783203,
      "eval_runtime": 0.943,
      "eval_samples_per_second": 26.512,
      "eval_steps_per_second": 13.786,
      "num_input_tokens_seen": 14201656,
      "step": 25000
    },
    {
      "epoch": 438.69026548672565,
      "grad_norm": 3.51023281552898e-08,
      "learning_rate": 0.09255395118137114,
      "loss": 0.0,
      "num_input_tokens_seen": 14204424,
      "step": 25005
    },
    {
      "epoch": 438.7787610619469,
      "grad_norm": 1.38833442520081e-08,
      "learning_rate": 0.09249954170005527,
      "loss": 0.0,
      "num_input_tokens_seen": 14207144,
      "step": 25010
    },
    {
      "epoch": 438.86725663716817,
      "grad_norm": 1.6037652983413864e-08,
      "learning_rate": 0.0924451410860327,
      "loss": 0.0,
      "num_input_tokens_seen": 14209656,
      "step": 25015
    },
    {
      "epoch": 438.95575221238937,
      "grad_norm": 1.7365231030908035e-08,
      "learning_rate": 0.09239074934769258,
      "loss": 0.0,
      "num_input_tokens_seen": 14212424,
      "step": 25020
    },
    {
      "epoch": 439.0353982300885,
      "grad_norm": 1.9683875862597233e-08,
      "learning_rate": 0.09233636649342288,
      "loss": 0.0,
      "num_input_tokens_seen": 14214888,
      "step": 25025
    },
    {
      "epoch": 439.12389380530976,
      "grad_norm": 4.296921574109547e-08,
      "learning_rate": 0.09228199253161017,
      "loss": 0.0,
      "num_input_tokens_seen": 14217768,
      "step": 25030
    },
    {
      "epoch": 439.21238938053096,
      "grad_norm": 2.348705230303949e-08,
      "learning_rate": 0.09222762747063949,
      "loss": 0.0,
      "num_input_tokens_seen": 14220328,
      "step": 25035
    },
    {
      "epoch": 439.3008849557522,
      "grad_norm": 1.1002473598864526e-08,
      "learning_rate": 0.09217327131889473,
      "loss": 0.0,
      "num_input_tokens_seen": 14222840,
      "step": 25040
    },
    {
      "epoch": 439.3893805309734,
      "grad_norm": 1.623202194878104e-08,
      "learning_rate": 0.09211892408475818,
      "loss": 0.0,
      "num_input_tokens_seen": 14225352,
      "step": 25045
    },
    {
      "epoch": 439.4778761061947,
      "grad_norm": 7.425308812258891e-09,
      "learning_rate": 0.09206458577661089,
      "loss": 0.0,
      "num_input_tokens_seen": 14228072,
      "step": 25050
    },
    {
      "epoch": 439.56637168141594,
      "grad_norm": 1.1506207542311131e-08,
      "learning_rate": 0.09201025640283263,
      "loss": 0.0,
      "num_input_tokens_seen": 14231080,
      "step": 25055
    },
    {
      "epoch": 439.65486725663715,
      "grad_norm": 1.501340030074516e-08,
      "learning_rate": 0.09195593597180148,
      "loss": 0.0,
      "num_input_tokens_seen": 14234200,
      "step": 25060
    },
    {
      "epoch": 439.7433628318584,
      "grad_norm": 1.4956277993860567e-08,
      "learning_rate": 0.09190162449189444,
      "loss": 0.0,
      "num_input_tokens_seen": 14237048,
      "step": 25065
    },
    {
      "epoch": 439.83185840707966,
      "grad_norm": 9.613093432392361e-09,
      "learning_rate": 0.09184732197148705,
      "loss": 0.0,
      "num_input_tokens_seen": 14239992,
      "step": 25070
    },
    {
      "epoch": 439.92035398230087,
      "grad_norm": 1.9428084030437276e-08,
      "learning_rate": 0.09179302841895343,
      "loss": 0.0,
      "num_input_tokens_seen": 14243480,
      "step": 25075
    },
    {
      "epoch": 440.0,
      "grad_norm": 2.007786648050569e-08,
      "learning_rate": 0.09173874384266625,
      "loss": 0.0,
      "num_input_tokens_seen": 14246096,
      "step": 25080
    },
    {
      "epoch": 440.08849557522126,
      "grad_norm": 1.276848404785369e-08,
      "learning_rate": 0.09168446825099695,
      "loss": 0.0,
      "num_input_tokens_seen": 14249216,
      "step": 25085
    },
    {
      "epoch": 440.17699115044246,
      "grad_norm": 1.1526397614147754e-08,
      "learning_rate": 0.09163020165231545,
      "loss": 0.0,
      "num_input_tokens_seen": 14252000,
      "step": 25090
    },
    {
      "epoch": 440.2654867256637,
      "grad_norm": 2.593488979130143e-08,
      "learning_rate": 0.09157594405499044,
      "loss": 0.0,
      "num_input_tokens_seen": 14255216,
      "step": 25095
    },
    {
      "epoch": 440.353982300885,
      "grad_norm": 3.653280344906307e-08,
      "learning_rate": 0.09152169546738899,
      "loss": 0.0,
      "num_input_tokens_seen": 14257904,
      "step": 25100
    },
    {
      "epoch": 440.4424778761062,
      "grad_norm": 2.470744497884425e-08,
      "learning_rate": 0.09146745589787698,
      "loss": 0.0,
      "num_input_tokens_seen": 14260496,
      "step": 25105
    },
    {
      "epoch": 440.53097345132744,
      "grad_norm": 1.726898801734933e-08,
      "learning_rate": 0.09141322535481891,
      "loss": 0.0,
      "num_input_tokens_seen": 14263680,
      "step": 25110
    },
    {
      "epoch": 440.6194690265487,
      "grad_norm": 3.3338444893615815e-08,
      "learning_rate": 0.0913590038465777,
      "loss": 0.0,
      "num_input_tokens_seen": 14266272,
      "step": 25115
    },
    {
      "epoch": 440.7079646017699,
      "grad_norm": 6.6506156137791095e-09,
      "learning_rate": 0.09130479138151505,
      "loss": 0.0,
      "num_input_tokens_seen": 14268784,
      "step": 25120
    },
    {
      "epoch": 440.79646017699116,
      "grad_norm": 7.137592294981232e-09,
      "learning_rate": 0.09125058796799114,
      "loss": 0.0,
      "num_input_tokens_seen": 14271952,
      "step": 25125
    },
    {
      "epoch": 440.88495575221236,
      "grad_norm": 2.135628029975578e-08,
      "learning_rate": 0.09119639361436485,
      "loss": 0.0,
      "num_input_tokens_seen": 14274608,
      "step": 25130
    },
    {
      "epoch": 440.9734513274336,
      "grad_norm": 6.746412317681916e-09,
      "learning_rate": 0.09114220832899368,
      "loss": 0.0,
      "num_input_tokens_seen": 14277520,
      "step": 25135
    },
    {
      "epoch": 441.05309734513276,
      "grad_norm": 1.4916572865786293e-08,
      "learning_rate": 0.0910880321202336,
      "loss": 0.0,
      "num_input_tokens_seen": 14280120,
      "step": 25140
    },
    {
      "epoch": 441.14159292035396,
      "grad_norm": 1.1636162255967974e-08,
      "learning_rate": 0.09103386499643933,
      "loss": 0.0,
      "num_input_tokens_seen": 14283176,
      "step": 25145
    },
    {
      "epoch": 441.2300884955752,
      "grad_norm": 1.0038572639814447e-08,
      "learning_rate": 0.09097970696596407,
      "loss": 0.0,
      "num_input_tokens_seen": 14285800,
      "step": 25150
    },
    {
      "epoch": 441.3185840707965,
      "grad_norm": 8.630494541250755e-09,
      "learning_rate": 0.09092555803715971,
      "loss": 0.0,
      "num_input_tokens_seen": 14288680,
      "step": 25155
    },
    {
      "epoch": 441.4070796460177,
      "grad_norm": 7.586899997136243e-09,
      "learning_rate": 0.0908714182183767,
      "loss": 0.0,
      "num_input_tokens_seen": 14291416,
      "step": 25160
    },
    {
      "epoch": 441.49557522123894,
      "grad_norm": 2.645566965497892e-08,
      "learning_rate": 0.090817287517964,
      "loss": 0.0,
      "num_input_tokens_seen": 14294424,
      "step": 25165
    },
    {
      "epoch": 441.5840707964602,
      "grad_norm": 6.3991913989980276e-09,
      "learning_rate": 0.09076316594426931,
      "loss": 0.0,
      "num_input_tokens_seen": 14297048,
      "step": 25170
    },
    {
      "epoch": 441.6725663716814,
      "grad_norm": 8.215237379260998e-09,
      "learning_rate": 0.09070905350563888,
      "loss": 0.0,
      "num_input_tokens_seen": 14299672,
      "step": 25175
    },
    {
      "epoch": 441.76106194690266,
      "grad_norm": 2.2645725294978547e-08,
      "learning_rate": 0.09065495021041745,
      "loss": 0.0,
      "num_input_tokens_seen": 14302568,
      "step": 25180
    },
    {
      "epoch": 441.8495575221239,
      "grad_norm": 2.10178647819248e-08,
      "learning_rate": 0.09060085606694851,
      "loss": 0.0,
      "num_input_tokens_seen": 14305448,
      "step": 25185
    },
    {
      "epoch": 441.9380530973451,
      "grad_norm": 2.1103296887758916e-08,
      "learning_rate": 0.09054677108357405,
      "loss": 0.0,
      "num_input_tokens_seen": 14308680,
      "step": 25190
    },
    {
      "epoch": 442.01769911504425,
      "grad_norm": 1.5178985179886695e-08,
      "learning_rate": 0.09049269526863457,
      "loss": 0.0,
      "num_input_tokens_seen": 14311808,
      "step": 25195
    },
    {
      "epoch": 442.1061946902655,
      "grad_norm": 6.334731850188291e-09,
      "learning_rate": 0.09043862863046935,
      "loss": 0.0,
      "num_input_tokens_seen": 14314736,
      "step": 25200
    },
    {
      "epoch": 442.1061946902655,
      "eval_loss": 0.723112940788269,
      "eval_runtime": 0.9394,
      "eval_samples_per_second": 26.614,
      "eval_steps_per_second": 13.839,
      "num_input_tokens_seen": 14314736,
      "step": 25200
    },
    {
      "epoch": 442.1946902654867,
      "grad_norm": 1.5193407421065785e-08,
      "learning_rate": 0.09038457117741602,
      "loss": 0.0,
      "num_input_tokens_seen": 14317232,
      "step": 25205
    },
    {
      "epoch": 442.283185840708,
      "grad_norm": 3.3318954706373916e-08,
      "learning_rate": 0.09033052291781099,
      "loss": 0.0,
      "num_input_tokens_seen": 14319824,
      "step": 25210
    },
    {
      "epoch": 442.37168141592923,
      "grad_norm": 2.2404291755151462e-08,
      "learning_rate": 0.09027648385998926,
      "loss": 0.0,
      "num_input_tokens_seen": 14323552,
      "step": 25215
    },
    {
      "epoch": 442.46017699115043,
      "grad_norm": 2.2368542573758532e-08,
      "learning_rate": 0.09022245401228417,
      "loss": 0.0,
      "num_input_tokens_seen": 14326240,
      "step": 25220
    },
    {
      "epoch": 442.5486725663717,
      "grad_norm": 2.388278375065056e-08,
      "learning_rate": 0.09016843338302792,
      "loss": 0.0,
      "num_input_tokens_seen": 14329824,
      "step": 25225
    },
    {
      "epoch": 442.6371681415929,
      "grad_norm": 1.9167471165815186e-08,
      "learning_rate": 0.09011442198055115,
      "loss": 0.0,
      "num_input_tokens_seen": 14332720,
      "step": 25230
    },
    {
      "epoch": 442.72566371681415,
      "grad_norm": 8.309196886102654e-09,
      "learning_rate": 0.09006041981318305,
      "loss": 0.0,
      "num_input_tokens_seen": 14335168,
      "step": 25235
    },
    {
      "epoch": 442.8141592920354,
      "grad_norm": 6.149508013919558e-09,
      "learning_rate": 0.09000642688925149,
      "loss": 0.0,
      "num_input_tokens_seen": 14337936,
      "step": 25240
    },
    {
      "epoch": 442.9026548672566,
      "grad_norm": 1.3960371525456594e-08,
      "learning_rate": 0.0899524432170828,
      "loss": 0.0,
      "num_input_tokens_seen": 14340288,
      "step": 25245
    },
    {
      "epoch": 442.9911504424779,
      "grad_norm": 1.8322767303402543e-08,
      "learning_rate": 0.08989846880500196,
      "loss": 0.0,
      "num_input_tokens_seen": 14343200,
      "step": 25250
    },
    {
      "epoch": 443.070796460177,
      "grad_norm": 1.4297336647928205e-08,
      "learning_rate": 0.08984450366133256,
      "loss": 0.0,
      "num_input_tokens_seen": 14345424,
      "step": 25255
    },
    {
      "epoch": 443.1592920353982,
      "grad_norm": 2.311042557323617e-08,
      "learning_rate": 0.08979054779439664,
      "loss": 0.0,
      "num_input_tokens_seen": 14348080,
      "step": 25260
    },
    {
      "epoch": 443.24778761061947,
      "grad_norm": 1.4854398600050445e-08,
      "learning_rate": 0.08973660121251485,
      "loss": 0.0,
      "num_input_tokens_seen": 14351088,
      "step": 25265
    },
    {
      "epoch": 443.3362831858407,
      "grad_norm": 3.098662659795082e-08,
      "learning_rate": 0.08968266392400655,
      "loss": 0.0,
      "num_input_tokens_seen": 14354304,
      "step": 25270
    },
    {
      "epoch": 443.42477876106193,
      "grad_norm": 1.7175400657265527e-08,
      "learning_rate": 0.0896287359371894,
      "loss": 0.0,
      "num_input_tokens_seen": 14356976,
      "step": 25275
    },
    {
      "epoch": 443.5132743362832,
      "grad_norm": 2.475711369243072e-08,
      "learning_rate": 0.08957481726037989,
      "loss": 0.0,
      "num_input_tokens_seen": 14359952,
      "step": 25280
    },
    {
      "epoch": 443.60176991150445,
      "grad_norm": 3.724558794715449e-08,
      "learning_rate": 0.08952090790189286,
      "loss": 0.0,
      "num_input_tokens_seen": 14362992,
      "step": 25285
    },
    {
      "epoch": 443.69026548672565,
      "grad_norm": 1.7159591081394865e-08,
      "learning_rate": 0.08946700787004187,
      "loss": 0.0,
      "num_input_tokens_seen": 14365696,
      "step": 25290
    },
    {
      "epoch": 443.7787610619469,
      "grad_norm": 3.156666394943386e-08,
      "learning_rate": 0.08941311717313899,
      "loss": 0.0,
      "num_input_tokens_seen": 14368800,
      "step": 25295
    },
    {
      "epoch": 443.86725663716817,
      "grad_norm": 6.053795242877413e-09,
      "learning_rate": 0.08935923581949483,
      "loss": 0.0,
      "num_input_tokens_seen": 14371440,
      "step": 25300
    },
    {
      "epoch": 443.95575221238937,
      "grad_norm": 4.214627224996548e-08,
      "learning_rate": 0.0893053638174185,
      "loss": 0.0,
      "num_input_tokens_seen": 14374960,
      "step": 25305
    },
    {
      "epoch": 444.0353982300885,
      "grad_norm": 1.394225890294365e-08,
      "learning_rate": 0.0892515011752179,
      "loss": 0.0,
      "num_input_tokens_seen": 14377168,
      "step": 25310
    },
    {
      "epoch": 444.12389380530976,
      "grad_norm": 2.4382996954841474e-08,
      "learning_rate": 0.08919764790119918,
      "loss": 0.0,
      "num_input_tokens_seen": 14380288,
      "step": 25315
    },
    {
      "epoch": 444.21238938053096,
      "grad_norm": 1.027482721127626e-08,
      "learning_rate": 0.08914380400366727,
      "loss": 0.0,
      "num_input_tokens_seen": 14382640,
      "step": 25320
    },
    {
      "epoch": 444.3008849557522,
      "grad_norm": 1.8817347680055718e-08,
      "learning_rate": 0.08908996949092551,
      "loss": 0.0,
      "num_input_tokens_seen": 14385440,
      "step": 25325
    },
    {
      "epoch": 444.3893805309734,
      "grad_norm": 1.8012066504979884e-08,
      "learning_rate": 0.08903614437127592,
      "loss": 0.0,
      "num_input_tokens_seen": 14388304,
      "step": 25330
    },
    {
      "epoch": 444.4778761061947,
      "grad_norm": 3.427703276770444e-08,
      "learning_rate": 0.088982328653019,
      "loss": 0.0,
      "num_input_tokens_seen": 14391376,
      "step": 25335
    },
    {
      "epoch": 444.56637168141594,
      "grad_norm": 7.205828467249376e-08,
      "learning_rate": 0.0889285223444538,
      "loss": 0.0,
      "num_input_tokens_seen": 14394336,
      "step": 25340
    },
    {
      "epoch": 444.65486725663715,
      "grad_norm": 2.50008191926554e-08,
      "learning_rate": 0.08887472545387787,
      "loss": 0.0,
      "num_input_tokens_seen": 14396880,
      "step": 25345
    },
    {
      "epoch": 444.7433628318584,
      "grad_norm": 1.9071165979767102e-08,
      "learning_rate": 0.08882093798958751,
      "loss": 0.0,
      "num_input_tokens_seen": 14399520,
      "step": 25350
    },
    {
      "epoch": 444.83185840707966,
      "grad_norm": 3.08715826236039e-08,
      "learning_rate": 0.08876715995987726,
      "loss": 0.0,
      "num_input_tokens_seen": 14402304,
      "step": 25355
    },
    {
      "epoch": 444.92035398230087,
      "grad_norm": 1.5056425439752275e-08,
      "learning_rate": 0.08871339137304052,
      "loss": 0.0,
      "num_input_tokens_seen": 14405232,
      "step": 25360
    },
    {
      "epoch": 445.0,
      "grad_norm": 1.1761803975218754e-08,
      "learning_rate": 0.0886596322373689,
      "loss": 0.0,
      "num_input_tokens_seen": 14407480,
      "step": 25365
    },
    {
      "epoch": 445.08849557522126,
      "grad_norm": 8.249334548793286e-09,
      "learning_rate": 0.08860588256115293,
      "loss": 0.0,
      "num_input_tokens_seen": 14410616,
      "step": 25370
    },
    {
      "epoch": 445.17699115044246,
      "grad_norm": 1.917528358319487e-08,
      "learning_rate": 0.0885521423526814,
      "loss": 0.0,
      "num_input_tokens_seen": 14413176,
      "step": 25375
    },
    {
      "epoch": 445.2654867256637,
      "grad_norm": 8.394606787476278e-09,
      "learning_rate": 0.08849841162024165,
      "loss": 0.0,
      "num_input_tokens_seen": 14416200,
      "step": 25380
    },
    {
      "epoch": 445.353982300885,
      "grad_norm": 3.043290419668665e-08,
      "learning_rate": 0.08844469037211973,
      "loss": 0.0,
      "num_input_tokens_seen": 14419464,
      "step": 25385
    },
    {
      "epoch": 445.4424778761062,
      "grad_norm": 1.053585663157719e-08,
      "learning_rate": 0.08839097861660014,
      "loss": 0.0,
      "num_input_tokens_seen": 14421992,
      "step": 25390
    },
    {
      "epoch": 445.53097345132744,
      "grad_norm": 1.0072643163994144e-08,
      "learning_rate": 0.08833727636196585,
      "loss": 0.0,
      "num_input_tokens_seen": 14425256,
      "step": 25395
    },
    {
      "epoch": 445.6194690265487,
      "grad_norm": 1.0567135610983769e-08,
      "learning_rate": 0.08828358361649848,
      "loss": 0.0,
      "num_input_tokens_seen": 14428104,
      "step": 25400
    },
    {
      "epoch": 445.6194690265487,
      "eval_loss": 0.7107198238372803,
      "eval_runtime": 0.9415,
      "eval_samples_per_second": 26.553,
      "eval_steps_per_second": 13.808,
      "num_input_tokens_seen": 14428104,
      "step": 25400
    },
    {
      "epoch": 445.7079646017699,
      "grad_norm": 2.0073770201634034e-08,
      "learning_rate": 0.08822990038847807,
      "loss": 0.0,
      "num_input_tokens_seen": 14430552,
      "step": 25405
    },
    {
      "epoch": 445.79646017699116,
      "grad_norm": 1.3487524874733481e-08,
      "learning_rate": 0.08817622668618325,
      "loss": 0.0,
      "num_input_tokens_seen": 14433528,
      "step": 25410
    },
    {
      "epoch": 445.88495575221236,
      "grad_norm": 2.2903414276242984e-08,
      "learning_rate": 0.08812256251789125,
      "loss": 0.0,
      "num_input_tokens_seen": 14436056,
      "step": 25415
    },
    {
      "epoch": 445.9734513274336,
      "grad_norm": 2.7575371319699116e-08,
      "learning_rate": 0.08806890789187766,
      "loss": 0.0,
      "num_input_tokens_seen": 14439160,
      "step": 25420
    },
    {
      "epoch": 446.05309734513276,
      "grad_norm": 2.2124336140905143e-08,
      "learning_rate": 0.08801526281641672,
      "loss": 0.0,
      "num_input_tokens_seen": 14441688,
      "step": 25425
    },
    {
      "epoch": 446.14159292035396,
      "grad_norm": 2.027275414206997e-08,
      "learning_rate": 0.0879616272997813,
      "loss": 0.0,
      "num_input_tokens_seen": 14444552,
      "step": 25430
    },
    {
      "epoch": 446.2300884955752,
      "grad_norm": 3.1768315977842576e-08,
      "learning_rate": 0.08790800135024247,
      "loss": 0.0,
      "num_input_tokens_seen": 14447160,
      "step": 25435
    },
    {
      "epoch": 446.3185840707965,
      "grad_norm": 8.499662307315248e-09,
      "learning_rate": 0.08785438497607023,
      "loss": 0.0,
      "num_input_tokens_seen": 14450104,
      "step": 25440
    },
    {
      "epoch": 446.4070796460177,
      "grad_norm": 2.39212880615014e-08,
      "learning_rate": 0.08780077818553277,
      "loss": 0.0,
      "num_input_tokens_seen": 14453128,
      "step": 25445
    },
    {
      "epoch": 446.49557522123894,
      "grad_norm": 2.442332025509586e-08,
      "learning_rate": 0.0877471809868969,
      "loss": 0.0,
      "num_input_tokens_seen": 14456008,
      "step": 25450
    },
    {
      "epoch": 446.5840707964602,
      "grad_norm": 1.4147370386297098e-08,
      "learning_rate": 0.08769359338842811,
      "loss": 0.0,
      "num_input_tokens_seen": 14459208,
      "step": 25455
    },
    {
      "epoch": 446.6725663716814,
      "grad_norm": 2.328372339377438e-08,
      "learning_rate": 0.08764001539839016,
      "loss": 0.0,
      "num_input_tokens_seen": 14461800,
      "step": 25460
    },
    {
      "epoch": 446.76106194690266,
      "grad_norm": 1.9413610274909843e-08,
      "learning_rate": 0.08758644702504548,
      "loss": 0.0,
      "num_input_tokens_seen": 14464744,
      "step": 25465
    },
    {
      "epoch": 446.8495575221239,
      "grad_norm": 4.3745405520212444e-08,
      "learning_rate": 0.0875328882766551,
      "loss": 0.0,
      "num_input_tokens_seen": 14467672,
      "step": 25470
    },
    {
      "epoch": 446.9380530973451,
      "grad_norm": 1.1377708553084176e-08,
      "learning_rate": 0.08747933916147828,
      "loss": 0.0,
      "num_input_tokens_seen": 14470456,
      "step": 25475
    },
    {
      "epoch": 447.01769911504425,
      "grad_norm": 6.901486493404718e-09,
      "learning_rate": 0.0874257996877731,
      "loss": 0.0,
      "num_input_tokens_seen": 14472848,
      "step": 25480
    },
    {
      "epoch": 447.1061946902655,
      "grad_norm": 1.051503151217048e-08,
      "learning_rate": 0.08737226986379593,
      "loss": 0.0,
      "num_input_tokens_seen": 14475520,
      "step": 25485
    },
    {
      "epoch": 447.1946902654867,
      "grad_norm": 3.569459039454159e-08,
      "learning_rate": 0.08731874969780173,
      "loss": 0.0,
      "num_input_tokens_seen": 14478144,
      "step": 25490
    },
    {
      "epoch": 447.283185840708,
      "grad_norm": 1.2714255426260479e-08,
      "learning_rate": 0.08726523919804412,
      "loss": 0.0,
      "num_input_tokens_seen": 14481280,
      "step": 25495
    },
    {
      "epoch": 447.37168141592923,
      "grad_norm": 1.4650731294807429e-08,
      "learning_rate": 0.08721173837277492,
      "loss": 0.0,
      "num_input_tokens_seen": 14484608,
      "step": 25500
    },
    {
      "epoch": 447.46017699115043,
      "grad_norm": 2.014829902918791e-08,
      "learning_rate": 0.08715824723024479,
      "loss": 0.0,
      "num_input_tokens_seen": 14487248,
      "step": 25505
    },
    {
      "epoch": 447.5486725663717,
      "grad_norm": 2.5937216818761044e-08,
      "learning_rate": 0.08710476577870258,
      "loss": 0.0,
      "num_input_tokens_seen": 14489776,
      "step": 25510
    },
    {
      "epoch": 447.6371681415929,
      "grad_norm": 1.6054478635396663e-08,
      "learning_rate": 0.08705129402639587,
      "loss": 0.0,
      "num_input_tokens_seen": 14492768,
      "step": 25515
    },
    {
      "epoch": 447.72566371681415,
      "grad_norm": 7.921170386282483e-09,
      "learning_rate": 0.08699783198157078,
      "loss": 0.0,
      "num_input_tokens_seen": 14495664,
      "step": 25520
    },
    {
      "epoch": 447.8141592920354,
      "grad_norm": 2.781213126468174e-08,
      "learning_rate": 0.08694437965247163,
      "loss": 0.0,
      "num_input_tokens_seen": 14498544,
      "step": 25525
    },
    {
      "epoch": 447.9026548672566,
      "grad_norm": 1.509962643808649e-08,
      "learning_rate": 0.08689093704734165,
      "loss": 0.0,
      "num_input_tokens_seen": 14501520,
      "step": 25530
    },
    {
      "epoch": 447.9911504424779,
      "grad_norm": 3.7440436528868304e-08,
      "learning_rate": 0.08683750417442222,
      "loss": 0.0,
      "num_input_tokens_seen": 14504656,
      "step": 25535
    },
    {
      "epoch": 448.070796460177,
      "grad_norm": 1.6950535197679528e-08,
      "learning_rate": 0.08678408104195334,
      "loss": 0.0,
      "num_input_tokens_seen": 14506960,
      "step": 25540
    },
    {
      "epoch": 448.1592920353982,
      "grad_norm": 1.8642110077848884e-08,
      "learning_rate": 0.08673066765817365,
      "loss": 0.0,
      "num_input_tokens_seen": 14510720,
      "step": 25545
    },
    {
      "epoch": 448.24778761061947,
      "grad_norm": 1.9420292929339666e-08,
      "learning_rate": 0.08667726403132005,
      "loss": 0.0,
      "num_input_tokens_seen": 14513552,
      "step": 25550
    },
    {
      "epoch": 448.3362831858407,
      "grad_norm": 2.0546952583799794e-08,
      "learning_rate": 0.0866238701696281,
      "loss": 0.0,
      "num_input_tokens_seen": 14516592,
      "step": 25555
    },
    {
      "epoch": 448.42477876106193,
      "grad_norm": 3.2289456441958464e-08,
      "learning_rate": 0.08657048608133185,
      "loss": 0.0,
      "num_input_tokens_seen": 14519216,
      "step": 25560
    },
    {
      "epoch": 448.5132743362832,
      "grad_norm": 2.1484304113528196e-08,
      "learning_rate": 0.08651711177466369,
      "loss": 0.0,
      "num_input_tokens_seen": 14522608,
      "step": 25565
    },
    {
      "epoch": 448.60176991150445,
      "grad_norm": 2.0192853611433748e-08,
      "learning_rate": 0.08646374725785466,
      "loss": 0.0,
      "num_input_tokens_seen": 14524928,
      "step": 25570
    },
    {
      "epoch": 448.69026548672565,
      "grad_norm": 4.986773305404313e-08,
      "learning_rate": 0.08641039253913434,
      "loss": 0.0,
      "num_input_tokens_seen": 14527616,
      "step": 25575
    },
    {
      "epoch": 448.7787610619469,
      "grad_norm": 1.8009783886441255e-08,
      "learning_rate": 0.08635704762673052,
      "loss": 0.0,
      "num_input_tokens_seen": 14530688,
      "step": 25580
    },
    {
      "epoch": 448.86725663716817,
      "grad_norm": 1.97724165929003e-08,
      "learning_rate": 0.08630371252886981,
      "loss": 0.0,
      "num_input_tokens_seen": 14533152,
      "step": 25585
    },
    {
      "epoch": 448.95575221238937,
      "grad_norm": 7.6031962947809e-09,
      "learning_rate": 0.08625038725377704,
      "loss": 0.0,
      "num_input_tokens_seen": 14535984,
      "step": 25590
    },
    {
      "epoch": 449.0353982300885,
      "grad_norm": 6.799034668603099e-09,
      "learning_rate": 0.08619707180967566,
      "loss": 0.0,
      "num_input_tokens_seen": 14537984,
      "step": 25595
    },
    {
      "epoch": 449.12389380530976,
      "grad_norm": 1.5876528536296064e-08,
      "learning_rate": 0.08614376620478768,
      "loss": 0.0,
      "num_input_tokens_seen": 14541136,
      "step": 25600
    },
    {
      "epoch": 449.12389380530976,
      "eval_loss": 0.7242066264152527,
      "eval_runtime": 0.9504,
      "eval_samples_per_second": 26.305,
      "eval_steps_per_second": 13.679,
      "num_input_tokens_seen": 14541136,
      "step": 25600
    },
    {
      "epoch": 449.21238938053096,
      "grad_norm": 3.781121904467e-08,
      "learning_rate": 0.08609047044733344,
      "loss": 0.0,
      "num_input_tokens_seen": 14543632,
      "step": 25605
    },
    {
      "epoch": 449.3008849557522,
      "grad_norm": 2.187057468461262e-08,
      "learning_rate": 0.08603718454553168,
      "loss": 0.0,
      "num_input_tokens_seen": 14546160,
      "step": 25610
    },
    {
      "epoch": 449.3893805309734,
      "grad_norm": 6.018976872468329e-09,
      "learning_rate": 0.08598390850759997,
      "loss": 0.0,
      "num_input_tokens_seen": 14548832,
      "step": 25615
    },
    {
      "epoch": 449.4778761061947,
      "grad_norm": 9.47996436906351e-09,
      "learning_rate": 0.08593064234175397,
      "loss": 0.0,
      "num_input_tokens_seen": 14551792,
      "step": 25620
    },
    {
      "epoch": 449.56637168141594,
      "grad_norm": 3.0490348024159175e-08,
      "learning_rate": 0.08587738605620815,
      "loss": 0.0,
      "num_input_tokens_seen": 14554608,
      "step": 25625
    },
    {
      "epoch": 449.65486725663715,
      "grad_norm": 2.614340210982391e-08,
      "learning_rate": 0.08582413965917512,
      "loss": 0.0,
      "num_input_tokens_seen": 14558080,
      "step": 25630
    },
    {
      "epoch": 449.7433628318584,
      "grad_norm": 4.157394073445175e-08,
      "learning_rate": 0.08577090315886628,
      "loss": 0.0,
      "num_input_tokens_seen": 14561024,
      "step": 25635
    },
    {
      "epoch": 449.83185840707966,
      "grad_norm": 3.051622954330924e-08,
      "learning_rate": 0.08571767656349136,
      "loss": 0.0,
      "num_input_tokens_seen": 14564240,
      "step": 25640
    },
    {
      "epoch": 449.92035398230087,
      "grad_norm": 6.09460641953774e-08,
      "learning_rate": 0.08566445988125847,
      "loss": 0.0,
      "num_input_tokens_seen": 14567040,
      "step": 25645
    },
    {
      "epoch": 450.0,
      "grad_norm": 2.030329682156662e-08,
      "learning_rate": 0.08561125312037436,
      "loss": 0.0,
      "num_input_tokens_seen": 14569608,
      "step": 25650
    },
    {
      "epoch": 450.08849557522126,
      "grad_norm": 1.5409142406497267e-08,
      "learning_rate": 0.08555805628904424,
      "loss": 0.0,
      "num_input_tokens_seen": 14572744,
      "step": 25655
    },
    {
      "epoch": 450.17699115044246,
      "grad_norm": 1.2041556196606962e-08,
      "learning_rate": 0.08550486939547161,
      "loss": 0.0,
      "num_input_tokens_seen": 14575960,
      "step": 25660
    },
    {
      "epoch": 450.2654867256637,
      "grad_norm": 2.0493065022719748e-08,
      "learning_rate": 0.08545169244785869,
      "loss": 0.0,
      "num_input_tokens_seen": 14578824,
      "step": 25665
    },
    {
      "epoch": 450.353982300885,
      "grad_norm": 9.6888017608876e-09,
      "learning_rate": 0.08539852545440589,
      "loss": 0.0,
      "num_input_tokens_seen": 14581752,
      "step": 25670
    },
    {
      "epoch": 450.4424778761062,
      "grad_norm": 1.802425586561185e-08,
      "learning_rate": 0.08534536842331235,
      "loss": 0.0,
      "num_input_tokens_seen": 14584520,
      "step": 25675
    },
    {
      "epoch": 450.53097345132744,
      "grad_norm": 1.569453900174267e-08,
      "learning_rate": 0.08529222136277545,
      "loss": 0.0,
      "num_input_tokens_seen": 14587320,
      "step": 25680
    },
    {
      "epoch": 450.6194690265487,
      "grad_norm": 1.657067016935798e-08,
      "learning_rate": 0.08523908428099125,
      "loss": 0.0,
      "num_input_tokens_seen": 14590216,
      "step": 25685
    },
    {
      "epoch": 450.7079646017699,
      "grad_norm": 2.4220101479954792e-08,
      "learning_rate": 0.08518595718615402,
      "loss": 0.0,
      "num_input_tokens_seen": 14592824,
      "step": 25690
    },
    {
      "epoch": 450.79646017699116,
      "grad_norm": 8.054455769013202e-09,
      "learning_rate": 0.08513284008645675,
      "loss": 0.0,
      "num_input_tokens_seen": 14595464,
      "step": 25695
    },
    {
      "epoch": 450.88495575221236,
      "grad_norm": 1.712121999730698e-08,
      "learning_rate": 0.08507973299009065,
      "loss": 0.0,
      "num_input_tokens_seen": 14598680,
      "step": 25700
    },
    {
      "epoch": 450.9734513274336,
      "grad_norm": 9.727110672486106e-09,
      "learning_rate": 0.08502663590524563,
      "loss": 0.0,
      "num_input_tokens_seen": 14601352,
      "step": 25705
    },
    {
      "epoch": 451.05309734513276,
      "grad_norm": 2.2093328055916572e-08,
      "learning_rate": 0.08497354884010981,
      "loss": 0.0,
      "num_input_tokens_seen": 14603616,
      "step": 25710
    },
    {
      "epoch": 451.14159292035396,
      "grad_norm": 1.7201818636181088e-08,
      "learning_rate": 0.0849204718028699,
      "loss": 0.0,
      "num_input_tokens_seen": 14606192,
      "step": 25715
    },
    {
      "epoch": 451.2300884955752,
      "grad_norm": 2.0157310487434188e-08,
      "learning_rate": 0.08486740480171118,
      "loss": 0.0,
      "num_input_tokens_seen": 14608576,
      "step": 25720
    },
    {
      "epoch": 451.3185840707965,
      "grad_norm": 1.3712981861147e-08,
      "learning_rate": 0.08481434784481706,
      "loss": 0.0,
      "num_input_tokens_seen": 14611856,
      "step": 25725
    },
    {
      "epoch": 451.4070796460177,
      "grad_norm": 2.4859916791797332e-08,
      "learning_rate": 0.08476130094036968,
      "loss": 0.0,
      "num_input_tokens_seen": 14614592,
      "step": 25730
    },
    {
      "epoch": 451.49557522123894,
      "grad_norm": 2.6413951914605605e-08,
      "learning_rate": 0.08470826409654961,
      "loss": 0.0,
      "num_input_tokens_seen": 14618064,
      "step": 25735
    },
    {
      "epoch": 451.5840707964602,
      "grad_norm": 1.3077826821472627e-08,
      "learning_rate": 0.08465523732153564,
      "loss": 0.0,
      "num_input_tokens_seen": 14620832,
      "step": 25740
    },
    {
      "epoch": 451.6725663716814,
      "grad_norm": 3.175239626784787e-08,
      "learning_rate": 0.08460222062350532,
      "loss": 0.0,
      "num_input_tokens_seen": 14623856,
      "step": 25745
    },
    {
      "epoch": 451.76106194690266,
      "grad_norm": 1.0286514751101095e-08,
      "learning_rate": 0.08454921401063442,
      "loss": 0.0,
      "num_input_tokens_seen": 14626592,
      "step": 25750
    },
    {
      "epoch": 451.8495575221239,
      "grad_norm": 2.5587311824892822e-08,
      "learning_rate": 0.08449621749109716,
      "loss": 0.0,
      "num_input_tokens_seen": 14629616,
      "step": 25755
    },
    {
      "epoch": 451.9380530973451,
      "grad_norm": 2.4857811808942643e-08,
      "learning_rate": 0.08444323107306641,
      "loss": 0.0,
      "num_input_tokens_seen": 14632288,
      "step": 25760
    },
    {
      "epoch": 452.01769911504425,
      "grad_norm": 1.9037411647104818e-08,
      "learning_rate": 0.0843902547647132,
      "loss": 0.0,
      "num_input_tokens_seen": 14635040,
      "step": 25765
    },
    {
      "epoch": 452.1061946902655,
      "grad_norm": 2.481990613034668e-08,
      "learning_rate": 0.0843372885742072,
      "loss": 0.0,
      "num_input_tokens_seen": 14638080,
      "step": 25770
    },
    {
      "epoch": 452.1946902654867,
      "grad_norm": 2.7649472045254697e-08,
      "learning_rate": 0.08428433250971652,
      "loss": 0.0,
      "num_input_tokens_seen": 14640768,
      "step": 25775
    },
    {
      "epoch": 452.283185840708,
      "grad_norm": 3.090655198434433e-08,
      "learning_rate": 0.08423138657940757,
      "loss": 0.0,
      "num_input_tokens_seen": 14643808,
      "step": 25780
    },
    {
      "epoch": 452.37168141592923,
      "grad_norm": 2.8886210756695618e-08,
      "learning_rate": 0.08417845079144536,
      "loss": 0.0,
      "num_input_tokens_seen": 14646704,
      "step": 25785
    },
    {
      "epoch": 452.46017699115043,
      "grad_norm": 1.8269835422302094e-08,
      "learning_rate": 0.08412552515399314,
      "loss": 0.0,
      "num_input_tokens_seen": 14649808,
      "step": 25790
    },
    {
      "epoch": 452.5486725663717,
      "grad_norm": 7.819893177440917e-09,
      "learning_rate": 0.08407260967521278,
      "loss": 0.0,
      "num_input_tokens_seen": 14652400,
      "step": 25795
    },
    {
      "epoch": 452.6371681415929,
      "grad_norm": 3.363176048765126e-08,
      "learning_rate": 0.08401970436326454,
      "loss": 0.0,
      "num_input_tokens_seen": 14655696,
      "step": 25800
    },
    {
      "epoch": 452.6371681415929,
      "eval_loss": 0.7248790264129639,
      "eval_runtime": 0.9145,
      "eval_samples_per_second": 27.338,
      "eval_steps_per_second": 14.216,
      "num_input_tokens_seen": 14655696,
      "step": 25800
    },
    {
      "epoch": 452.72566371681415,
      "grad_norm": 6.869238067253036e-09,
      "learning_rate": 0.08396680922630702,
      "loss": 0.0,
      "num_input_tokens_seen": 14658528,
      "step": 25805
    },
    {
      "epoch": 452.8141592920354,
      "grad_norm": 1.1808471533925058e-08,
      "learning_rate": 0.08391392427249732,
      "loss": 0.0,
      "num_input_tokens_seen": 14661792,
      "step": 25810
    },
    {
      "epoch": 452.9026548672566,
      "grad_norm": 7.543462743342388e-09,
      "learning_rate": 0.08386104950999107,
      "loss": 0.0,
      "num_input_tokens_seen": 14664272,
      "step": 25815
    },
    {
      "epoch": 452.9911504424779,
      "grad_norm": 1.2344845146117223e-08,
      "learning_rate": 0.0838081849469421,
      "loss": 0.0,
      "num_input_tokens_seen": 14666800,
      "step": 25820
    },
    {
      "epoch": 453.070796460177,
      "grad_norm": 2.0235297881754377e-08,
      "learning_rate": 0.08375533059150281,
      "loss": 0.0,
      "num_input_tokens_seen": 14669096,
      "step": 25825
    },
    {
      "epoch": 453.1592920353982,
      "grad_norm": 1.372174995850628e-08,
      "learning_rate": 0.08370248645182406,
      "loss": 0.0,
      "num_input_tokens_seen": 14672008,
      "step": 25830
    },
    {
      "epoch": 453.24778761061947,
      "grad_norm": 1.8422300129827818e-08,
      "learning_rate": 0.083649652536055,
      "loss": 0.0,
      "num_input_tokens_seen": 14674920,
      "step": 25835
    },
    {
      "epoch": 453.3362831858407,
      "grad_norm": 7.225873677185746e-09,
      "learning_rate": 0.08359682885234339,
      "loss": 0.0,
      "num_input_tokens_seen": 14678232,
      "step": 25840
    },
    {
      "epoch": 453.42477876106193,
      "grad_norm": 1.55314907601678e-08,
      "learning_rate": 0.08354401540883516,
      "loss": 0.0,
      "num_input_tokens_seen": 14680744,
      "step": 25845
    },
    {
      "epoch": 453.5132743362832,
      "grad_norm": 2.195238124613752e-08,
      "learning_rate": 0.0834912122136749,
      "loss": 0.0,
      "num_input_tokens_seen": 14683560,
      "step": 25850
    },
    {
      "epoch": 453.60176991150445,
      "grad_norm": 8.774865278837751e-09,
      "learning_rate": 0.0834384192750056,
      "loss": 0.0,
      "num_input_tokens_seen": 14686808,
      "step": 25855
    },
    {
      "epoch": 453.69026548672565,
      "grad_norm": 2.3718182973198054e-08,
      "learning_rate": 0.08338563660096844,
      "loss": 0.0,
      "num_input_tokens_seen": 14689640,
      "step": 25860
    },
    {
      "epoch": 453.7787610619469,
      "grad_norm": 2.1569453778624847e-08,
      "learning_rate": 0.08333286419970329,
      "loss": 0.0,
      "num_input_tokens_seen": 14692248,
      "step": 25865
    },
    {
      "epoch": 453.86725663716817,
      "grad_norm": 1.2287381778719464e-08,
      "learning_rate": 0.08328010207934824,
      "loss": 0.0,
      "num_input_tokens_seen": 14695032,
      "step": 25870
    },
    {
      "epoch": 453.95575221238937,
      "grad_norm": 1.7634054216841832e-08,
      "learning_rate": 0.08322735024803989,
      "loss": 0.0,
      "num_input_tokens_seen": 14697960,
      "step": 25875
    },
    {
      "epoch": 454.0353982300885,
      "grad_norm": 2.3953031558221483e-08,
      "learning_rate": 0.08317460871391331,
      "loss": 0.0,
      "num_input_tokens_seen": 14700632,
      "step": 25880
    },
    {
      "epoch": 454.12389380530976,
      "grad_norm": 9.206029716324338e-09,
      "learning_rate": 0.08312187748510179,
      "loss": 0.0,
      "num_input_tokens_seen": 14703112,
      "step": 25885
    },
    {
      "epoch": 454.21238938053096,
      "grad_norm": 9.430844549740414e-09,
      "learning_rate": 0.08306915656973726,
      "loss": 0.0,
      "num_input_tokens_seen": 14705736,
      "step": 25890
    },
    {
      "epoch": 454.3008849557522,
      "grad_norm": 1.4949970150723857e-08,
      "learning_rate": 0.08301644597594988,
      "loss": 0.0,
      "num_input_tokens_seen": 14708888,
      "step": 25895
    },
    {
      "epoch": 454.3893805309734,
      "grad_norm": 2.4575541601734585e-08,
      "learning_rate": 0.08296374571186826,
      "loss": 0.0,
      "num_input_tokens_seen": 14711496,
      "step": 25900
    },
    {
      "epoch": 454.4778761061947,
      "grad_norm": 2.778141094950115e-08,
      "learning_rate": 0.08291105578561955,
      "loss": 0.0,
      "num_input_tokens_seen": 14713992,
      "step": 25905
    },
    {
      "epoch": 454.56637168141594,
      "grad_norm": 2.2121483311821066e-08,
      "learning_rate": 0.08285837620532904,
      "loss": 0.0,
      "num_input_tokens_seen": 14716824,
      "step": 25910
    },
    {
      "epoch": 454.65486725663715,
      "grad_norm": 2.3367876522684128e-08,
      "learning_rate": 0.0828057069791207,
      "loss": 0.0,
      "num_input_tokens_seen": 14719784,
      "step": 25915
    },
    {
      "epoch": 454.7433628318584,
      "grad_norm": 3.2769953861588874e-08,
      "learning_rate": 0.0827530481151168,
      "loss": 0.0,
      "num_input_tokens_seen": 14722456,
      "step": 25920
    },
    {
      "epoch": 454.83185840707966,
      "grad_norm": 2.4188160807625536e-08,
      "learning_rate": 0.08270039962143792,
      "loss": 0.0,
      "num_input_tokens_seen": 14725944,
      "step": 25925
    },
    {
      "epoch": 454.92035398230087,
      "grad_norm": 1.6900735033686942e-08,
      "learning_rate": 0.08264776150620314,
      "loss": 0.0,
      "num_input_tokens_seen": 14728984,
      "step": 25930
    },
    {
      "epoch": 455.0,
      "grad_norm": 5.6147001714634825e-08,
      "learning_rate": 0.08259513377753,
      "loss": 0.0,
      "num_input_tokens_seen": 14731520,
      "step": 25935
    },
    {
      "epoch": 455.08849557522126,
      "grad_norm": 6.20616269486618e-09,
      "learning_rate": 0.08254251644353423,
      "loss": 0.0,
      "num_input_tokens_seen": 14733952,
      "step": 25940
    },
    {
      "epoch": 455.17699115044246,
      "grad_norm": 9.417130186761824e-09,
      "learning_rate": 0.08248990951233022,
      "loss": 0.0,
      "num_input_tokens_seen": 14737200,
      "step": 25945
    },
    {
      "epoch": 455.2654867256637,
      "grad_norm": 3.5309636103875164e-08,
      "learning_rate": 0.08243731299203048,
      "loss": 0.0,
      "num_input_tokens_seen": 14740208,
      "step": 25950
    },
    {
      "epoch": 455.353982300885,
      "grad_norm": 1.8698736781175285e-08,
      "learning_rate": 0.08238472689074612,
      "loss": 0.0,
      "num_input_tokens_seen": 14742944,
      "step": 25955
    },
    {
      "epoch": 455.4424778761062,
      "grad_norm": 2.1812951445099316e-08,
      "learning_rate": 0.08233215121658666,
      "loss": 0.0,
      "num_input_tokens_seen": 14745824,
      "step": 25960
    },
    {
      "epoch": 455.53097345132744,
      "grad_norm": 2.241643848321928e-08,
      "learning_rate": 0.08227958597765982,
      "loss": 0.0,
      "num_input_tokens_seen": 14748576,
      "step": 25965
    },
    {
      "epoch": 455.6194690265487,
      "grad_norm": 2.4458254088699505e-08,
      "learning_rate": 0.08222703118207181,
      "loss": 0.0,
      "num_input_tokens_seen": 14750992,
      "step": 25970
    },
    {
      "epoch": 455.7079646017699,
      "grad_norm": 4.6724650815122004e-08,
      "learning_rate": 0.08217448683792734,
      "loss": 0.0,
      "num_input_tokens_seen": 14753936,
      "step": 25975
    },
    {
      "epoch": 455.79646017699116,
      "grad_norm": 6.966250687412412e-09,
      "learning_rate": 0.08212195295332926,
      "loss": 0.0,
      "num_input_tokens_seen": 14756976,
      "step": 25980
    },
    {
      "epoch": 455.88495575221236,
      "grad_norm": 2.328061654566227e-08,
      "learning_rate": 0.08206942953637915,
      "loss": 0.0,
      "num_input_tokens_seen": 14760288,
      "step": 25985
    },
    {
      "epoch": 455.9734513274336,
      "grad_norm": 1.4692023597717707e-08,
      "learning_rate": 0.08201691659517658,
      "loss": 0.0,
      "num_input_tokens_seen": 14763120,
      "step": 25990
    },
    {
      "epoch": 456.05309734513276,
      "grad_norm": 1.766782986578619e-08,
      "learning_rate": 0.08196441413781981,
      "loss": 0.0,
      "num_input_tokens_seen": 14765480,
      "step": 25995
    },
    {
      "epoch": 456.14159292035396,
      "grad_norm": 3.0788282145977064e-08,
      "learning_rate": 0.08191192217240544,
      "loss": 0.0,
      "num_input_tokens_seen": 14768168,
      "step": 26000
    },
    {
      "epoch": 456.14159292035396,
      "eval_loss": 0.7227019667625427,
      "eval_runtime": 0.9419,
      "eval_samples_per_second": 26.543,
      "eval_steps_per_second": 13.803,
      "num_input_tokens_seen": 14768168,
      "step": 26000
    },
    {
      "epoch": 456.2300884955752,
      "grad_norm": 1.551413220113318e-08,
      "learning_rate": 0.08185944070702823,
      "loss": 0.0,
      "num_input_tokens_seen": 14770904,
      "step": 26005
    },
    {
      "epoch": 456.3185840707965,
      "grad_norm": 6.904313565314624e-09,
      "learning_rate": 0.08180696974978159,
      "loss": 0.0,
      "num_input_tokens_seen": 14774024,
      "step": 26010
    },
    {
      "epoch": 456.4070796460177,
      "grad_norm": 2.792873310397681e-08,
      "learning_rate": 0.08175450930875724,
      "loss": 0.0,
      "num_input_tokens_seen": 14777208,
      "step": 26015
    },
    {
      "epoch": 456.49557522123894,
      "grad_norm": 1.9269453588321994e-08,
      "learning_rate": 0.08170205939204513,
      "loss": 0.0,
      "num_input_tokens_seen": 14779928,
      "step": 26020
    },
    {
      "epoch": 456.5840707964602,
      "grad_norm": 2.111347008337816e-08,
      "learning_rate": 0.08164962000773379,
      "loss": 0.0,
      "num_input_tokens_seen": 14783064,
      "step": 26025
    },
    {
      "epoch": 456.6725663716814,
      "grad_norm": 1.8315503780286235e-08,
      "learning_rate": 0.08159719116390995,
      "loss": 0.0,
      "num_input_tokens_seen": 14786072,
      "step": 26030
    },
    {
      "epoch": 456.76106194690266,
      "grad_norm": 9.732796790729026e-09,
      "learning_rate": 0.08154477286865887,
      "loss": 0.0,
      "num_input_tokens_seen": 14788584,
      "step": 26035
    },
    {
      "epoch": 456.8495575221239,
      "grad_norm": 1.4343155108065275e-08,
      "learning_rate": 0.08149236513006404,
      "loss": 0.0,
      "num_input_tokens_seen": 14791480,
      "step": 26040
    },
    {
      "epoch": 456.9380530973451,
      "grad_norm": 2.4700414158473905e-08,
      "learning_rate": 0.08143996795620746,
      "loss": 0.0,
      "num_input_tokens_seen": 14794696,
      "step": 26045
    },
    {
      "epoch": 457.01769911504425,
      "grad_norm": 1.5974714884237073e-08,
      "learning_rate": 0.08138758135516938,
      "loss": 0.0,
      "num_input_tokens_seen": 14796992,
      "step": 26050
    },
    {
      "epoch": 457.1061946902655,
      "grad_norm": 1.850927766611221e-08,
      "learning_rate": 0.08133520533502851,
      "loss": 0.0,
      "num_input_tokens_seen": 14800080,
      "step": 26055
    },
    {
      "epoch": 457.1946902654867,
      "grad_norm": 2.138373567106555e-08,
      "learning_rate": 0.08128283990386184,
      "loss": 0.0,
      "num_input_tokens_seen": 14802624,
      "step": 26060
    },
    {
      "epoch": 457.283185840708,
      "grad_norm": 2.281599620346242e-08,
      "learning_rate": 0.08123048506974488,
      "loss": 0.0,
      "num_input_tokens_seen": 14805408,
      "step": 26065
    },
    {
      "epoch": 457.37168141592923,
      "grad_norm": 1.7417775666217494e-08,
      "learning_rate": 0.08117814084075124,
      "loss": 0.0,
      "num_input_tokens_seen": 14808416,
      "step": 26070
    },
    {
      "epoch": 457.46017699115043,
      "grad_norm": 1.2906270718815449e-08,
      "learning_rate": 0.08112580722495318,
      "loss": 0.0,
      "num_input_tokens_seen": 14810992,
      "step": 26075
    },
    {
      "epoch": 457.5486725663717,
      "grad_norm": 1.875088351255272e-08,
      "learning_rate": 0.08107348423042122,
      "loss": 0.0,
      "num_input_tokens_seen": 14813776,
      "step": 26080
    },
    {
      "epoch": 457.6371681415929,
      "grad_norm": 2.4430056200230865e-08,
      "learning_rate": 0.08102117186522413,
      "loss": 0.0,
      "num_input_tokens_seen": 14817152,
      "step": 26085
    },
    {
      "epoch": 457.72566371681415,
      "grad_norm": 2.7215504516675537e-08,
      "learning_rate": 0.08096887013742916,
      "loss": 0.0,
      "num_input_tokens_seen": 14819984,
      "step": 26090
    },
    {
      "epoch": 457.8141592920354,
      "grad_norm": 1.548033345954991e-08,
      "learning_rate": 0.08091657905510198,
      "loss": 0.0,
      "num_input_tokens_seen": 14823344,
      "step": 26095
    },
    {
      "epoch": 457.9026548672566,
      "grad_norm": 1.9701548836792426e-08,
      "learning_rate": 0.08086429862630642,
      "loss": 0.0,
      "num_input_tokens_seen": 14826016,
      "step": 26100
    },
    {
      "epoch": 457.9911504424779,
      "grad_norm": 1.562801088539345e-08,
      "learning_rate": 0.08081202885910488,
      "loss": 0.0,
      "num_input_tokens_seen": 14828992,
      "step": 26105
    },
    {
      "epoch": 458.070796460177,
      "grad_norm": 1.2045366482027475e-08,
      "learning_rate": 0.08075976976155795,
      "loss": 0.0,
      "num_input_tokens_seen": 14831296,
      "step": 26110
    },
    {
      "epoch": 458.1592920353982,
      "grad_norm": 3.930703229571009e-08,
      "learning_rate": 0.08070752134172461,
      "loss": 0.0,
      "num_input_tokens_seen": 14834464,
      "step": 26115
    },
    {
      "epoch": 458.24778761061947,
      "grad_norm": 3.178202589992907e-08,
      "learning_rate": 0.08065528360766229,
      "loss": 0.0,
      "num_input_tokens_seen": 14837632,
      "step": 26120
    },
    {
      "epoch": 458.3362831858407,
      "grad_norm": 1.0115079440708996e-08,
      "learning_rate": 0.08060305656742664,
      "loss": 0.0,
      "num_input_tokens_seen": 14840336,
      "step": 26125
    },
    {
      "epoch": 458.42477876106193,
      "grad_norm": 1.1643413344586406e-08,
      "learning_rate": 0.08055084022907182,
      "loss": 0.0,
      "num_input_tokens_seen": 14842496,
      "step": 26130
    },
    {
      "epoch": 458.5132743362832,
      "grad_norm": 1.5300811284646443e-08,
      "learning_rate": 0.08049863460065014,
      "loss": 0.0,
      "num_input_tokens_seen": 14845360,
      "step": 26135
    },
    {
      "epoch": 458.60176991150445,
      "grad_norm": 1.4275751247794233e-08,
      "learning_rate": 0.0804464396902124,
      "loss": 0.0,
      "num_input_tokens_seen": 14848080,
      "step": 26140
    },
    {
      "epoch": 458.69026548672565,
      "grad_norm": 1.2652974668014849e-08,
      "learning_rate": 0.08039425550580777,
      "loss": 0.0,
      "num_input_tokens_seen": 14850864,
      "step": 26145
    },
    {
      "epoch": 458.7787610619469,
      "grad_norm": 1.4648738222433622e-08,
      "learning_rate": 0.08034208205548363,
      "loss": 0.0,
      "num_input_tokens_seen": 14854176,
      "step": 26150
    },
    {
      "epoch": 458.86725663716817,
      "grad_norm": 1.172292751761006e-08,
      "learning_rate": 0.08028991934728581,
      "loss": 0.0,
      "num_input_tokens_seen": 14857696,
      "step": 26155
    },
    {
      "epoch": 458.95575221238937,
      "grad_norm": 2.5387524971165476e-08,
      "learning_rate": 0.0802377673892585,
      "loss": 0.0,
      "num_input_tokens_seen": 14860256,
      "step": 26160
    },
    {
      "epoch": 459.0353982300885,
      "grad_norm": 1.1503896502063071e-08,
      "learning_rate": 0.0801856261894441,
      "loss": 0.0,
      "num_input_tokens_seen": 14862496,
      "step": 26165
    },
    {
      "epoch": 459.12389380530976,
      "grad_norm": 1.0334208155882152e-08,
      "learning_rate": 0.08013349575588354,
      "loss": 0.0,
      "num_input_tokens_seen": 14865568,
      "step": 26170
    },
    {
      "epoch": 459.21238938053096,
      "grad_norm": 9.992651150980691e-09,
      "learning_rate": 0.08008137609661586,
      "loss": 0.0,
      "num_input_tokens_seen": 14868448,
      "step": 26175
    },
    {
      "epoch": 459.3008849557522,
      "grad_norm": 5.888314902335878e-08,
      "learning_rate": 0.08002926721967872,
      "loss": 0.0,
      "num_input_tokens_seen": 14871296,
      "step": 26180
    },
    {
      "epoch": 459.3893805309734,
      "grad_norm": 6.226792947927606e-08,
      "learning_rate": 0.07997716913310782,
      "loss": 0.0,
      "num_input_tokens_seen": 14873776,
      "step": 26185
    },
    {
      "epoch": 459.4778761061947,
      "grad_norm": 1.9160124153927427e-08,
      "learning_rate": 0.07992508184493745,
      "loss": 0.0,
      "num_input_tokens_seen": 14876544,
      "step": 26190
    },
    {
      "epoch": 459.56637168141594,
      "grad_norm": 1.3366991957752816e-08,
      "learning_rate": 0.07987300536320001,
      "loss": 0.0,
      "num_input_tokens_seen": 14879248,
      "step": 26195
    },
    {
      "epoch": 459.65486725663715,
      "grad_norm": 2.0903412334405402e-08,
      "learning_rate": 0.07982093969592649,
      "loss": 0.0,
      "num_input_tokens_seen": 14882048,
      "step": 26200
    },
    {
      "epoch": 459.65486725663715,
      "eval_loss": 0.7370618581771851,
      "eval_runtime": 0.9384,
      "eval_samples_per_second": 26.642,
      "eval_steps_per_second": 13.854,
      "num_input_tokens_seen": 14882048,
      "step": 26200
    },
    {
      "epoch": 459.7433628318584,
      "grad_norm": 1.832665397216715e-08,
      "learning_rate": 0.07976888485114592,
      "loss": 0.0,
      "num_input_tokens_seen": 14885712,
      "step": 26205
    },
    {
      "epoch": 459.83185840707966,
      "grad_norm": 2.630437556661036e-08,
      "learning_rate": 0.07971684083688595,
      "loss": 0.0,
      "num_input_tokens_seen": 14888896,
      "step": 26210
    },
    {
      "epoch": 459.92035398230087,
      "grad_norm": 1.0203955902454709e-08,
      "learning_rate": 0.0796648076611723,
      "loss": 0.0,
      "num_input_tokens_seen": 14891792,
      "step": 26215
    },
    {
      "epoch": 460.0,
      "grad_norm": 3.76410191904597e-09,
      "learning_rate": 0.07961278533202922,
      "loss": 0.0,
      "num_input_tokens_seen": 14893968,
      "step": 26220
    },
    {
      "epoch": 460.08849557522126,
      "grad_norm": 2.9959412728430834e-08,
      "learning_rate": 0.07956077385747919,
      "loss": 0.0,
      "num_input_tokens_seen": 14897008,
      "step": 26225
    },
    {
      "epoch": 460.17699115044246,
      "grad_norm": 1.8504596965840392e-08,
      "learning_rate": 0.079508773245543,
      "loss": 0.0,
      "num_input_tokens_seen": 14899680,
      "step": 26230
    },
    {
      "epoch": 460.2654867256637,
      "grad_norm": 1.532199611631313e-08,
      "learning_rate": 0.07945678350423982,
      "loss": 0.0,
      "num_input_tokens_seen": 14902800,
      "step": 26235
    },
    {
      "epoch": 460.353982300885,
      "grad_norm": 2.5165359573975365e-08,
      "learning_rate": 0.07940480464158717,
      "loss": 0.0,
      "num_input_tokens_seen": 14905440,
      "step": 26240
    },
    {
      "epoch": 460.4424778761062,
      "grad_norm": 3.718771068861315e-08,
      "learning_rate": 0.07935283666560076,
      "loss": 0.0,
      "num_input_tokens_seen": 14908336,
      "step": 26245
    },
    {
      "epoch": 460.53097345132744,
      "grad_norm": 3.408083770750636e-08,
      "learning_rate": 0.07930087958429478,
      "loss": 0.0,
      "num_input_tokens_seen": 14910976,
      "step": 26250
    },
    {
      "epoch": 460.6194690265487,
      "grad_norm": 3.698157158282811e-08,
      "learning_rate": 0.07924893340568159,
      "loss": 0.0,
      "num_input_tokens_seen": 14913952,
      "step": 26255
    },
    {
      "epoch": 460.7079646017699,
      "grad_norm": 7.631489218340448e-09,
      "learning_rate": 0.07919699813777205,
      "loss": 0.0,
      "num_input_tokens_seen": 14917056,
      "step": 26260
    },
    {
      "epoch": 460.79646017699116,
      "grad_norm": 2.438881274713367e-08,
      "learning_rate": 0.07914507378857515,
      "loss": 0.0,
      "num_input_tokens_seen": 14919856,
      "step": 26265
    },
    {
      "epoch": 460.88495575221236,
      "grad_norm": 1.0818485662866806e-08,
      "learning_rate": 0.07909316036609822,
      "loss": 0.0,
      "num_input_tokens_seen": 14922096,
      "step": 26270
    },
    {
      "epoch": 460.9734513274336,
      "grad_norm": 2.102343543697316e-08,
      "learning_rate": 0.07904125787834704,
      "loss": 0.0,
      "num_input_tokens_seen": 14925632,
      "step": 26275
    },
    {
      "epoch": 461.05309734513276,
      "grad_norm": 2.0087298935322906e-08,
      "learning_rate": 0.07898936633332569,
      "loss": 0.0,
      "num_input_tokens_seen": 14927848,
      "step": 26280
    },
    {
      "epoch": 461.14159292035396,
      "grad_norm": 1.7843444055642976e-08,
      "learning_rate": 0.07893748573903635,
      "loss": 0.0,
      "num_input_tokens_seen": 14930584,
      "step": 26285
    },
    {
      "epoch": 461.2300884955752,
      "grad_norm": 1.0353384816141897e-08,
      "learning_rate": 0.0788856161034798,
      "loss": 0.0,
      "num_input_tokens_seen": 14933784,
      "step": 26290
    },
    {
      "epoch": 461.3185840707965,
      "grad_norm": 2.6429315624909577e-08,
      "learning_rate": 0.07883375743465487,
      "loss": 0.0,
      "num_input_tokens_seen": 14936216,
      "step": 26295
    },
    {
      "epoch": 461.4070796460177,
      "grad_norm": 1.3660075737220723e-08,
      "learning_rate": 0.07878190974055888,
      "loss": 0.0,
      "num_input_tokens_seen": 14939304,
      "step": 26300
    },
    {
      "epoch": 461.49557522123894,
      "grad_norm": 2.382215491536499e-08,
      "learning_rate": 0.07873007302918746,
      "loss": 0.0,
      "num_input_tokens_seen": 14942392,
      "step": 26305
    },
    {
      "epoch": 461.5840707964602,
      "grad_norm": 1.3382996932875812e-08,
      "learning_rate": 0.07867824730853433,
      "loss": 0.0,
      "num_input_tokens_seen": 14945288,
      "step": 26310
    },
    {
      "epoch": 461.6725663716814,
      "grad_norm": 1.7320793688213598e-08,
      "learning_rate": 0.07862643258659176,
      "loss": 0.0,
      "num_input_tokens_seen": 14948376,
      "step": 26315
    },
    {
      "epoch": 461.76106194690266,
      "grad_norm": 1.594868948018302e-08,
      "learning_rate": 0.07857462887135026,
      "loss": 0.0,
      "num_input_tokens_seen": 14951320,
      "step": 26320
    },
    {
      "epoch": 461.8495575221239,
      "grad_norm": 1.8506549182006893e-08,
      "learning_rate": 0.0785228361707986,
      "loss": 0.0,
      "num_input_tokens_seen": 14954360,
      "step": 26325
    },
    {
      "epoch": 461.9380530973451,
      "grad_norm": 7.034090199198317e-09,
      "learning_rate": 0.07847105449292378,
      "loss": 0.0,
      "num_input_tokens_seen": 14956952,
      "step": 26330
    },
    {
      "epoch": 462.01769911504425,
      "grad_norm": 2.4126149966718913e-08,
      "learning_rate": 0.0784192838457113,
      "loss": 0.0,
      "num_input_tokens_seen": 14959064,
      "step": 26335
    },
    {
      "epoch": 462.1061946902655,
      "grad_norm": 1.0865077726407435e-08,
      "learning_rate": 0.07836752423714473,
      "loss": 0.0,
      "num_input_tokens_seen": 14961480,
      "step": 26340
    },
    {
      "epoch": 462.1946902654867,
      "grad_norm": 1.7342996372349262e-08,
      "learning_rate": 0.07831577567520616,
      "loss": 0.0,
      "num_input_tokens_seen": 14964344,
      "step": 26345
    },
    {
      "epoch": 462.283185840708,
      "grad_norm": 1.8700065496091156e-08,
      "learning_rate": 0.07826403816787579,
      "loss": 0.0,
      "num_input_tokens_seen": 14966744,
      "step": 26350
    },
    {
      "epoch": 462.37168141592923,
      "grad_norm": 1.0280844620069729e-08,
      "learning_rate": 0.0782123117231322,
      "loss": 0.0,
      "num_input_tokens_seen": 14970328,
      "step": 26355
    },
    {
      "epoch": 462.46017699115043,
      "grad_norm": 1.3751360050662242e-08,
      "learning_rate": 0.07816059634895237,
      "loss": 0.0,
      "num_input_tokens_seen": 14973160,
      "step": 26360
    },
    {
      "epoch": 462.5486725663717,
      "grad_norm": 2.022829548309346e-08,
      "learning_rate": 0.0781088920533113,
      "loss": 0.0,
      "num_input_tokens_seen": 14976136,
      "step": 26365
    },
    {
      "epoch": 462.6371681415929,
      "grad_norm": 1.3748574723138063e-08,
      "learning_rate": 0.07805719884418257,
      "loss": 0.0,
      "num_input_tokens_seen": 14978712,
      "step": 26370
    },
    {
      "epoch": 462.72566371681415,
      "grad_norm": 2.3566180118450575e-08,
      "learning_rate": 0.07800551672953779,
      "loss": 0.0,
      "num_input_tokens_seen": 14981528,
      "step": 26375
    },
    {
      "epoch": 462.8141592920354,
      "grad_norm": 1.177709840760599e-08,
      "learning_rate": 0.07795384571734709,
      "loss": 0.0,
      "num_input_tokens_seen": 14984488,
      "step": 26380
    },
    {
      "epoch": 462.9026548672566,
      "grad_norm": 1.5148618359717148e-08,
      "learning_rate": 0.07790218581557883,
      "loss": 0.0,
      "num_input_tokens_seen": 14987208,
      "step": 26385
    },
    {
      "epoch": 462.9911504424779,
      "grad_norm": 1.6355846454985112e-08,
      "learning_rate": 0.07785053703219949,
      "loss": 0.0,
      "num_input_tokens_seen": 14990504,
      "step": 26390
    },
    {
      "epoch": 463.070796460177,
      "grad_norm": 9.710634074622249e-09,
      "learning_rate": 0.07779889937517409,
      "loss": 0.0,
      "num_input_tokens_seen": 14992888,
      "step": 26395
    },
    {
      "epoch": 463.1592920353982,
      "grad_norm": 4.3124899207214185e-08,
      "learning_rate": 0.0777472728524657,
      "loss": 0.0,
      "num_input_tokens_seen": 14996008,
      "step": 26400
    },
    {
      "epoch": 463.1592920353982,
      "eval_loss": 0.755020022392273,
      "eval_runtime": 0.9399,
      "eval_samples_per_second": 26.599,
      "eval_steps_per_second": 13.831,
      "num_input_tokens_seen": 14996008,
      "step": 26400
    },
    {
      "epoch": 463.24778761061947,
      "grad_norm": 1.828844631290849e-08,
      "learning_rate": 0.07769565747203584,
      "loss": 0.0,
      "num_input_tokens_seen": 14998904,
      "step": 26405
    },
    {
      "epoch": 463.3362831858407,
      "grad_norm": 3.2458103760291124e-08,
      "learning_rate": 0.07764405324184427,
      "loss": 0.0,
      "num_input_tokens_seen": 15001768,
      "step": 26410
    },
    {
      "epoch": 463.42477876106193,
      "grad_norm": 1.8316868022338895e-08,
      "learning_rate": 0.07759246016984889,
      "loss": 0.0,
      "num_input_tokens_seen": 15004696,
      "step": 26415
    },
    {
      "epoch": 463.5132743362832,
      "grad_norm": 1.9511675830585773e-08,
      "learning_rate": 0.07754087826400609,
      "loss": 0.0,
      "num_input_tokens_seen": 15007400,
      "step": 26420
    },
    {
      "epoch": 463.60176991150445,
      "grad_norm": 2.8701963472599346e-08,
      "learning_rate": 0.0774893075322705,
      "loss": 0.0,
      "num_input_tokens_seen": 15010040,
      "step": 26425
    },
    {
      "epoch": 463.69026548672565,
      "grad_norm": 2.4164316769770267e-08,
      "learning_rate": 0.07743774798259484,
      "loss": 0.0,
      "num_input_tokens_seen": 15012792,
      "step": 26430
    },
    {
      "epoch": 463.7787610619469,
      "grad_norm": 9.449704130304326e-09,
      "learning_rate": 0.07738619962293032,
      "loss": 0.0,
      "num_input_tokens_seen": 15015944,
      "step": 26435
    },
    {
      "epoch": 463.86725663716817,
      "grad_norm": 1.4060736575061128e-08,
      "learning_rate": 0.0773346624612264,
      "loss": 0.0,
      "num_input_tokens_seen": 15018952,
      "step": 26440
    },
    {
      "epoch": 463.95575221238937,
      "grad_norm": 1.0691037388710356e-08,
      "learning_rate": 0.07728313650543066,
      "loss": 0.0,
      "num_input_tokens_seen": 15021608,
      "step": 26445
    },
    {
      "epoch": 464.0353982300885,
      "grad_norm": 1.4473712894869095e-08,
      "learning_rate": 0.07723162176348913,
      "loss": 0.0,
      "num_input_tokens_seen": 15024248,
      "step": 26450
    },
    {
      "epoch": 464.12389380530976,
      "grad_norm": 2.1448563813919463e-08,
      "learning_rate": 0.07718011824334593,
      "loss": 0.0,
      "num_input_tokens_seen": 15027448,
      "step": 26455
    },
    {
      "epoch": 464.21238938053096,
      "grad_norm": 2.0578115211833392e-08,
      "learning_rate": 0.07712862595294363,
      "loss": 0.0,
      "num_input_tokens_seen": 15030568,
      "step": 26460
    },
    {
      "epoch": 464.3008849557522,
      "grad_norm": 1.1305594682653464e-08,
      "learning_rate": 0.07707714490022301,
      "loss": 0.0,
      "num_input_tokens_seen": 15032920,
      "step": 26465
    },
    {
      "epoch": 464.3893805309734,
      "grad_norm": 2.9708688842333686e-08,
      "learning_rate": 0.07702567509312298,
      "loss": 0.0,
      "num_input_tokens_seen": 15036120,
      "step": 26470
    },
    {
      "epoch": 464.4778761061947,
      "grad_norm": 2.9727674544233196e-08,
      "learning_rate": 0.07697421653958098,
      "loss": 0.0,
      "num_input_tokens_seen": 15039080,
      "step": 26475
    },
    {
      "epoch": 464.56637168141594,
      "grad_norm": 3.5697777178711476e-08,
      "learning_rate": 0.07692276924753247,
      "loss": 0.0,
      "num_input_tokens_seen": 15041496,
      "step": 26480
    },
    {
      "epoch": 464.65486725663715,
      "grad_norm": 3.0138807005641866e-08,
      "learning_rate": 0.07687133322491124,
      "loss": 0.0,
      "num_input_tokens_seen": 15044024,
      "step": 26485
    },
    {
      "epoch": 464.7433628318584,
      "grad_norm": 1.884195377499509e-08,
      "learning_rate": 0.07681990847964948,
      "loss": 0.0,
      "num_input_tokens_seen": 15046536,
      "step": 26490
    },
    {
      "epoch": 464.83185840707966,
      "grad_norm": 4.120063579193811e-08,
      "learning_rate": 0.0767684950196774,
      "loss": 0.0,
      "num_input_tokens_seen": 15050024,
      "step": 26495
    },
    {
      "epoch": 464.92035398230087,
      "grad_norm": 3.2930216775639565e-08,
      "learning_rate": 0.0767170928529237,
      "loss": 0.0,
      "num_input_tokens_seen": 15053384,
      "step": 26500
    },
    {
      "epoch": 465.0,
      "grad_norm": 8.251271665926652e-09,
      "learning_rate": 0.07666570198731526,
      "loss": 0.0,
      "num_input_tokens_seen": 15055576,
      "step": 26505
    },
    {
      "epoch": 465.08849557522126,
      "grad_norm": 2.2037793812046402e-08,
      "learning_rate": 0.07661432243077708,
      "loss": 0.0,
      "num_input_tokens_seen": 15058232,
      "step": 26510
    },
    {
      "epoch": 465.17699115044246,
      "grad_norm": 2.0737275008286815e-08,
      "learning_rate": 0.0765629541912326,
      "loss": 0.0,
      "num_input_tokens_seen": 15061512,
      "step": 26515
    },
    {
      "epoch": 465.2654867256637,
      "grad_norm": 2.7345794961775027e-08,
      "learning_rate": 0.07651159727660352,
      "loss": 0.0,
      "num_input_tokens_seen": 15064536,
      "step": 26520
    },
    {
      "epoch": 465.353982300885,
      "grad_norm": 1.7652560302394704e-08,
      "learning_rate": 0.07646025169480959,
      "loss": 0.0,
      "num_input_tokens_seen": 15067416,
      "step": 26525
    },
    {
      "epoch": 465.4424778761062,
      "grad_norm": 2.7957186787830324e-08,
      "learning_rate": 0.07640891745376908,
      "loss": 0.0,
      "num_input_tokens_seen": 15070200,
      "step": 26530
    },
    {
      "epoch": 465.53097345132744,
      "grad_norm": 2.381276686946876e-08,
      "learning_rate": 0.07635759456139822,
      "loss": 0.0,
      "num_input_tokens_seen": 15072808,
      "step": 26535
    },
    {
      "epoch": 465.6194690265487,
      "grad_norm": 5.04851875859913e-08,
      "learning_rate": 0.0763062830256118,
      "loss": 0.0,
      "num_input_tokens_seen": 15075800,
      "step": 26540
    },
    {
      "epoch": 465.7079646017699,
      "grad_norm": 2.7456808382453346e-08,
      "learning_rate": 0.07625498285432258,
      "loss": 0.0,
      "num_input_tokens_seen": 15078232,
      "step": 26545
    },
    {
      "epoch": 465.79646017699116,
      "grad_norm": 1.2724507669759078e-08,
      "learning_rate": 0.07620369405544176,
      "loss": 0.0,
      "num_input_tokens_seen": 15081096,
      "step": 26550
    },
    {
      "epoch": 465.88495575221236,
      "grad_norm": 6.733528401525746e-09,
      "learning_rate": 0.07615241663687868,
      "loss": 0.0,
      "num_input_tokens_seen": 15084616,
      "step": 26555
    },
    {
      "epoch": 465.9734513274336,
      "grad_norm": 3.3804905541501284e-08,
      "learning_rate": 0.07610115060654106,
      "loss": 0.0,
      "num_input_tokens_seen": 15087160,
      "step": 26560
    },
    {
      "epoch": 466.05309734513276,
      "grad_norm": 6.3425731333666135e-09,
      "learning_rate": 0.07604989597233458,
      "loss": 0.0,
      "num_input_tokens_seen": 15089240,
      "step": 26565
    },
    {
      "epoch": 466.14159292035396,
      "grad_norm": 1.1826374546330953e-08,
      "learning_rate": 0.07599865274216352,
      "loss": 0.0,
      "num_input_tokens_seen": 15092200,
      "step": 26570
    },
    {
      "epoch": 466.2300884955752,
      "grad_norm": 2.0637283881796975e-08,
      "learning_rate": 0.07594742092393013,
      "loss": 0.0,
      "num_input_tokens_seen": 15094808,
      "step": 26575
    },
    {
      "epoch": 466.3185840707965,
      "grad_norm": 2.0089132135581167e-08,
      "learning_rate": 0.07589620052553503,
      "loss": 0.0,
      "num_input_tokens_seen": 15097560,
      "step": 26580
    },
    {
      "epoch": 466.4070796460177,
      "grad_norm": 2.7731783092121987e-08,
      "learning_rate": 0.0758449915548771,
      "loss": 0.0,
      "num_input_tokens_seen": 15100824,
      "step": 26585
    },
    {
      "epoch": 466.49557522123894,
      "grad_norm": 3.522254132803937e-08,
      "learning_rate": 0.07579379401985332,
      "loss": 0.0,
      "num_input_tokens_seen": 15103368,
      "step": 26590
    },
    {
      "epoch": 466.5840707964602,
      "grad_norm": 8.521155336893571e-09,
      "learning_rate": 0.07574260792835905,
      "loss": 0.0,
      "num_input_tokens_seen": 15106504,
      "step": 26595
    },
    {
      "epoch": 466.6725663716814,
      "grad_norm": 1.8836969317703733e-08,
      "learning_rate": 0.07569143328828784,
      "loss": 0.0,
      "num_input_tokens_seen": 15109352,
      "step": 26600
    },
    {
      "epoch": 466.6725663716814,
      "eval_loss": 0.7279479503631592,
      "eval_runtime": 0.9243,
      "eval_samples_per_second": 27.046,
      "eval_steps_per_second": 14.064,
      "num_input_tokens_seen": 15109352,
      "step": 26600
    },
    {
      "epoch": 466.76106194690266,
      "grad_norm": 2.5071402731668968e-08,
      "learning_rate": 0.0756402701075314,
      "loss": 0.0,
      "num_input_tokens_seen": 15112392,
      "step": 26605
    },
    {
      "epoch": 466.8495575221239,
      "grad_norm": 2.4407542653648306e-08,
      "learning_rate": 0.07558911839397982,
      "loss": 0.0,
      "num_input_tokens_seen": 15115688,
      "step": 26610
    },
    {
      "epoch": 466.9380530973451,
      "grad_norm": 1.8362836584628894e-08,
      "learning_rate": 0.07553797815552123,
      "loss": 0.0,
      "num_input_tokens_seen": 15118104,
      "step": 26615
    },
    {
      "epoch": 467.01769911504425,
      "grad_norm": 1.4815571880433254e-08,
      "learning_rate": 0.07548684940004222,
      "loss": 0.0,
      "num_input_tokens_seen": 15120496,
      "step": 26620
    },
    {
      "epoch": 467.1061946902655,
      "grad_norm": 3.0804685025032086e-08,
      "learning_rate": 0.07543573213542744,
      "loss": 0.0,
      "num_input_tokens_seen": 15123328,
      "step": 26625
    },
    {
      "epoch": 467.1946902654867,
      "grad_norm": 1.3218560468430951e-08,
      "learning_rate": 0.0753846263695597,
      "loss": 0.0,
      "num_input_tokens_seen": 15125856,
      "step": 26630
    },
    {
      "epoch": 467.283185840708,
      "grad_norm": 1.424525208904015e-08,
      "learning_rate": 0.07533353211032029,
      "loss": 0.0,
      "num_input_tokens_seen": 15128512,
      "step": 26635
    },
    {
      "epoch": 467.37168141592923,
      "grad_norm": 2.451504244049829e-08,
      "learning_rate": 0.07528244936558857,
      "loss": 0.0,
      "num_input_tokens_seen": 15131520,
      "step": 26640
    },
    {
      "epoch": 467.46017699115043,
      "grad_norm": 2.2337729888022295e-08,
      "learning_rate": 0.07523137814324206,
      "loss": 0.0,
      "num_input_tokens_seen": 15134128,
      "step": 26645
    },
    {
      "epoch": 467.5486725663717,
      "grad_norm": 1.8173603066884425e-08,
      "learning_rate": 0.07518031845115672,
      "loss": 0.0,
      "num_input_tokens_seen": 15136816,
      "step": 26650
    },
    {
      "epoch": 467.6371681415929,
      "grad_norm": 1.2570815499657328e-08,
      "learning_rate": 0.07512927029720647,
      "loss": 0.0,
      "num_input_tokens_seen": 15140256,
      "step": 26655
    },
    {
      "epoch": 467.72566371681415,
      "grad_norm": 2.6296046229390413e-08,
      "learning_rate": 0.0750782336892636,
      "loss": 0.0,
      "num_input_tokens_seen": 15143248,
      "step": 26660
    },
    {
      "epoch": 467.8141592920354,
      "grad_norm": 2.4679909671476707e-08,
      "learning_rate": 0.0750272086351987,
      "loss": 0.0,
      "num_input_tokens_seen": 15146240,
      "step": 26665
    },
    {
      "epoch": 467.9026548672566,
      "grad_norm": 2.5128377600935892e-08,
      "learning_rate": 0.07497619514288031,
      "loss": 0.0,
      "num_input_tokens_seen": 15149520,
      "step": 26670
    },
    {
      "epoch": 467.9911504424779,
      "grad_norm": 2.746814153908872e-08,
      "learning_rate": 0.07492519322017545,
      "loss": 0.0,
      "num_input_tokens_seen": 15152272,
      "step": 26675
    },
    {
      "epoch": 468.070796460177,
      "grad_norm": 2.348684446928928e-08,
      "learning_rate": 0.0748742028749493,
      "loss": 0.0,
      "num_input_tokens_seen": 15155144,
      "step": 26680
    },
    {
      "epoch": 468.1592920353982,
      "grad_norm": 1.4589614849569443e-08,
      "learning_rate": 0.0748232241150651,
      "loss": 0.0,
      "num_input_tokens_seen": 15158280,
      "step": 26685
    },
    {
      "epoch": 468.24778761061947,
      "grad_norm": 2.7930594725944502e-08,
      "learning_rate": 0.07477225694838453,
      "loss": 0.0,
      "num_input_tokens_seen": 15161736,
      "step": 26690
    },
    {
      "epoch": 468.3362831858407,
      "grad_norm": 1.2642954239083792e-08,
      "learning_rate": 0.07472130138276731,
      "loss": 0.0,
      "num_input_tokens_seen": 15164072,
      "step": 26695
    },
    {
      "epoch": 468.42477876106193,
      "grad_norm": 1.5674876507887348e-08,
      "learning_rate": 0.07467035742607138,
      "loss": 0.0,
      "num_input_tokens_seen": 15167192,
      "step": 26700
    },
    {
      "epoch": 468.5132743362832,
      "grad_norm": 1.1083265860634128e-08,
      "learning_rate": 0.07461942508615303,
      "loss": 0.0,
      "num_input_tokens_seen": 15170264,
      "step": 26705
    },
    {
      "epoch": 468.60176991150445,
      "grad_norm": 1.2504946411695528e-08,
      "learning_rate": 0.07456850437086657,
      "loss": 0.0,
      "num_input_tokens_seen": 15173768,
      "step": 26710
    },
    {
      "epoch": 468.69026548672565,
      "grad_norm": 4.4026275958231054e-08,
      "learning_rate": 0.07451759528806468,
      "loss": 0.0,
      "num_input_tokens_seen": 15176584,
      "step": 26715
    },
    {
      "epoch": 468.7787610619469,
      "grad_norm": 2.2286126721837718e-08,
      "learning_rate": 0.0744666978455982,
      "loss": 0.0,
      "num_input_tokens_seen": 15179032,
      "step": 26720
    },
    {
      "epoch": 468.86725663716817,
      "grad_norm": 1.0850369491777201e-08,
      "learning_rate": 0.07441581205131609,
      "loss": 0.0,
      "num_input_tokens_seen": 15181624,
      "step": 26725
    },
    {
      "epoch": 468.95575221238937,
      "grad_norm": 6.022408882699892e-08,
      "learning_rate": 0.07436493791306566,
      "loss": 0.0,
      "num_input_tokens_seen": 15184088,
      "step": 26730
    },
    {
      "epoch": 469.0353982300885,
      "grad_norm": 2.8226319059854177e-08,
      "learning_rate": 0.07431407543869223,
      "loss": 0.0,
      "num_input_tokens_seen": 15186016,
      "step": 26735
    },
    {
      "epoch": 469.12389380530976,
      "grad_norm": 9.274656598279307e-09,
      "learning_rate": 0.0742632246360395,
      "loss": 0.0,
      "num_input_tokens_seen": 15188768,
      "step": 26740
    },
    {
      "epoch": 469.21238938053096,
      "grad_norm": 9.977712878139755e-09,
      "learning_rate": 0.07421238551294934,
      "loss": 0.0,
      "num_input_tokens_seen": 15191872,
      "step": 26745
    },
    {
      "epoch": 469.3008849557522,
      "grad_norm": 1.5472718217779402e-08,
      "learning_rate": 0.07416155807726171,
      "loss": 0.0,
      "num_input_tokens_seen": 15194592,
      "step": 26750
    },
    {
      "epoch": 469.3893805309734,
      "grad_norm": 2.7133379987276385e-08,
      "learning_rate": 0.07411074233681492,
      "loss": 0.0,
      "num_input_tokens_seen": 15197680,
      "step": 26755
    },
    {
      "epoch": 469.4778761061947,
      "grad_norm": 1.0527031690799049e-08,
      "learning_rate": 0.07405993829944528,
      "loss": 0.0,
      "num_input_tokens_seen": 15200560,
      "step": 26760
    },
    {
      "epoch": 469.56637168141594,
      "grad_norm": 2.1121369542242974e-08,
      "learning_rate": 0.07400914597298755,
      "loss": 0.0,
      "num_input_tokens_seen": 15203216,
      "step": 26765
    },
    {
      "epoch": 469.65486725663715,
      "grad_norm": 1.1805872723869015e-08,
      "learning_rate": 0.07395836536527445,
      "loss": 0.0,
      "num_input_tokens_seen": 15205920,
      "step": 26770
    },
    {
      "epoch": 469.7433628318584,
      "grad_norm": 2.970399926027767e-08,
      "learning_rate": 0.07390759648413696,
      "loss": 0.0,
      "num_input_tokens_seen": 15208880,
      "step": 26775
    },
    {
      "epoch": 469.83185840707966,
      "grad_norm": 2.852095093430762e-08,
      "learning_rate": 0.07385683933740435,
      "loss": 0.0,
      "num_input_tokens_seen": 15211968,
      "step": 26780
    },
    {
      "epoch": 469.92035398230087,
      "grad_norm": 2.3176815133751916e-08,
      "learning_rate": 0.07380609393290402,
      "loss": 0.0,
      "num_input_tokens_seen": 15215120,
      "step": 26785
    },
    {
      "epoch": 470.0,
      "grad_norm": 2.9119251010456537e-08,
      "learning_rate": 0.07375536027846147,
      "loss": 0.0,
      "num_input_tokens_seen": 15217640,
      "step": 26790
    },
    {
      "epoch": 470.08849557522126,
      "grad_norm": 3.951613436470325e-08,
      "learning_rate": 0.07370463838190057,
      "loss": 0.0,
      "num_input_tokens_seen": 15220568,
      "step": 26795
    },
    {
      "epoch": 470.17699115044246,
      "grad_norm": 1.4186101182644961e-08,
      "learning_rate": 0.07365392825104317,
      "loss": 0.0,
      "num_input_tokens_seen": 15223592,
      "step": 26800
    },
    {
      "epoch": 470.17699115044246,
      "eval_loss": 0.738929033279419,
      "eval_runtime": 0.9374,
      "eval_samples_per_second": 26.67,
      "eval_steps_per_second": 13.869,
      "num_input_tokens_seen": 15223592,
      "step": 26800
    },
    {
      "epoch": 470.2654867256637,
      "grad_norm": 8.684468255637512e-09,
      "learning_rate": 0.07360322989370945,
      "loss": 0.0,
      "num_input_tokens_seen": 15226888,
      "step": 26805
    },
    {
      "epoch": 470.353982300885,
      "grad_norm": 2.8926530504236325e-08,
      "learning_rate": 0.07355254331771781,
      "loss": 0.0,
      "num_input_tokens_seen": 15229496,
      "step": 26810
    },
    {
      "epoch": 470.4424778761062,
      "grad_norm": 1.4983911000854278e-08,
      "learning_rate": 0.07350186853088461,
      "loss": 0.0,
      "num_input_tokens_seen": 15232088,
      "step": 26815
    },
    {
      "epoch": 470.53097345132744,
      "grad_norm": 2.0816749213281582e-08,
      "learning_rate": 0.07345120554102462,
      "loss": 0.0,
      "num_input_tokens_seen": 15235576,
      "step": 26820
    },
    {
      "epoch": 470.6194690265487,
      "grad_norm": 1.5178038381691294e-08,
      "learning_rate": 0.07340055435595079,
      "loss": 0.0,
      "num_input_tokens_seen": 15238120,
      "step": 26825
    },
    {
      "epoch": 470.7079646017699,
      "grad_norm": 3.084651822859996e-08,
      "learning_rate": 0.07334991498347401,
      "loss": 0.0,
      "num_input_tokens_seen": 15241112,
      "step": 26830
    },
    {
      "epoch": 470.79646017699116,
      "grad_norm": 1.7046366096451493e-08,
      "learning_rate": 0.07329928743140365,
      "loss": 0.0,
      "num_input_tokens_seen": 15243688,
      "step": 26835
    },
    {
      "epoch": 470.88495575221236,
      "grad_norm": 2.3278268201920582e-08,
      "learning_rate": 0.07324867170754705,
      "loss": 0.0,
      "num_input_tokens_seen": 15246424,
      "step": 26840
    },
    {
      "epoch": 470.9734513274336,
      "grad_norm": 3.077843047094575e-08,
      "learning_rate": 0.07319806781970974,
      "loss": 0.0,
      "num_input_tokens_seen": 15249320,
      "step": 26845
    },
    {
      "epoch": 471.05309734513276,
      "grad_norm": 2.55559182704701e-08,
      "learning_rate": 0.07314747577569555,
      "loss": 0.0,
      "num_input_tokens_seen": 15251360,
      "step": 26850
    },
    {
      "epoch": 471.14159292035396,
      "grad_norm": 4.134213327233738e-08,
      "learning_rate": 0.07309689558330636,
      "loss": 0.0,
      "num_input_tokens_seen": 15254256,
      "step": 26855
    },
    {
      "epoch": 471.2300884955752,
      "grad_norm": 6.970477528511765e-09,
      "learning_rate": 0.0730463272503423,
      "loss": 0.0,
      "num_input_tokens_seen": 15257424,
      "step": 26860
    },
    {
      "epoch": 471.3185840707965,
      "grad_norm": 9.620774399365928e-09,
      "learning_rate": 0.07299577078460168,
      "loss": 0.0,
      "num_input_tokens_seen": 15260464,
      "step": 26865
    },
    {
      "epoch": 471.4070796460177,
      "grad_norm": 2.5922693325242108e-08,
      "learning_rate": 0.07294522619388083,
      "loss": 0.0,
      "num_input_tokens_seen": 15263712,
      "step": 26870
    },
    {
      "epoch": 471.49557522123894,
      "grad_norm": 8.512454741094189e-09,
      "learning_rate": 0.07289469348597452,
      "loss": 0.0,
      "num_input_tokens_seen": 15266256,
      "step": 26875
    },
    {
      "epoch": 471.5840707964602,
      "grad_norm": 1.3382488894819744e-08,
      "learning_rate": 0.07284417266867535,
      "loss": 0.0,
      "num_input_tokens_seen": 15269312,
      "step": 26880
    },
    {
      "epoch": 471.6725663716814,
      "grad_norm": 9.108938492374818e-09,
      "learning_rate": 0.07279366374977439,
      "loss": 0.0,
      "num_input_tokens_seen": 15272032,
      "step": 26885
    },
    {
      "epoch": 471.76106194690266,
      "grad_norm": 7.0129599905044415e-09,
      "learning_rate": 0.07274316673706074,
      "loss": 0.0,
      "num_input_tokens_seen": 15274912,
      "step": 26890
    },
    {
      "epoch": 471.8495575221239,
      "grad_norm": 3.0220579816386817e-08,
      "learning_rate": 0.07269268163832161,
      "loss": 0.0,
      "num_input_tokens_seen": 15277952,
      "step": 26895
    },
    {
      "epoch": 471.9380530973451,
      "grad_norm": 2.4196570080903257e-08,
      "learning_rate": 0.07264220846134248,
      "loss": 0.0,
      "num_input_tokens_seen": 15280512,
      "step": 26900
    },
    {
      "epoch": 472.01769911504425,
      "grad_norm": 1.8187554573501075e-08,
      "learning_rate": 0.07259174721390699,
      "loss": 0.0,
      "num_input_tokens_seen": 15282872,
      "step": 26905
    },
    {
      "epoch": 472.1061946902655,
      "grad_norm": 1.2322418641019794e-08,
      "learning_rate": 0.07254129790379686,
      "loss": 0.0,
      "num_input_tokens_seen": 15286456,
      "step": 26910
    },
    {
      "epoch": 472.1946902654867,
      "grad_norm": 1.468863786158181e-08,
      "learning_rate": 0.072490860538792,
      "loss": 0.0,
      "num_input_tokens_seen": 15289368,
      "step": 26915
    },
    {
      "epoch": 472.283185840708,
      "grad_norm": 1.8187281014547807e-08,
      "learning_rate": 0.07244043512667042,
      "loss": 0.0,
      "num_input_tokens_seen": 15292488,
      "step": 26920
    },
    {
      "epoch": 472.37168141592923,
      "grad_norm": 1.2780495772801714e-08,
      "learning_rate": 0.07239002167520843,
      "loss": 0.0,
      "num_input_tokens_seen": 15294952,
      "step": 26925
    },
    {
      "epoch": 472.46017699115043,
      "grad_norm": 2.922030439833634e-08,
      "learning_rate": 0.07233962019218045,
      "loss": 0.0,
      "num_input_tokens_seen": 15297928,
      "step": 26930
    },
    {
      "epoch": 472.5486725663717,
      "grad_norm": 5.622412313499581e-09,
      "learning_rate": 0.07228923068535892,
      "loss": 0.0,
      "num_input_tokens_seen": 15300760,
      "step": 26935
    },
    {
      "epoch": 472.6371681415929,
      "grad_norm": 3.978725970910091e-08,
      "learning_rate": 0.0722388531625146,
      "loss": 0.0,
      "num_input_tokens_seen": 15303640,
      "step": 26940
    },
    {
      "epoch": 472.72566371681415,
      "grad_norm": 1.4614865762041518e-08,
      "learning_rate": 0.07218848763141639,
      "loss": 0.0,
      "num_input_tokens_seen": 15306424,
      "step": 26945
    },
    {
      "epoch": 472.8141592920354,
      "grad_norm": 3.219632560558239e-08,
      "learning_rate": 0.07213813409983118,
      "loss": 0.0,
      "num_input_tokens_seen": 15309560,
      "step": 26950
    },
    {
      "epoch": 472.9026548672566,
      "grad_norm": 1.4026952932510994e-08,
      "learning_rate": 0.0720877925755242,
      "loss": 0.0,
      "num_input_tokens_seen": 15312440,
      "step": 26955
    },
    {
      "epoch": 472.9911504424779,
      "grad_norm": 3.1585955184709746e-08,
      "learning_rate": 0.07203746306625866,
      "loss": 0.0,
      "num_input_tokens_seen": 15314872,
      "step": 26960
    },
    {
      "epoch": 473.070796460177,
      "grad_norm": 2.6663244057090196e-08,
      "learning_rate": 0.07198714557979606,
      "loss": 0.0,
      "num_input_tokens_seen": 15317736,
      "step": 26965
    },
    {
      "epoch": 473.1592920353982,
      "grad_norm": 2.5832925132363016e-08,
      "learning_rate": 0.07193684012389602,
      "loss": 0.0,
      "num_input_tokens_seen": 15320888,
      "step": 26970
    },
    {
      "epoch": 473.24778761061947,
      "grad_norm": 1.977051944379582e-08,
      "learning_rate": 0.07188654670631621,
      "loss": 0.0,
      "num_input_tokens_seen": 15323944,
      "step": 26975
    },
    {
      "epoch": 473.3362831858407,
      "grad_norm": 1.0030976937969172e-08,
      "learning_rate": 0.07183626533481258,
      "loss": 0.0,
      "num_input_tokens_seen": 15327048,
      "step": 26980
    },
    {
      "epoch": 473.42477876106193,
      "grad_norm": 8.7425551242859e-09,
      "learning_rate": 0.07178599601713909,
      "loss": 0.0,
      "num_input_tokens_seen": 15329816,
      "step": 26985
    },
    {
      "epoch": 473.5132743362832,
      "grad_norm": 2.6188081037048505e-08,
      "learning_rate": 0.07173573876104786,
      "loss": 0.0,
      "num_input_tokens_seen": 15332728,
      "step": 26990
    },
    {
      "epoch": 473.60176991150445,
      "grad_norm": 1.183982867303257e-08,
      "learning_rate": 0.0716854935742893,
      "loss": 0.0,
      "num_input_tokens_seen": 15335192,
      "step": 26995
    },
    {
      "epoch": 473.69026548672565,
      "grad_norm": 1.4243016543957765e-08,
      "learning_rate": 0.07163526046461174,
      "loss": 0.0,
      "num_input_tokens_seen": 15338072,
      "step": 27000
    },
    {
      "epoch": 473.69026548672565,
      "eval_loss": 0.7722395658493042,
      "eval_runtime": 0.9415,
      "eval_samples_per_second": 26.552,
      "eval_steps_per_second": 13.807,
      "num_input_tokens_seen": 15338072,
      "step": 27000
    },
    {
      "epoch": 473.7787610619469,
      "grad_norm": 1.4597269171190419e-08,
      "learning_rate": 0.07158503943976181,
      "loss": 0.0,
      "num_input_tokens_seen": 15341128,
      "step": 27005
    },
    {
      "epoch": 473.86725663716817,
      "grad_norm": 2.5061135389137235e-08,
      "learning_rate": 0.07153483050748427,
      "loss": 0.0,
      "num_input_tokens_seen": 15343592,
      "step": 27010
    },
    {
      "epoch": 473.95575221238937,
      "grad_norm": 1.848863995235206e-08,
      "learning_rate": 0.07148463367552188,
      "loss": 0.0,
      "num_input_tokens_seen": 15346248,
      "step": 27015
    },
    {
      "epoch": 474.0353982300885,
      "grad_norm": 2.8274918406623328e-08,
      "learning_rate": 0.07143444895161565,
      "loss": 0.0,
      "num_input_tokens_seen": 15348456,
      "step": 27020
    },
    {
      "epoch": 474.12389380530976,
      "grad_norm": 3.677578419569727e-08,
      "learning_rate": 0.07138427634350476,
      "loss": 0.0,
      "num_input_tokens_seen": 15351736,
      "step": 27025
    },
    {
      "epoch": 474.21238938053096,
      "grad_norm": 1.264080129459444e-08,
      "learning_rate": 0.07133411585892636,
      "loss": 0.0,
      "num_input_tokens_seen": 15354824,
      "step": 27030
    },
    {
      "epoch": 474.3008849557522,
      "grad_norm": 3.4439036511457743e-08,
      "learning_rate": 0.07128396750561593,
      "loss": 0.0,
      "num_input_tokens_seen": 15357480,
      "step": 27035
    },
    {
      "epoch": 474.3893805309734,
      "grad_norm": 3.204685938840157e-08,
      "learning_rate": 0.07123383129130685,
      "loss": 0.0,
      "num_input_tokens_seen": 15360664,
      "step": 27040
    },
    {
      "epoch": 474.4778761061947,
      "grad_norm": 2.138198951229242e-08,
      "learning_rate": 0.07118370722373084,
      "loss": 0.0,
      "num_input_tokens_seen": 15363368,
      "step": 27045
    },
    {
      "epoch": 474.56637168141594,
      "grad_norm": 1.7911188976427184e-08,
      "learning_rate": 0.07113359531061769,
      "loss": 0.0,
      "num_input_tokens_seen": 15366024,
      "step": 27050
    },
    {
      "epoch": 474.65486725663715,
      "grad_norm": 4.1598248401442106e-08,
      "learning_rate": 0.07108349555969525,
      "loss": 0.0,
      "num_input_tokens_seen": 15368904,
      "step": 27055
    },
    {
      "epoch": 474.7433628318584,
      "grad_norm": 2.206866156484466e-08,
      "learning_rate": 0.07103340797868944,
      "loss": 0.0,
      "num_input_tokens_seen": 15371720,
      "step": 27060
    },
    {
      "epoch": 474.83185840707966,
      "grad_norm": 2.886568140070267e-08,
      "learning_rate": 0.07098333257532453,
      "loss": 0.0,
      "num_input_tokens_seen": 15375016,
      "step": 27065
    },
    {
      "epoch": 474.92035398230087,
      "grad_norm": 4.514050289117222e-08,
      "learning_rate": 0.07093326935732269,
      "loss": 0.0,
      "num_input_tokens_seen": 15377832,
      "step": 27070
    },
    {
      "epoch": 475.0,
      "grad_norm": 6.404533792192524e-08,
      "learning_rate": 0.0708832183324044,
      "loss": 0.0,
      "num_input_tokens_seen": 15379912,
      "step": 27075
    },
    {
      "epoch": 475.08849557522126,
      "grad_norm": 1.5245694484633532e-08,
      "learning_rate": 0.07083317950828799,
      "loss": 0.0,
      "num_input_tokens_seen": 15382568,
      "step": 27080
    },
    {
      "epoch": 475.17699115044246,
      "grad_norm": 1.146172134980361e-08,
      "learning_rate": 0.0707831528926902,
      "loss": 0.0,
      "num_input_tokens_seen": 15385384,
      "step": 27085
    },
    {
      "epoch": 475.2654867256637,
      "grad_norm": 5.0678274021720426e-08,
      "learning_rate": 0.07073313849332578,
      "loss": 0.0,
      "num_input_tokens_seen": 15388488,
      "step": 27090
    },
    {
      "epoch": 475.353982300885,
      "grad_norm": 2.6282231502250397e-08,
      "learning_rate": 0.07068313631790749,
      "loss": 0.0,
      "num_input_tokens_seen": 15391576,
      "step": 27095
    },
    {
      "epoch": 475.4424778761062,
      "grad_norm": 1.5744435089004583e-08,
      "learning_rate": 0.07063314637414632,
      "loss": 0.0,
      "num_input_tokens_seen": 15394008,
      "step": 27100
    },
    {
      "epoch": 475.53097345132744,
      "grad_norm": 1.731838850105305e-08,
      "learning_rate": 0.07058316866975144,
      "loss": 0.0,
      "num_input_tokens_seen": 15396680,
      "step": 27105
    },
    {
      "epoch": 475.6194690265487,
      "grad_norm": 2.116708408550494e-08,
      "learning_rate": 0.0705332032124299,
      "loss": 0.0,
      "num_input_tokens_seen": 15399592,
      "step": 27110
    },
    {
      "epoch": 475.7079646017699,
      "grad_norm": 1.059663201630201e-08,
      "learning_rate": 0.0704832500098871,
      "loss": 0.0,
      "num_input_tokens_seen": 15402920,
      "step": 27115
    },
    {
      "epoch": 475.79646017699116,
      "grad_norm": 2.0187647109537465e-08,
      "learning_rate": 0.07043330906982641,
      "loss": 0.0,
      "num_input_tokens_seen": 15405896,
      "step": 27120
    },
    {
      "epoch": 475.88495575221236,
      "grad_norm": 9.376956988660368e-09,
      "learning_rate": 0.07038338039994936,
      "loss": 0.0,
      "num_input_tokens_seen": 15408936,
      "step": 27125
    },
    {
      "epoch": 475.9734513274336,
      "grad_norm": 2.42500473035534e-08,
      "learning_rate": 0.07033346400795562,
      "loss": 0.0,
      "num_input_tokens_seen": 15411768,
      "step": 27130
    },
    {
      "epoch": 476.05309734513276,
      "grad_norm": 1.4219494026690427e-08,
      "learning_rate": 0.07028355990154282,
      "loss": 0.0,
      "num_input_tokens_seen": 15414264,
      "step": 27135
    },
    {
      "epoch": 476.14159292035396,
      "grad_norm": 1.4152766070196776e-08,
      "learning_rate": 0.07023366808840685,
      "loss": 0.0,
      "num_input_tokens_seen": 15417240,
      "step": 27140
    },
    {
      "epoch": 476.2300884955752,
      "grad_norm": 2.3683616845460165e-08,
      "learning_rate": 0.07018378857624172,
      "loss": 0.0,
      "num_input_tokens_seen": 15420360,
      "step": 27145
    },
    {
      "epoch": 476.3185840707965,
      "grad_norm": 3.933042691528499e-08,
      "learning_rate": 0.0701339213727394,
      "loss": 0.0,
      "num_input_tokens_seen": 15423784,
      "step": 27150
    },
    {
      "epoch": 476.4070796460177,
      "grad_norm": 2.159858603079101e-08,
      "learning_rate": 0.07008406648559008,
      "loss": 0.0,
      "num_input_tokens_seen": 15426936,
      "step": 27155
    },
    {
      "epoch": 476.49557522123894,
      "grad_norm": 4.904070394218252e-08,
      "learning_rate": 0.07003422392248196,
      "loss": 0.0,
      "num_input_tokens_seen": 15429864,
      "step": 27160
    },
    {
      "epoch": 476.5840707964602,
      "grad_norm": 3.336716503099524e-08,
      "learning_rate": 0.06998439369110142,
      "loss": 0.0,
      "num_input_tokens_seen": 15432792,
      "step": 27165
    },
    {
      "epoch": 476.6725663716814,
      "grad_norm": 1.5013210230563345e-08,
      "learning_rate": 0.06993457579913295,
      "loss": 0.0,
      "num_input_tokens_seen": 15436024,
      "step": 27170
    },
    {
      "epoch": 476.76106194690266,
      "grad_norm": 2.735541571041722e-08,
      "learning_rate": 0.06988477025425903,
      "loss": 0.0,
      "num_input_tokens_seen": 15438648,
      "step": 27175
    },
    {
      "epoch": 476.8495575221239,
      "grad_norm": 2.179154989789822e-08,
      "learning_rate": 0.06983497706416032,
      "loss": 0.0,
      "num_input_tokens_seen": 15441224,
      "step": 27180
    },
    {
      "epoch": 476.9380530973451,
      "grad_norm": 2.5198746200771893e-08,
      "learning_rate": 0.0697851962365156,
      "loss": 0.0,
      "num_input_tokens_seen": 15443704,
      "step": 27185
    },
    {
      "epoch": 477.01769911504425,
      "grad_norm": 3.6157786098556244e-08,
      "learning_rate": 0.06973542777900163,
      "loss": 0.0,
      "num_input_tokens_seen": 15445760,
      "step": 27190
    },
    {
      "epoch": 477.1061946902655,
      "grad_norm": 1.7848517330776303e-08,
      "learning_rate": 0.06968567169929342,
      "loss": 0.0,
      "num_input_tokens_seen": 15448704,
      "step": 27195
    },
    {
      "epoch": 477.1946902654867,
      "grad_norm": 1.987315023654901e-08,
      "learning_rate": 0.06963592800506392,
      "loss": 0.0,
      "num_input_tokens_seen": 15451312,
      "step": 27200
    },
    {
      "epoch": 477.1946902654867,
      "eval_loss": 0.754585862159729,
      "eval_runtime": 0.9422,
      "eval_samples_per_second": 26.535,
      "eval_steps_per_second": 13.798,
      "num_input_tokens_seen": 15451312,
      "step": 27200
    },
    {
      "epoch": 477.283185840708,
      "grad_norm": 1.98118659255897e-08,
      "learning_rate": 0.06958619670398417,
      "loss": 0.0,
      "num_input_tokens_seen": 15454432,
      "step": 27205
    },
    {
      "epoch": 477.37168141592923,
      "grad_norm": 3.153532546207316e-08,
      "learning_rate": 0.0695364778037235,
      "loss": 0.0,
      "num_input_tokens_seen": 15456992,
      "step": 27210
    },
    {
      "epoch": 477.46017699115043,
      "grad_norm": 1.1336310556941953e-08,
      "learning_rate": 0.06948677131194907,
      "loss": 0.0,
      "num_input_tokens_seen": 15460000,
      "step": 27215
    },
    {
      "epoch": 477.5486725663717,
      "grad_norm": 4.671350595231161e-08,
      "learning_rate": 0.06943707723632629,
      "loss": 0.0,
      "num_input_tokens_seen": 15462736,
      "step": 27220
    },
    {
      "epoch": 477.6371681415929,
      "grad_norm": 3.82123666042844e-08,
      "learning_rate": 0.06938739558451867,
      "loss": 0.0,
      "num_input_tokens_seen": 15465216,
      "step": 27225
    },
    {
      "epoch": 477.72566371681415,
      "grad_norm": 2.7914021316632898e-08,
      "learning_rate": 0.06933772636418763,
      "loss": 0.0,
      "num_input_tokens_seen": 15468384,
      "step": 27230
    },
    {
      "epoch": 477.8141592920354,
      "grad_norm": 1.2741340427169234e-08,
      "learning_rate": 0.06928806958299293,
      "loss": 0.0,
      "num_input_tokens_seen": 15471680,
      "step": 27235
    },
    {
      "epoch": 477.9026548672566,
      "grad_norm": 2.0837330083622874e-08,
      "learning_rate": 0.06923842524859211,
      "loss": 0.0,
      "num_input_tokens_seen": 15474256,
      "step": 27240
    },
    {
      "epoch": 477.9911504424779,
      "grad_norm": 1.810075467290062e-08,
      "learning_rate": 0.06918879336864105,
      "loss": 0.0,
      "num_input_tokens_seen": 15477216,
      "step": 27245
    },
    {
      "epoch": 478.070796460177,
      "grad_norm": 2.6881158632363622e-08,
      "learning_rate": 0.06913917395079362,
      "loss": 0.0,
      "num_input_tokens_seen": 15479632,
      "step": 27250
    },
    {
      "epoch": 478.1592920353982,
      "grad_norm": 1.3797913922530824e-08,
      "learning_rate": 0.0690895670027017,
      "loss": 0.0,
      "num_input_tokens_seen": 15482416,
      "step": 27255
    },
    {
      "epoch": 478.24778761061947,
      "grad_norm": 2.6186812718265173e-08,
      "learning_rate": 0.06903997253201531,
      "loss": 0.0,
      "num_input_tokens_seen": 15485280,
      "step": 27260
    },
    {
      "epoch": 478.3362831858407,
      "grad_norm": 2.8028910747934788e-08,
      "learning_rate": 0.06899039054638263,
      "loss": 0.0,
      "num_input_tokens_seen": 15487776,
      "step": 27265
    },
    {
      "epoch": 478.42477876106193,
      "grad_norm": 1.4396940528627056e-08,
      "learning_rate": 0.06894082105344976,
      "loss": 0.0,
      "num_input_tokens_seen": 15490944,
      "step": 27270
    },
    {
      "epoch": 478.5132743362832,
      "grad_norm": 1.340068678246098e-08,
      "learning_rate": 0.06889126406086087,
      "loss": 0.0,
      "num_input_tokens_seen": 15493584,
      "step": 27275
    },
    {
      "epoch": 478.60176991150445,
      "grad_norm": 2.012478361734793e-08,
      "learning_rate": 0.0688417195762584,
      "loss": 0.0,
      "num_input_tokens_seen": 15496672,
      "step": 27280
    },
    {
      "epoch": 478.69026548672565,
      "grad_norm": 7.284918002170571e-09,
      "learning_rate": 0.06879218760728262,
      "loss": 0.0,
      "num_input_tokens_seen": 15499664,
      "step": 27285
    },
    {
      "epoch": 478.7787610619469,
      "grad_norm": 6.207403036029291e-09,
      "learning_rate": 0.06874266816157207,
      "loss": 0.0,
      "num_input_tokens_seen": 15502736,
      "step": 27290
    },
    {
      "epoch": 478.86725663716817,
      "grad_norm": 1.0570317066083135e-08,
      "learning_rate": 0.06869316124676321,
      "loss": 0.0,
      "num_input_tokens_seen": 15505280,
      "step": 27295
    },
    {
      "epoch": 478.95575221238937,
      "grad_norm": 2.7189454243625732e-08,
      "learning_rate": 0.06864366687049062,
      "loss": 0.0,
      "num_input_tokens_seen": 15508256,
      "step": 27300
    },
    {
      "epoch": 479.0353982300885,
      "grad_norm": 4.533600872491661e-08,
      "learning_rate": 0.06859418504038704,
      "loss": 0.0,
      "num_input_tokens_seen": 15510952,
      "step": 27305
    },
    {
      "epoch": 479.12389380530976,
      "grad_norm": 1.766427359939371e-08,
      "learning_rate": 0.06854471576408311,
      "loss": 0.0,
      "num_input_tokens_seen": 15513672,
      "step": 27310
    },
    {
      "epoch": 479.21238938053096,
      "grad_norm": 2.8404043561636172e-08,
      "learning_rate": 0.06849525904920767,
      "loss": 0.0,
      "num_input_tokens_seen": 15516376,
      "step": 27315
    },
    {
      "epoch": 479.3008849557522,
      "grad_norm": 1.4365200584620652e-08,
      "learning_rate": 0.06844581490338748,
      "loss": 0.0,
      "num_input_tokens_seen": 15519176,
      "step": 27320
    },
    {
      "epoch": 479.3893805309734,
      "grad_norm": 3.500647238752208e-08,
      "learning_rate": 0.06839638333424752,
      "loss": 0.0,
      "num_input_tokens_seen": 15522648,
      "step": 27325
    },
    {
      "epoch": 479.4778761061947,
      "grad_norm": 1.5155226407159716e-08,
      "learning_rate": 0.06834696434941082,
      "loss": 0.0,
      "num_input_tokens_seen": 15525672,
      "step": 27330
    },
    {
      "epoch": 479.56637168141594,
      "grad_norm": 1.918274605827719e-08,
      "learning_rate": 0.06829755795649824,
      "loss": 0.0,
      "num_input_tokens_seen": 15528648,
      "step": 27335
    },
    {
      "epoch": 479.65486725663715,
      "grad_norm": 6.46174953544687e-08,
      "learning_rate": 0.06824816416312904,
      "loss": 0.0,
      "num_input_tokens_seen": 15531416,
      "step": 27340
    },
    {
      "epoch": 479.7433628318584,
      "grad_norm": 2.2823300582786032e-08,
      "learning_rate": 0.06819878297692027,
      "loss": 0.0,
      "num_input_tokens_seen": 15533992,
      "step": 27345
    },
    {
      "epoch": 479.83185840707966,
      "grad_norm": 1.2284036898790873e-08,
      "learning_rate": 0.0681494144054871,
      "loss": 0.0,
      "num_input_tokens_seen": 15536600,
      "step": 27350
    },
    {
      "epoch": 479.92035398230087,
      "grad_norm": 1.5650218898599633e-08,
      "learning_rate": 0.06810005845644286,
      "loss": 0.0,
      "num_input_tokens_seen": 15539992,
      "step": 27355
    },
    {
      "epoch": 480.0,
      "grad_norm": 2.9691902270201354e-09,
      "learning_rate": 0.06805071513739878,
      "loss": 0.0,
      "num_input_tokens_seen": 15542312,
      "step": 27360
    },
    {
      "epoch": 480.08849557522126,
      "grad_norm": 1.4223238586907883e-08,
      "learning_rate": 0.06800138445596428,
      "loss": 0.0,
      "num_input_tokens_seen": 15545032,
      "step": 27365
    },
    {
      "epoch": 480.17699115044246,
      "grad_norm": 4.109620377334977e-08,
      "learning_rate": 0.06795206641974678,
      "loss": 0.0,
      "num_input_tokens_seen": 15548616,
      "step": 27370
    },
    {
      "epoch": 480.2654867256637,
      "grad_norm": 1.5817493093095436e-08,
      "learning_rate": 0.06790276103635169,
      "loss": 0.0,
      "num_input_tokens_seen": 15551624,
      "step": 27375
    },
    {
      "epoch": 480.353982300885,
      "grad_norm": 2.0078244844512483e-08,
      "learning_rate": 0.0678534683133826,
      "loss": 0.0,
      "num_input_tokens_seen": 15555016,
      "step": 27380
    },
    {
      "epoch": 480.4424778761062,
      "grad_norm": 4.0415947921701445e-08,
      "learning_rate": 0.06780418825844095,
      "loss": 0.0,
      "num_input_tokens_seen": 15557608,
      "step": 27385
    },
    {
      "epoch": 480.53097345132744,
      "grad_norm": 1.6421818571643598e-08,
      "learning_rate": 0.0677549208791264,
      "loss": 0.0,
      "num_input_tokens_seen": 15560360,
      "step": 27390
    },
    {
      "epoch": 480.6194690265487,
      "grad_norm": 1.2078074540511352e-08,
      "learning_rate": 0.06770566618303668,
      "loss": 0.0,
      "num_input_tokens_seen": 15562952,
      "step": 27395
    },
    {
      "epoch": 480.7079646017699,
      "grad_norm": 1.5483969662000163e-08,
      "learning_rate": 0.06765642417776736,
      "loss": 0.0,
      "num_input_tokens_seen": 15565784,
      "step": 27400
    },
    {
      "epoch": 480.7079646017699,
      "eval_loss": 0.7863607406616211,
      "eval_runtime": 0.9418,
      "eval_samples_per_second": 26.546,
      "eval_steps_per_second": 13.804,
      "num_input_tokens_seen": 15565784,
      "step": 27400
    },
    {
      "epoch": 480.79646017699116,
      "grad_norm": 3.369859769009054e-08,
      "learning_rate": 0.0676071948709122,
      "loss": 0.0,
      "num_input_tokens_seen": 15568328,
      "step": 27405
    },
    {
      "epoch": 480.88495575221236,
      "grad_norm": 8.449854149716884e-09,
      "learning_rate": 0.06755797827006307,
      "loss": 0.0,
      "num_input_tokens_seen": 15571400,
      "step": 27410
    },
    {
      "epoch": 480.9734513274336,
      "grad_norm": 2.1604401823083208e-08,
      "learning_rate": 0.06750877438280974,
      "loss": 0.0,
      "num_input_tokens_seen": 15574088,
      "step": 27415
    },
    {
      "epoch": 481.05309734513276,
      "grad_norm": 3.30394094305575e-08,
      "learning_rate": 0.06745958321673998,
      "loss": 0.0,
      "num_input_tokens_seen": 15576984,
      "step": 27420
    },
    {
      "epoch": 481.14159292035396,
      "grad_norm": 2.9118734090616272e-08,
      "learning_rate": 0.0674104047794398,
      "loss": 0.0,
      "num_input_tokens_seen": 15579496,
      "step": 27425
    },
    {
      "epoch": 481.2300884955752,
      "grad_norm": 1.8822717606781225e-08,
      "learning_rate": 0.06736123907849303,
      "loss": 0.0,
      "num_input_tokens_seen": 15582376,
      "step": 27430
    },
    {
      "epoch": 481.3185840707965,
      "grad_norm": 1.6837292449167762e-08,
      "learning_rate": 0.06731208612148178,
      "loss": 0.0,
      "num_input_tokens_seen": 15585416,
      "step": 27435
    },
    {
      "epoch": 481.4070796460177,
      "grad_norm": 1.7802994634052993e-08,
      "learning_rate": 0.0672629459159859,
      "loss": 0.0,
      "num_input_tokens_seen": 15588184,
      "step": 27440
    },
    {
      "epoch": 481.49557522123894,
      "grad_norm": 6.902975968614555e-08,
      "learning_rate": 0.0672138184695835,
      "loss": 0.0,
      "num_input_tokens_seen": 15590984,
      "step": 27445
    },
    {
      "epoch": 481.5840707964602,
      "grad_norm": 1.8512443134000023e-08,
      "learning_rate": 0.0671647037898507,
      "loss": 0.0,
      "num_input_tokens_seen": 15593608,
      "step": 27450
    },
    {
      "epoch": 481.6725663716814,
      "grad_norm": 1.5314611800931743e-08,
      "learning_rate": 0.0671156018843615,
      "loss": 0.0,
      "num_input_tokens_seen": 15596728,
      "step": 27455
    },
    {
      "epoch": 481.76106194690266,
      "grad_norm": 4.142725984479512e-08,
      "learning_rate": 0.06706651276068812,
      "loss": 0.0,
      "num_input_tokens_seen": 15599736,
      "step": 27460
    },
    {
      "epoch": 481.8495575221239,
      "grad_norm": 4.1768966951849507e-08,
      "learning_rate": 0.06701743642640064,
      "loss": 0.0,
      "num_input_tokens_seen": 15602520,
      "step": 27465
    },
    {
      "epoch": 481.9380530973451,
      "grad_norm": 1.752091449702675e-08,
      "learning_rate": 0.06696837288906729,
      "loss": 0.0,
      "num_input_tokens_seen": 15605320,
      "step": 27470
    },
    {
      "epoch": 482.01769911504425,
      "grad_norm": 3.1443633474737e-08,
      "learning_rate": 0.06691932215625432,
      "loss": 0.0,
      "num_input_tokens_seen": 15607816,
      "step": 27475
    },
    {
      "epoch": 482.1061946902655,
      "grad_norm": 1.9127451622580338e-08,
      "learning_rate": 0.06687028423552589,
      "loss": 0.0,
      "num_input_tokens_seen": 15610568,
      "step": 27480
    },
    {
      "epoch": 482.1946902654867,
      "grad_norm": 2.356654782431633e-08,
      "learning_rate": 0.06682125913444435,
      "loss": 0.0,
      "num_input_tokens_seen": 15613480,
      "step": 27485
    },
    {
      "epoch": 482.283185840708,
      "grad_norm": 7.140675606365221e-09,
      "learning_rate": 0.0667722468605699,
      "loss": 0.0,
      "num_input_tokens_seen": 15616344,
      "step": 27490
    },
    {
      "epoch": 482.37168141592923,
      "grad_norm": 2.4066673987022114e-08,
      "learning_rate": 0.06672324742146094,
      "loss": 0.0,
      "num_input_tokens_seen": 15618984,
      "step": 27495
    },
    {
      "epoch": 482.46017699115043,
      "grad_norm": 2.5545725534925623e-08,
      "learning_rate": 0.06667426082467373,
      "loss": 0.0,
      "num_input_tokens_seen": 15621560,
      "step": 27500
    },
    {
      "epoch": 482.5486725663717,
      "grad_norm": 2.811770194455221e-08,
      "learning_rate": 0.0666252870777626,
      "loss": 0.0,
      "num_input_tokens_seen": 15624664,
      "step": 27505
    },
    {
      "epoch": 482.6371681415929,
      "grad_norm": 2.852794445118434e-08,
      "learning_rate": 0.06657632618827995,
      "loss": 0.0,
      "num_input_tokens_seen": 15627640,
      "step": 27510
    },
    {
      "epoch": 482.72566371681415,
      "grad_norm": 3.7351075121705435e-08,
      "learning_rate": 0.06652737816377623,
      "loss": 0.0,
      "num_input_tokens_seen": 15631112,
      "step": 27515
    },
    {
      "epoch": 482.8141592920354,
      "grad_norm": 5.0921499905598466e-08,
      "learning_rate": 0.06647844301179971,
      "loss": 0.0,
      "num_input_tokens_seen": 15633832,
      "step": 27520
    },
    {
      "epoch": 482.9026548672566,
      "grad_norm": 1.5320024360221396e-08,
      "learning_rate": 0.06642952073989689,
      "loss": 0.0,
      "num_input_tokens_seen": 15636792,
      "step": 27525
    },
    {
      "epoch": 482.9911504424779,
      "grad_norm": 2.2708201541377093e-08,
      "learning_rate": 0.06638061135561223,
      "loss": 0.0,
      "num_input_tokens_seen": 15639560,
      "step": 27530
    },
    {
      "epoch": 483.070796460177,
      "grad_norm": 2.5551171844995224e-08,
      "learning_rate": 0.06633171486648808,
      "loss": 0.0,
      "num_input_tokens_seen": 15642000,
      "step": 27535
    },
    {
      "epoch": 483.1592920353982,
      "grad_norm": 2.150577138593235e-08,
      "learning_rate": 0.06628283128006499,
      "loss": 0.0,
      "num_input_tokens_seen": 15645056,
      "step": 27540
    },
    {
      "epoch": 483.24778761061947,
      "grad_norm": 4.3691319007166385e-08,
      "learning_rate": 0.0662339606038813,
      "loss": 0.0,
      "num_input_tokens_seen": 15648016,
      "step": 27545
    },
    {
      "epoch": 483.3362831858407,
      "grad_norm": 5.009442816117371e-08,
      "learning_rate": 0.06618510284547358,
      "loss": 0.0,
      "num_input_tokens_seen": 15650528,
      "step": 27550
    },
    {
      "epoch": 483.42477876106193,
      "grad_norm": 2.7134323232758106e-08,
      "learning_rate": 0.06613625801237633,
      "loss": 0.0,
      "num_input_tokens_seen": 15653824,
      "step": 27555
    },
    {
      "epoch": 483.5132743362832,
      "grad_norm": 2.52503191688902e-08,
      "learning_rate": 0.066087426112122,
      "loss": 0.0,
      "num_input_tokens_seen": 15656496,
      "step": 27560
    },
    {
      "epoch": 483.60176991150445,
      "grad_norm": 3.938269799164118e-08,
      "learning_rate": 0.06603860715224101,
      "loss": 0.0,
      "num_input_tokens_seen": 15659120,
      "step": 27565
    },
    {
      "epoch": 483.69026548672565,
      "grad_norm": 4.00681834378247e-08,
      "learning_rate": 0.06598980114026198,
      "loss": 0.0,
      "num_input_tokens_seen": 15662032,
      "step": 27570
    },
    {
      "epoch": 483.7787610619469,
      "grad_norm": 3.038986662318166e-08,
      "learning_rate": 0.06594100808371128,
      "loss": 0.0,
      "num_input_tokens_seen": 15664544,
      "step": 27575
    },
    {
      "epoch": 483.86725663716817,
      "grad_norm": 2.09384296567805e-08,
      "learning_rate": 0.06589222799011357,
      "loss": 0.0,
      "num_input_tokens_seen": 15667200,
      "step": 27580
    },
    {
      "epoch": 483.95575221238937,
      "grad_norm": 1.6550583126218044e-08,
      "learning_rate": 0.0658434608669912,
      "loss": 0.0,
      "num_input_tokens_seen": 15670000,
      "step": 27585
    },
    {
      "epoch": 484.0353982300885,
      "grad_norm": 2.7876955854821972e-08,
      "learning_rate": 0.06579470672186473,
      "loss": 0.0,
      "num_input_tokens_seen": 15672776,
      "step": 27590
    },
    {
      "epoch": 484.12389380530976,
      "grad_norm": 3.6109415901819375e-08,
      "learning_rate": 0.06574596556225275,
      "loss": 0.0,
      "num_input_tokens_seen": 15676328,
      "step": 27595
    },
    {
      "epoch": 484.21238938053096,
      "grad_norm": 1.9713812804411646e-08,
      "learning_rate": 0.06569723739567161,
      "loss": 0.0,
      "num_input_tokens_seen": 15679720,
      "step": 27600
    },
    {
      "epoch": 484.21238938053096,
      "eval_loss": 0.7917570471763611,
      "eval_runtime": 0.9429,
      "eval_samples_per_second": 26.515,
      "eval_steps_per_second": 13.788,
      "num_input_tokens_seen": 15679720,
      "step": 27600
    },
    {
      "epoch": 484.3008849557522,
      "grad_norm": 3.1416043100307434e-08,
      "learning_rate": 0.06564852222963588,
      "loss": 0.0,
      "num_input_tokens_seen": 15682680,
      "step": 27605
    },
    {
      "epoch": 484.3893805309734,
      "grad_norm": 1.3314604530023644e-08,
      "learning_rate": 0.06559982007165813,
      "loss": 0.0,
      "num_input_tokens_seen": 15685336,
      "step": 27610
    },
    {
      "epoch": 484.4778761061947,
      "grad_norm": 1.3376977747725505e-08,
      "learning_rate": 0.06555113092924868,
      "loss": 0.0,
      "num_input_tokens_seen": 15687944,
      "step": 27615
    },
    {
      "epoch": 484.56637168141594,
      "grad_norm": 8.574632559543716e-09,
      "learning_rate": 0.06550245480991615,
      "loss": 0.0,
      "num_input_tokens_seen": 15690776,
      "step": 27620
    },
    {
      "epoch": 484.65486725663715,
      "grad_norm": 2.8876039337433212e-08,
      "learning_rate": 0.0654537917211669,
      "loss": 0.0,
      "num_input_tokens_seen": 15693704,
      "step": 27625
    },
    {
      "epoch": 484.7433628318584,
      "grad_norm": 2.2639429886339713e-08,
      "learning_rate": 0.0654051416705055,
      "loss": 0.0,
      "num_input_tokens_seen": 15696456,
      "step": 27630
    },
    {
      "epoch": 484.83185840707966,
      "grad_norm": 2.965112599895292e-08,
      "learning_rate": 0.06535650466543427,
      "loss": 0.0,
      "num_input_tokens_seen": 15699304,
      "step": 27635
    },
    {
      "epoch": 484.92035398230087,
      "grad_norm": 2.835233559039807e-08,
      "learning_rate": 0.0653078807134538,
      "loss": 0.0,
      "num_input_tokens_seen": 15701880,
      "step": 27640
    },
    {
      "epoch": 485.0,
      "grad_norm": 1.805410754229797e-08,
      "learning_rate": 0.06525926982206236,
      "loss": 0.0,
      "num_input_tokens_seen": 15704328,
      "step": 27645
    },
    {
      "epoch": 485.08849557522126,
      "grad_norm": 2.415598743255032e-08,
      "learning_rate": 0.06521067199875648,
      "loss": 0.0,
      "num_input_tokens_seen": 15707336,
      "step": 27650
    },
    {
      "epoch": 485.17699115044246,
      "grad_norm": 6.43802451349984e-08,
      "learning_rate": 0.06516208725103047,
      "loss": 0.0,
      "num_input_tokens_seen": 15709944,
      "step": 27655
    },
    {
      "epoch": 485.2654867256637,
      "grad_norm": 1.2468158949729968e-08,
      "learning_rate": 0.06511351558637678,
      "loss": 0.0,
      "num_input_tokens_seen": 15712472,
      "step": 27660
    },
    {
      "epoch": 485.353982300885,
      "grad_norm": 1.5650506668407616e-08,
      "learning_rate": 0.06506495701228569,
      "loss": 0.0,
      "num_input_tokens_seen": 15715112,
      "step": 27665
    },
    {
      "epoch": 485.4424778761062,
      "grad_norm": 9.567106218355548e-09,
      "learning_rate": 0.06501641153624559,
      "loss": 0.0,
      "num_input_tokens_seen": 15717704,
      "step": 27670
    },
    {
      "epoch": 485.53097345132744,
      "grad_norm": 2.889923322868526e-08,
      "learning_rate": 0.06496787916574286,
      "loss": 0.0,
      "num_input_tokens_seen": 15720360,
      "step": 27675
    },
    {
      "epoch": 485.6194690265487,
      "grad_norm": 1.3565387924074912e-08,
      "learning_rate": 0.06491935990826168,
      "loss": 0.0,
      "num_input_tokens_seen": 15723352,
      "step": 27680
    },
    {
      "epoch": 485.7079646017699,
      "grad_norm": 3.368909062828607e-08,
      "learning_rate": 0.0648708537712844,
      "loss": 0.0,
      "num_input_tokens_seen": 15726600,
      "step": 27685
    },
    {
      "epoch": 485.79646017699116,
      "grad_norm": 1.4761341482483203e-08,
      "learning_rate": 0.06482236076229132,
      "loss": 0.0,
      "num_input_tokens_seen": 15729752,
      "step": 27690
    },
    {
      "epoch": 485.88495575221236,
      "grad_norm": 2.0190856986346262e-08,
      "learning_rate": 0.06477388088876056,
      "loss": 0.0,
      "num_input_tokens_seen": 15732568,
      "step": 27695
    },
    {
      "epoch": 485.9734513274336,
      "grad_norm": 2.1439051423044475e-08,
      "learning_rate": 0.06472541415816846,
      "loss": 0.0,
      "num_input_tokens_seen": 15735464,
      "step": 27700
    },
    {
      "epoch": 486.05309734513276,
      "grad_norm": 1.1702649516109886e-08,
      "learning_rate": 0.06467696057798909,
      "loss": 0.0,
      "num_input_tokens_seen": 15737976,
      "step": 27705
    },
    {
      "epoch": 486.14159292035396,
      "grad_norm": 2.1986894083170228e-08,
      "learning_rate": 0.0646285201556946,
      "loss": 0.0,
      "num_input_tokens_seen": 15740776,
      "step": 27710
    },
    {
      "epoch": 486.2300884955752,
      "grad_norm": 2.2645885167094093e-08,
      "learning_rate": 0.06458009289875521,
      "loss": 0.0,
      "num_input_tokens_seen": 15744008,
      "step": 27715
    },
    {
      "epoch": 486.3185840707965,
      "grad_norm": 1.5002578734879535e-08,
      "learning_rate": 0.0645316788146389,
      "loss": 0.0,
      "num_input_tokens_seen": 15746824,
      "step": 27720
    },
    {
      "epoch": 486.4070796460177,
      "grad_norm": 2.9855009131551924e-08,
      "learning_rate": 0.06448327791081175,
      "loss": 0.0,
      "num_input_tokens_seen": 15749496,
      "step": 27725
    },
    {
      "epoch": 486.49557522123894,
      "grad_norm": 1.9493542779969175e-08,
      "learning_rate": 0.0644348901947379,
      "loss": 0.0,
      "num_input_tokens_seen": 15751848,
      "step": 27730
    },
    {
      "epoch": 486.5840707964602,
      "grad_norm": 4.596095593001337e-08,
      "learning_rate": 0.06438651567387917,
      "loss": 0.0,
      "num_input_tokens_seen": 15755016,
      "step": 27735
    },
    {
      "epoch": 486.6725663716814,
      "grad_norm": 1.6629547516799903e-08,
      "learning_rate": 0.0643381543556957,
      "loss": 0.0,
      "num_input_tokens_seen": 15758056,
      "step": 27740
    },
    {
      "epoch": 486.76106194690266,
      "grad_norm": 2.371603180506554e-08,
      "learning_rate": 0.06428980624764526,
      "loss": 0.0,
      "num_input_tokens_seen": 15760712,
      "step": 27745
    },
    {
      "epoch": 486.8495575221239,
      "grad_norm": 4.484798310500082e-08,
      "learning_rate": 0.06424147135718378,
      "loss": 0.0,
      "num_input_tokens_seen": 15763544,
      "step": 27750
    },
    {
      "epoch": 486.9380530973451,
      "grad_norm": 2.0263472677584105e-08,
      "learning_rate": 0.06419314969176519,
      "loss": 0.0,
      "num_input_tokens_seen": 15766648,
      "step": 27755
    },
    {
      "epoch": 487.01769911504425,
      "grad_norm": 1.879711497565495e-08,
      "learning_rate": 0.06414484125884118,
      "loss": 0.0,
      "num_input_tokens_seen": 15769352,
      "step": 27760
    },
    {
      "epoch": 487.1061946902655,
      "grad_norm": 7.728368167647659e-09,
      "learning_rate": 0.06409654606586157,
      "loss": 0.0,
      "num_input_tokens_seen": 15771960,
      "step": 27765
    },
    {
      "epoch": 487.1946902654867,
      "grad_norm": 1.8745836882771982e-08,
      "learning_rate": 0.06404826412027415,
      "loss": 0.0,
      "num_input_tokens_seen": 15774872,
      "step": 27770
    },
    {
      "epoch": 487.283185840708,
      "grad_norm": 2.445794144989577e-08,
      "learning_rate": 0.06399999542952453,
      "loss": 0.0,
      "num_input_tokens_seen": 15777672,
      "step": 27775
    },
    {
      "epoch": 487.37168141592923,
      "grad_norm": 9.112805798849877e-08,
      "learning_rate": 0.0639517400010563,
      "loss": 0.0,
      "num_input_tokens_seen": 15780872,
      "step": 27780
    },
    {
      "epoch": 487.46017699115043,
      "grad_norm": 3.012747740172017e-08,
      "learning_rate": 0.06390349784231118,
      "loss": 0.0,
      "num_input_tokens_seen": 15784040,
      "step": 27785
    },
    {
      "epoch": 487.5486725663717,
      "grad_norm": 4.162907529803306e-08,
      "learning_rate": 0.06385526896072859,
      "loss": 0.0,
      "num_input_tokens_seen": 15786696,
      "step": 27790
    },
    {
      "epoch": 487.6371681415929,
      "grad_norm": 4.07349531883483e-08,
      "learning_rate": 0.06380705336374613,
      "loss": 0.0,
      "num_input_tokens_seen": 15789272,
      "step": 27795
    },
    {
      "epoch": 487.72566371681415,
      "grad_norm": 1.9945302298651768e-08,
      "learning_rate": 0.06375885105879918,
      "loss": 0.0,
      "num_input_tokens_seen": 15792680,
      "step": 27800
    },
    {
      "epoch": 487.72566371681415,
      "eval_loss": 0.8236088156700134,
      "eval_runtime": 0.9399,
      "eval_samples_per_second": 26.599,
      "eval_steps_per_second": 13.831,
      "num_input_tokens_seen": 15792680,
      "step": 27800
    },
    {
      "epoch": 487.8141592920354,
      "grad_norm": 1.9408627593975325e-08,
      "learning_rate": 0.06371066205332115,
      "loss": 0.0,
      "num_input_tokens_seen": 15795880,
      "step": 27805
    },
    {
      "epoch": 487.9026548672566,
      "grad_norm": 9.965536840184086e-09,
      "learning_rate": 0.06366248635474347,
      "loss": 0.0,
      "num_input_tokens_seen": 15798424,
      "step": 27810
    },
    {
      "epoch": 487.9911504424779,
      "grad_norm": 2.6575495581937503e-08,
      "learning_rate": 0.06361432397049532,
      "loss": 0.0,
      "num_input_tokens_seen": 15801624,
      "step": 27815
    },
    {
      "epoch": 488.070796460177,
      "grad_norm": 2.4796168673901775e-08,
      "learning_rate": 0.06356617490800408,
      "loss": 0.0,
      "num_input_tokens_seen": 15803736,
      "step": 27820
    },
    {
      "epoch": 488.1592920353982,
      "grad_norm": 4.6748819926278884e-08,
      "learning_rate": 0.06351803917469478,
      "loss": 0.0,
      "num_input_tokens_seen": 15806408,
      "step": 27825
    },
    {
      "epoch": 488.24778761061947,
      "grad_norm": 2.189625725179667e-08,
      "learning_rate": 0.06346991677799067,
      "loss": 0.0,
      "num_input_tokens_seen": 15809416,
      "step": 27830
    },
    {
      "epoch": 488.3362831858407,
      "grad_norm": 2.7210502295815786e-08,
      "learning_rate": 0.06342180772531283,
      "loss": 0.0,
      "num_input_tokens_seen": 15812392,
      "step": 27835
    },
    {
      "epoch": 488.42477876106193,
      "grad_norm": 3.458385222643301e-08,
      "learning_rate": 0.06337371202408021,
      "loss": 0.0,
      "num_input_tokens_seen": 15815896,
      "step": 27840
    },
    {
      "epoch": 488.5132743362832,
      "grad_norm": 1.2597342724518512e-08,
      "learning_rate": 0.06332562968170984,
      "loss": 0.0,
      "num_input_tokens_seen": 15819336,
      "step": 27845
    },
    {
      "epoch": 488.60176991150445,
      "grad_norm": 3.673816806326613e-08,
      "learning_rate": 0.06327756070561656,
      "loss": 0.0,
      "num_input_tokens_seen": 15822296,
      "step": 27850
    },
    {
      "epoch": 488.69026548672565,
      "grad_norm": 2.837514223585913e-08,
      "learning_rate": 0.06322950510321329,
      "loss": 0.0,
      "num_input_tokens_seen": 15825352,
      "step": 27855
    },
    {
      "epoch": 488.7787610619469,
      "grad_norm": 4.3403726834867484e-08,
      "learning_rate": 0.06318146288191076,
      "loss": 0.0,
      "num_input_tokens_seen": 15827672,
      "step": 27860
    },
    {
      "epoch": 488.86725663716817,
      "grad_norm": 3.095751210935305e-08,
      "learning_rate": 0.06313343404911763,
      "loss": 0.0,
      "num_input_tokens_seen": 15830072,
      "step": 27865
    },
    {
      "epoch": 488.95575221238937,
      "grad_norm": 1.8648185218239632e-08,
      "learning_rate": 0.0630854186122406,
      "loss": 0.0,
      "num_input_tokens_seen": 15832920,
      "step": 27870
    },
    {
      "epoch": 489.0353982300885,
      "grad_norm": 2.1929654536734233e-08,
      "learning_rate": 0.06303741657868431,
      "loss": 0.0,
      "num_input_tokens_seen": 15835384,
      "step": 27875
    },
    {
      "epoch": 489.12389380530976,
      "grad_norm": 5.489443566375485e-08,
      "learning_rate": 0.06298942795585115,
      "loss": 0.0,
      "num_input_tokens_seen": 15838136,
      "step": 27880
    },
    {
      "epoch": 489.21238938053096,
      "grad_norm": 1.785471681614581e-08,
      "learning_rate": 0.06294145275114167,
      "loss": 0.0,
      "num_input_tokens_seen": 15841240,
      "step": 27885
    },
    {
      "epoch": 489.3008849557522,
      "grad_norm": 2.340670413047974e-08,
      "learning_rate": 0.06289349097195428,
      "loss": 0.0,
      "num_input_tokens_seen": 15843800,
      "step": 27890
    },
    {
      "epoch": 489.3893805309734,
      "grad_norm": 5.529674496074222e-08,
      "learning_rate": 0.06284554262568516,
      "loss": 0.0,
      "num_input_tokens_seen": 15846568,
      "step": 27895
    },
    {
      "epoch": 489.4778761061947,
      "grad_norm": 1.0953157492110677e-08,
      "learning_rate": 0.06279760771972868,
      "loss": 0.0,
      "num_input_tokens_seen": 15849160,
      "step": 27900
    },
    {
      "epoch": 489.56637168141594,
      "grad_norm": 6.143937980596093e-08,
      "learning_rate": 0.06274968626147688,
      "loss": 0.0,
      "num_input_tokens_seen": 15852296,
      "step": 27905
    },
    {
      "epoch": 489.65486725663715,
      "grad_norm": 2.6424698873483976e-08,
      "learning_rate": 0.06270177825831993,
      "loss": 0.0,
      "num_input_tokens_seen": 15855480,
      "step": 27910
    },
    {
      "epoch": 489.7433628318584,
      "grad_norm": 2.735449911028809e-08,
      "learning_rate": 0.06265388371764587,
      "loss": 0.0,
      "num_input_tokens_seen": 15857880,
      "step": 27915
    },
    {
      "epoch": 489.83185840707966,
      "grad_norm": 9.226087449576426e-09,
      "learning_rate": 0.0626060026468406,
      "loss": 0.0,
      "num_input_tokens_seen": 15861416,
      "step": 27920
    },
    {
      "epoch": 489.92035398230087,
      "grad_norm": 2.228159168282673e-08,
      "learning_rate": 0.06255813505328794,
      "loss": 0.0,
      "num_input_tokens_seen": 15863928,
      "step": 27925
    },
    {
      "epoch": 490.0,
      "grad_norm": 2.348622274439549e-08,
      "learning_rate": 0.06251028094436978,
      "loss": 0.0,
      "num_input_tokens_seen": 15866288,
      "step": 27930
    },
    {
      "epoch": 490.08849557522126,
      "grad_norm": 2.2664737642230648e-08,
      "learning_rate": 0.06246244032746568,
      "loss": 0.0,
      "num_input_tokens_seen": 15868624,
      "step": 27935
    },
    {
      "epoch": 490.17699115044246,
      "grad_norm": 6.58412124820984e-09,
      "learning_rate": 0.06241461320995342,
      "loss": 0.0,
      "num_input_tokens_seen": 15871104,
      "step": 27940
    },
    {
      "epoch": 490.2654867256637,
      "grad_norm": 2.4284016575393252e-08,
      "learning_rate": 0.062366799599208426,
      "loss": 0.0,
      "num_input_tokens_seen": 15874144,
      "step": 27945
    },
    {
      "epoch": 490.353982300885,
      "grad_norm": 1.7138152230700143e-08,
      "learning_rate": 0.06231899950260418,
      "loss": 0.0,
      "num_input_tokens_seen": 15877168,
      "step": 27950
    },
    {
      "epoch": 490.4424778761062,
      "grad_norm": 2.356945394410559e-08,
      "learning_rate": 0.06227121292751214,
      "loss": 0.0,
      "num_input_tokens_seen": 15879920,
      "step": 27955
    },
    {
      "epoch": 490.53097345132744,
      "grad_norm": 3.791276981246483e-08,
      "learning_rate": 0.062223439881301496,
      "loss": 0.0,
      "num_input_tokens_seen": 15882768,
      "step": 27960
    },
    {
      "epoch": 490.6194690265487,
      "grad_norm": 1.990859743727924e-08,
      "learning_rate": 0.06217568037133948,
      "loss": 0.0,
      "num_input_tokens_seen": 15885952,
      "step": 27965
    },
    {
      "epoch": 490.7079646017699,
      "grad_norm": 4.34998845832979e-08,
      "learning_rate": 0.06212793440499126,
      "loss": 0.0,
      "num_input_tokens_seen": 15889264,
      "step": 27970
    },
    {
      "epoch": 490.79646017699116,
      "grad_norm": 7.3873769323995475e-09,
      "learning_rate": 0.062080201989619783,
      "loss": 0.0,
      "num_input_tokens_seen": 15891904,
      "step": 27975
    },
    {
      "epoch": 490.88495575221236,
      "grad_norm": 1.5405792197498158e-08,
      "learning_rate": 0.062032483132586094,
      "loss": 0.0,
      "num_input_tokens_seen": 15895296,
      "step": 27980
    },
    {
      "epoch": 490.9734513274336,
      "grad_norm": 9.528178246398511e-09,
      "learning_rate": 0.0619847778412489,
      "loss": 0.0,
      "num_input_tokens_seen": 15897952,
      "step": 27985
    },
    {
      "epoch": 491.05309734513276,
      "grad_norm": 5.402615954608336e-08,
      "learning_rate": 0.06193708612296509,
      "loss": 0.0,
      "num_input_tokens_seen": 15900272,
      "step": 27990
    },
    {
      "epoch": 491.14159292035396,
      "grad_norm": 3.92483592293047e-08,
      "learning_rate": 0.06188940798508923,
      "loss": 0.0,
      "num_input_tokens_seen": 15903680,
      "step": 27995
    },
    {
      "epoch": 491.2300884955752,
      "grad_norm": 1.2660685833054686e-08,
      "learning_rate": 0.06184174343497397,
      "loss": 0.0,
      "num_input_tokens_seen": 15906624,
      "step": 28000
    },
    {
      "epoch": 491.2300884955752,
      "eval_loss": 0.8142685294151306,
      "eval_runtime": 0.9239,
      "eval_samples_per_second": 27.059,
      "eval_steps_per_second": 14.071,
      "num_input_tokens_seen": 15906624,
      "step": 28000
    },
    {
      "epoch": 491.3185840707965,
      "grad_norm": 8.694552633414787e-09,
      "learning_rate": 0.061794092479969726,
      "loss": 0.0,
      "num_input_tokens_seen": 15909952,
      "step": 28005
    },
    {
      "epoch": 491.4070796460177,
      "grad_norm": 1.289255457948002e-08,
      "learning_rate": 0.06174645512742485,
      "loss": 0.0,
      "num_input_tokens_seen": 15912832,
      "step": 28010
    },
    {
      "epoch": 491.49557522123894,
      "grad_norm": 1.0240805536909647e-08,
      "learning_rate": 0.06169883138468565,
      "loss": 0.0,
      "num_input_tokens_seen": 15915616,
      "step": 28015
    },
    {
      "epoch": 491.5840707964602,
      "grad_norm": 2.0840547065859027e-08,
      "learning_rate": 0.06165122125909637,
      "loss": 0.0,
      "num_input_tokens_seen": 15918304,
      "step": 28020
    },
    {
      "epoch": 491.6725663716814,
      "grad_norm": 5.386306867194435e-08,
      "learning_rate": 0.061603624757998965,
      "loss": 0.0,
      "num_input_tokens_seen": 15921024,
      "step": 28025
    },
    {
      "epoch": 491.76106194690266,
      "grad_norm": 1.7394617302102233e-08,
      "learning_rate": 0.0615560418887335,
      "loss": 0.0,
      "num_input_tokens_seen": 15924064,
      "step": 28030
    },
    {
      "epoch": 491.8495575221239,
      "grad_norm": 4.1831796693259093e-08,
      "learning_rate": 0.06150847265863787,
      "loss": 0.0,
      "num_input_tokens_seen": 15926880,
      "step": 28035
    },
    {
      "epoch": 491.9380530973451,
      "grad_norm": 3.253825298088486e-08,
      "learning_rate": 0.061460917075047757,
      "loss": 0.0,
      "num_input_tokens_seen": 15929184,
      "step": 28040
    },
    {
      "epoch": 492.01769911504425,
      "grad_norm": 2.224761352920268e-08,
      "learning_rate": 0.06141337514529694,
      "loss": 0.0,
      "num_input_tokens_seen": 15931552,
      "step": 28045
    },
    {
      "epoch": 492.1061946902655,
      "grad_norm": 1.1628559448695341e-08,
      "learning_rate": 0.06136584687671687,
      "loss": 0.0,
      "num_input_tokens_seen": 15934192,
      "step": 28050
    },
    {
      "epoch": 492.1946902654867,
      "grad_norm": 1.1007839972876354e-08,
      "learning_rate": 0.061318332276637064,
      "loss": 0.0,
      "num_input_tokens_seen": 15937104,
      "step": 28055
    },
    {
      "epoch": 492.283185840708,
      "grad_norm": 3.031972894973478e-08,
      "learning_rate": 0.06127083135238491,
      "loss": 0.0,
      "num_input_tokens_seen": 15939728,
      "step": 28060
    },
    {
      "epoch": 492.37168141592923,
      "grad_norm": 2.0503948761074753e-08,
      "learning_rate": 0.06122334411128555,
      "loss": 0.0,
      "num_input_tokens_seen": 15943264,
      "step": 28065
    },
    {
      "epoch": 492.46017699115043,
      "grad_norm": 2.228597573150637e-08,
      "learning_rate": 0.06117587056066223,
      "loss": 0.0,
      "num_input_tokens_seen": 15945904,
      "step": 28070
    },
    {
      "epoch": 492.5486725663717,
      "grad_norm": 1.5795006191865468e-08,
      "learning_rate": 0.06112841070783589,
      "loss": 0.0,
      "num_input_tokens_seen": 15949104,
      "step": 28075
    },
    {
      "epoch": 492.6371681415929,
      "grad_norm": 7.878153240881147e-08,
      "learning_rate": 0.061080964560125406,
      "loss": 0.0,
      "num_input_tokens_seen": 15952032,
      "step": 28080
    },
    {
      "epoch": 492.72566371681415,
      "grad_norm": 3.1131495603631265e-08,
      "learning_rate": 0.06103353212484766,
      "loss": 0.0,
      "num_input_tokens_seen": 15954944,
      "step": 28085
    },
    {
      "epoch": 492.8141592920354,
      "grad_norm": 6.106213135126382e-08,
      "learning_rate": 0.06098611340931722,
      "loss": 0.0,
      "num_input_tokens_seen": 15957568,
      "step": 28090
    },
    {
      "epoch": 492.9026548672566,
      "grad_norm": 2.354839345741766e-08,
      "learning_rate": 0.06093870842084672,
      "loss": 0.0,
      "num_input_tokens_seen": 15960352,
      "step": 28095
    },
    {
      "epoch": 492.9911504424779,
      "grad_norm": 1.8488929498516882e-08,
      "learning_rate": 0.06089131716674666,
      "loss": 0.0,
      "num_input_tokens_seen": 15963360,
      "step": 28100
    },
    {
      "epoch": 493.070796460177,
      "grad_norm": 3.493559219691633e-08,
      "learning_rate": 0.060843939654325226,
      "loss": 0.0,
      "num_input_tokens_seen": 15966096,
      "step": 28105
    },
    {
      "epoch": 493.1592920353982,
      "grad_norm": 2.542613941614036e-08,
      "learning_rate": 0.06079657589088873,
      "loss": 0.0,
      "num_input_tokens_seen": 15968816,
      "step": 28110
    },
    {
      "epoch": 493.24778761061947,
      "grad_norm": 2.5576811779615127e-08,
      "learning_rate": 0.06074922588374126,
      "loss": 0.0,
      "num_input_tokens_seen": 15972160,
      "step": 28115
    },
    {
      "epoch": 493.3362831858407,
      "grad_norm": 2.2905075169887823e-08,
      "learning_rate": 0.06070188964018472,
      "loss": 0.0,
      "num_input_tokens_seen": 15974976,
      "step": 28120
    },
    {
      "epoch": 493.42477876106193,
      "grad_norm": 9.6494066070818e-09,
      "learning_rate": 0.06065456716751902,
      "loss": 0.0,
      "num_input_tokens_seen": 15977984,
      "step": 28125
    },
    {
      "epoch": 493.5132743362832,
      "grad_norm": 2.1603394628755268e-08,
      "learning_rate": 0.06060725847304182,
      "loss": 0.0,
      "num_input_tokens_seen": 15981376,
      "step": 28130
    },
    {
      "epoch": 493.60176991150445,
      "grad_norm": 3.144153026823915e-08,
      "learning_rate": 0.06055996356404877,
      "loss": 0.0,
      "num_input_tokens_seen": 15983952,
      "step": 28135
    },
    {
      "epoch": 493.69026548672565,
      "grad_norm": 2.370323670675134e-08,
      "learning_rate": 0.06051268244783327,
      "loss": 0.0,
      "num_input_tokens_seen": 15987056,
      "step": 28140
    },
    {
      "epoch": 493.7787610619469,
      "grad_norm": 4.5777778012734416e-08,
      "learning_rate": 0.06046541513168676,
      "loss": 0.0,
      "num_input_tokens_seen": 15989728,
      "step": 28145
    },
    {
      "epoch": 493.86725663716817,
      "grad_norm": 1.752186307157899e-08,
      "learning_rate": 0.060418161622898356,
      "loss": 0.0,
      "num_input_tokens_seen": 15992624,
      "step": 28150
    },
    {
      "epoch": 493.95575221238937,
      "grad_norm": 1.6456723983537813e-08,
      "learning_rate": 0.06037092192875521,
      "loss": 0.0,
      "num_input_tokens_seen": 15995184,
      "step": 28155
    },
    {
      "epoch": 494.0353982300885,
      "grad_norm": 1.7588114076261263e-08,
      "learning_rate": 0.060323696056542225,
      "loss": 0.0,
      "num_input_tokens_seen": 15997584,
      "step": 28160
    },
    {
      "epoch": 494.12389380530976,
      "grad_norm": 2.1456079579706966e-08,
      "learning_rate": 0.06027648401354229,
      "loss": 0.0,
      "num_input_tokens_seen": 16000352,
      "step": 28165
    },
    {
      "epoch": 494.21238938053096,
      "grad_norm": 2.761967188291692e-08,
      "learning_rate": 0.06022928580703601,
      "loss": 0.0,
      "num_input_tokens_seen": 16003136,
      "step": 28170
    },
    {
      "epoch": 494.3008849557522,
      "grad_norm": 3.057186859223293e-08,
      "learning_rate": 0.060182101444301986,
      "loss": 0.0,
      "num_input_tokens_seen": 16005744,
      "step": 28175
    },
    {
      "epoch": 494.3893805309734,
      "grad_norm": 2.4746283244780898e-08,
      "learning_rate": 0.06013493093261669,
      "loss": 0.0,
      "num_input_tokens_seen": 16008176,
      "step": 28180
    },
    {
      "epoch": 494.4778761061947,
      "grad_norm": 1.6186842088927733e-08,
      "learning_rate": 0.06008777427925432,
      "loss": 0.0,
      "num_input_tokens_seen": 16010960,
      "step": 28185
    },
    {
      "epoch": 494.56637168141594,
      "grad_norm": 2.602211068847282e-08,
      "learning_rate": 0.06004063149148705,
      "loss": 0.0,
      "num_input_tokens_seen": 16013568,
      "step": 28190
    },
    {
      "epoch": 494.65486725663715,
      "grad_norm": 1.6462561092112082e-08,
      "learning_rate": 0.05999350257658497,
      "loss": 0.0,
      "num_input_tokens_seen": 16016880,
      "step": 28195
    },
    {
      "epoch": 494.7433628318584,
      "grad_norm": 3.4576402185848565e-08,
      "learning_rate": 0.05994638754181582,
      "loss": 0.0,
      "num_input_tokens_seen": 16019936,
      "step": 28200
    },
    {
      "epoch": 494.7433628318584,
      "eval_loss": 0.8141090869903564,
      "eval_runtime": 0.9456,
      "eval_samples_per_second": 26.439,
      "eval_steps_per_second": 13.748,
      "num_input_tokens_seen": 16019936,
      "step": 28200
    },
    {
      "epoch": 494.83185840707966,
      "grad_norm": 7.953757652501281e-09,
      "learning_rate": 0.059899286394445445,
      "loss": 0.0,
      "num_input_tokens_seen": 16022880,
      "step": 28205
    },
    {
      "epoch": 494.92035398230087,
      "grad_norm": 2.8022602904798077e-08,
      "learning_rate": 0.059852199141737346,
      "loss": 0.0,
      "num_input_tokens_seen": 16025536,
      "step": 28210
    },
    {
      "epoch": 495.0,
      "grad_norm": 6.728201107364384e-09,
      "learning_rate": 0.05980512579095304,
      "loss": 0.0,
      "num_input_tokens_seen": 16028400,
      "step": 28215
    },
    {
      "epoch": 495.08849557522126,
      "grad_norm": 3.6913878176392245e-08,
      "learning_rate": 0.05975806634935181,
      "loss": 0.0,
      "num_input_tokens_seen": 16031440,
      "step": 28220
    },
    {
      "epoch": 495.17699115044246,
      "grad_norm": 1.9348675550645567e-08,
      "learning_rate": 0.05971102082419076,
      "loss": 0.0,
      "num_input_tokens_seen": 16034208,
      "step": 28225
    },
    {
      "epoch": 495.2654867256637,
      "grad_norm": 5.969614846890181e-08,
      "learning_rate": 0.05966398922272492,
      "loss": 0.0,
      "num_input_tokens_seen": 16037008,
      "step": 28230
    },
    {
      "epoch": 495.353982300885,
      "grad_norm": 1.1078350681259508e-08,
      "learning_rate": 0.059616971552207236,
      "loss": 0.0,
      "num_input_tokens_seen": 16039616,
      "step": 28235
    },
    {
      "epoch": 495.4424778761062,
      "grad_norm": 9.736626616074773e-09,
      "learning_rate": 0.059569967819888305,
      "loss": 0.0,
      "num_input_tokens_seen": 16042784,
      "step": 28240
    },
    {
      "epoch": 495.53097345132744,
      "grad_norm": 1.0250367665776139e-08,
      "learning_rate": 0.05952297803301681,
      "loss": 0.0,
      "num_input_tokens_seen": 16045504,
      "step": 28245
    },
    {
      "epoch": 495.6194690265487,
      "grad_norm": 3.726715647189849e-08,
      "learning_rate": 0.059476002198839056,
      "loss": 0.0,
      "num_input_tokens_seen": 16048560,
      "step": 28250
    },
    {
      "epoch": 495.7079646017699,
      "grad_norm": 2.0539919987072608e-08,
      "learning_rate": 0.05942904032459935,
      "loss": 0.0,
      "num_input_tokens_seen": 16051424,
      "step": 28255
    },
    {
      "epoch": 495.79646017699116,
      "grad_norm": 4.3088018486514557e-08,
      "learning_rate": 0.05938209241753987,
      "loss": 0.0,
      "num_input_tokens_seen": 16054320,
      "step": 28260
    },
    {
      "epoch": 495.88495575221236,
      "grad_norm": 4.867461100843684e-08,
      "learning_rate": 0.05933515848490046,
      "loss": 0.0,
      "num_input_tokens_seen": 16057616,
      "step": 28265
    },
    {
      "epoch": 495.9734513274336,
      "grad_norm": 1.3896435113736061e-08,
      "learning_rate": 0.059288238533918985,
      "loss": 0.0,
      "num_input_tokens_seen": 16060192,
      "step": 28270
    },
    {
      "epoch": 496.05309734513276,
      "grad_norm": 1.1646048569957657e-08,
      "learning_rate": 0.05924133257183113,
      "loss": 0.0,
      "num_input_tokens_seen": 16062768,
      "step": 28275
    },
    {
      "epoch": 496.14159292035396,
      "grad_norm": 3.000115711415674e-08,
      "learning_rate": 0.059194440605870285,
      "loss": 0.0,
      "num_input_tokens_seen": 16065344,
      "step": 28280
    },
    {
      "epoch": 496.2300884955752,
      "grad_norm": 2.2600032067998654e-08,
      "learning_rate": 0.059147562643267884,
      "loss": 0.0,
      "num_input_tokens_seen": 16068416,
      "step": 28285
    },
    {
      "epoch": 496.3185840707965,
      "grad_norm": 1.3439295010186925e-08,
      "learning_rate": 0.059100698691253055,
      "loss": 0.0,
      "num_input_tokens_seen": 16071088,
      "step": 28290
    },
    {
      "epoch": 496.4070796460177,
      "grad_norm": 2.6902037930653933e-08,
      "learning_rate": 0.05905384875705273,
      "loss": 0.0,
      "num_input_tokens_seen": 16073616,
      "step": 28295
    },
    {
      "epoch": 496.49557522123894,
      "grad_norm": 4.9219959663560076e-08,
      "learning_rate": 0.05900701284789189,
      "loss": 0.0,
      "num_input_tokens_seen": 16076064,
      "step": 28300
    },
    {
      "epoch": 496.5840707964602,
      "grad_norm": 3.5713867418962764e-08,
      "learning_rate": 0.058960190970993115,
      "loss": 0.0,
      "num_input_tokens_seen": 16079136,
      "step": 28305
    },
    {
      "epoch": 496.6725663716814,
      "grad_norm": 3.053920138995636e-08,
      "learning_rate": 0.058913383133576955,
      "loss": 0.0,
      "num_input_tokens_seen": 16081920,
      "step": 28310
    },
    {
      "epoch": 496.76106194690266,
      "grad_norm": 2.0759999941333263e-08,
      "learning_rate": 0.05886658934286185,
      "loss": 0.0,
      "num_input_tokens_seen": 16085072,
      "step": 28315
    },
    {
      "epoch": 496.8495575221239,
      "grad_norm": 1.439461971841638e-08,
      "learning_rate": 0.058819809606063846,
      "loss": 0.0,
      "num_input_tokens_seen": 16088336,
      "step": 28320
    },
    {
      "epoch": 496.9380530973451,
      "grad_norm": 2.7589235784830635e-08,
      "learning_rate": 0.05877304393039711,
      "loss": 0.0,
      "num_input_tokens_seen": 16091408,
      "step": 28325
    },
    {
      "epoch": 497.01769911504425,
      "grad_norm": 1.142614980409462e-08,
      "learning_rate": 0.05872629232307338,
      "loss": 0.0,
      "num_input_tokens_seen": 16094072,
      "step": 28330
    },
    {
      "epoch": 497.1061946902655,
      "grad_norm": 1.3095178275079888e-08,
      "learning_rate": 0.05867955479130239,
      "loss": 0.0,
      "num_input_tokens_seen": 16096984,
      "step": 28335
    },
    {
      "epoch": 497.1946902654867,
      "grad_norm": 1.2280130690101032e-08,
      "learning_rate": 0.058632831342291705,
      "loss": 0.0,
      "num_input_tokens_seen": 16099432,
      "step": 28340
    },
    {
      "epoch": 497.283185840708,
      "grad_norm": 1.3576397783765515e-08,
      "learning_rate": 0.05858612198324655,
      "loss": 0.0,
      "num_input_tokens_seen": 16102472,
      "step": 28345
    },
    {
      "epoch": 497.37168141592923,
      "grad_norm": 3.565315154219206e-08,
      "learning_rate": 0.05853942672137025,
      "loss": 0.0,
      "num_input_tokens_seen": 16105208,
      "step": 28350
    },
    {
      "epoch": 497.46017699115043,
      "grad_norm": 2.4666229947456486e-08,
      "learning_rate": 0.05849274556386363,
      "loss": 0.0,
      "num_input_tokens_seen": 16108248,
      "step": 28355
    },
    {
      "epoch": 497.5486725663717,
      "grad_norm": 2.1424915175316528e-08,
      "learning_rate": 0.05844607851792567,
      "loss": 0.0,
      "num_input_tokens_seen": 16111864,
      "step": 28360
    },
    {
      "epoch": 497.6371681415929,
      "grad_norm": 1.9844481613517928e-08,
      "learning_rate": 0.058399425590752924,
      "loss": 0.0,
      "num_input_tokens_seen": 16114456,
      "step": 28365
    },
    {
      "epoch": 497.72566371681415,
      "grad_norm": 2.148550493075163e-08,
      "learning_rate": 0.05835278678953985,
      "loss": 0.0,
      "num_input_tokens_seen": 16117240,
      "step": 28370
    },
    {
      "epoch": 497.8141592920354,
      "grad_norm": 1.4044058360695999e-08,
      "learning_rate": 0.05830616212147874,
      "loss": 0.0,
      "num_input_tokens_seen": 16119992,
      "step": 28375
    },
    {
      "epoch": 497.9026548672566,
      "grad_norm": 1.9756232205736524e-08,
      "learning_rate": 0.058259551593759784,
      "loss": 0.0,
      "num_input_tokens_seen": 16122744,
      "step": 28380
    },
    {
      "epoch": 497.9911504424779,
      "grad_norm": 3.442849916268642e-08,
      "learning_rate": 0.058212955213570804,
      "loss": 0.0,
      "num_input_tokens_seen": 16125528,
      "step": 28385
    },
    {
      "epoch": 498.070796460177,
      "grad_norm": 1.3224599193506492e-08,
      "learning_rate": 0.0581663729880976,
      "loss": 0.0,
      "num_input_tokens_seen": 16127848,
      "step": 28390
    },
    {
      "epoch": 498.1592920353982,
      "grad_norm": 3.631077305499275e-08,
      "learning_rate": 0.05811980492452379,
      "loss": 0.0,
      "num_input_tokens_seen": 16130488,
      "step": 28395
    },
    {
      "epoch": 498.24778761061947,
      "grad_norm": 2.4789184038809253e-08,
      "learning_rate": 0.058073251030030644,
      "loss": 0.0,
      "num_input_tokens_seen": 16133784,
      "step": 28400
    },
    {
      "epoch": 498.24778761061947,
      "eval_loss": 0.8328125476837158,
      "eval_runtime": 0.9273,
      "eval_samples_per_second": 26.961,
      "eval_steps_per_second": 14.02,
      "num_input_tokens_seen": 16133784,
      "step": 28400
    },
    {
      "epoch": 498.3362831858407,
      "grad_norm": 1.4849216078971494e-08,
      "learning_rate": 0.05802671131179747,
      "loss": 0.0,
      "num_input_tokens_seen": 16137032,
      "step": 28405
    },
    {
      "epoch": 498.42477876106193,
      "grad_norm": 3.991146613202545e-08,
      "learning_rate": 0.057980185777001154,
      "loss": 0.0,
      "num_input_tokens_seen": 16139944,
      "step": 28410
    },
    {
      "epoch": 498.5132743362832,
      "grad_norm": 1.8264449508365033e-08,
      "learning_rate": 0.057933674432816606,
      "loss": 0.0,
      "num_input_tokens_seen": 16142856,
      "step": 28415
    },
    {
      "epoch": 498.60176991150445,
      "grad_norm": 2.6703254718540848e-08,
      "learning_rate": 0.05788717728641648,
      "loss": 0.0,
      "num_input_tokens_seen": 16145848,
      "step": 28420
    },
    {
      "epoch": 498.69026548672565,
      "grad_norm": 3.6586325080634197e-08,
      "learning_rate": 0.057840694344971126,
      "loss": 0.0,
      "num_input_tokens_seen": 16148488,
      "step": 28425
    },
    {
      "epoch": 498.7787610619469,
      "grad_norm": 2.8330678247812102e-08,
      "learning_rate": 0.0577942256156489,
      "loss": 0.0,
      "num_input_tokens_seen": 16151384,
      "step": 28430
    },
    {
      "epoch": 498.86725663716817,
      "grad_norm": 2.2354765150112144e-08,
      "learning_rate": 0.057747771105615804,
      "loss": 0.0,
      "num_input_tokens_seen": 16154248,
      "step": 28435
    },
    {
      "epoch": 498.95575221238937,
      "grad_norm": 2.8928404560701892e-08,
      "learning_rate": 0.05770133082203568,
      "loss": 0.0,
      "num_input_tokens_seen": 16157208,
      "step": 28440
    },
    {
      "epoch": 499.0353982300885,
      "grad_norm": 2.0818939461264563e-08,
      "learning_rate": 0.0576549047720703,
      "loss": 0.0,
      "num_input_tokens_seen": 16159192,
      "step": 28445
    },
    {
      "epoch": 499.12389380530976,
      "grad_norm": 2.9270646351164942e-08,
      "learning_rate": 0.05760849296287902,
      "loss": 0.0,
      "num_input_tokens_seen": 16161672,
      "step": 28450
    },
    {
      "epoch": 499.21238938053096,
      "grad_norm": 4.931879260539063e-08,
      "learning_rate": 0.05756209540161919,
      "loss": 0.0,
      "num_input_tokens_seen": 16164632,
      "step": 28455
    },
    {
      "epoch": 499.3008849557522,
      "grad_norm": 3.217629895857499e-08,
      "learning_rate": 0.05751571209544595,
      "loss": 0.0,
      "num_input_tokens_seen": 16167432,
      "step": 28460
    },
    {
      "epoch": 499.3893805309734,
      "grad_norm": 2.0592622718140774e-08,
      "learning_rate": 0.057469343051512085,
      "loss": 0.0,
      "num_input_tokens_seen": 16170008,
      "step": 28465
    },
    {
      "epoch": 499.4778761061947,
      "grad_norm": 2.5735415576377818e-08,
      "learning_rate": 0.057422988276968324,
      "loss": 0.0,
      "num_input_tokens_seen": 16172872,
      "step": 28470
    },
    {
      "epoch": 499.56637168141594,
      "grad_norm": 2.3149690164814274e-08,
      "learning_rate": 0.05737664777896323,
      "loss": 0.0,
      "num_input_tokens_seen": 16176408,
      "step": 28475
    },
    {
      "epoch": 499.65486725663715,
      "grad_norm": 2.550867961303993e-08,
      "learning_rate": 0.057330321564642975,
      "loss": 0.0,
      "num_input_tokens_seen": 16179272,
      "step": 28480
    },
    {
      "epoch": 499.7433628318584,
      "grad_norm": 3.6891190546839425e-08,
      "learning_rate": 0.05728400964115174,
      "loss": 0.0,
      "num_input_tokens_seen": 16182472,
      "step": 28485
    },
    {
      "epoch": 499.83185840707966,
      "grad_norm": 2.0965389424532077e-08,
      "learning_rate": 0.057237712015631305,
      "loss": 0.0,
      "num_input_tokens_seen": 16185496,
      "step": 28490
    },
    {
      "epoch": 499.92035398230087,
      "grad_norm": 1.5196134128814265e-08,
      "learning_rate": 0.057191428695221425,
      "loss": 0.0,
      "num_input_tokens_seen": 16188728,
      "step": 28495
    },
    {
      "epoch": 500.0,
      "grad_norm": 3.021751382448201e-08,
      "learning_rate": 0.05714515968705958,
      "loss": 0.0,
      "num_input_tokens_seen": 16190856,
      "step": 28500
    },
    {
      "epoch": 500.08849557522126,
      "grad_norm": 1.1447442105350092e-08,
      "learning_rate": 0.05709890499828099,
      "loss": 0.0,
      "num_input_tokens_seen": 16194104,
      "step": 28505
    },
    {
      "epoch": 500.17699115044246,
      "grad_norm": 2.6149480802928338e-08,
      "learning_rate": 0.05705266463601868,
      "loss": 0.0,
      "num_input_tokens_seen": 16196776,
      "step": 28510
    },
    {
      "epoch": 500.2654867256637,
      "grad_norm": 3.982164642479802e-08,
      "learning_rate": 0.057006438607403565,
      "loss": 0.0,
      "num_input_tokens_seen": 16199848,
      "step": 28515
    },
    {
      "epoch": 500.353982300885,
      "grad_norm": 1.975698182832275e-08,
      "learning_rate": 0.056960226919564205,
      "loss": 0.0,
      "num_input_tokens_seen": 16202504,
      "step": 28520
    },
    {
      "epoch": 500.4424778761062,
      "grad_norm": 2.6764048755012482e-08,
      "learning_rate": 0.05691402957962713,
      "loss": 0.0,
      "num_input_tokens_seen": 16205448,
      "step": 28525
    },
    {
      "epoch": 500.53097345132744,
      "grad_norm": 3.057009223539353e-08,
      "learning_rate": 0.05686784659471642,
      "loss": 0.0,
      "num_input_tokens_seen": 16208456,
      "step": 28530
    },
    {
      "epoch": 500.6194690265487,
      "grad_norm": 4.706728162773288e-08,
      "learning_rate": 0.056821677971954136,
      "loss": 0.0,
      "num_input_tokens_seen": 16211032,
      "step": 28535
    },
    {
      "epoch": 500.7079646017699,
      "grad_norm": 1.940224159113768e-08,
      "learning_rate": 0.05677552371846012,
      "loss": 0.0,
      "num_input_tokens_seen": 16214136,
      "step": 28540
    },
    {
      "epoch": 500.79646017699116,
      "grad_norm": 3.322316288745242e-08,
      "learning_rate": 0.05672938384135182,
      "loss": 0.0,
      "num_input_tokens_seen": 16216984,
      "step": 28545
    },
    {
      "epoch": 500.88495575221236,
      "grad_norm": 2.349199590412354e-08,
      "learning_rate": 0.05668325834774465,
      "loss": 0.0,
      "num_input_tokens_seen": 16220072,
      "step": 28550
    },
    {
      "epoch": 500.9734513274336,
      "grad_norm": 2.1922977211374928e-08,
      "learning_rate": 0.05663714724475177,
      "loss": 0.0,
      "num_input_tokens_seen": 16222648,
      "step": 28555
    },
    {
      "epoch": 501.05309734513276,
      "grad_norm": 4.224552085929645e-08,
      "learning_rate": 0.05659105053948403,
      "loss": 0.0,
      "num_input_tokens_seen": 16225112,
      "step": 28560
    },
    {
      "epoch": 501.14159292035396,
      "grad_norm": 2.9150140079536868e-08,
      "learning_rate": 0.056544968239050176,
      "loss": 0.0,
      "num_input_tokens_seen": 16228200,
      "step": 28565
    },
    {
      "epoch": 501.2300884955752,
      "grad_norm": 6.210702707676319e-08,
      "learning_rate": 0.056498900350556616,
      "loss": 0.0,
      "num_input_tokens_seen": 16231048,
      "step": 28570
    },
    {
      "epoch": 501.3185840707965,
      "grad_norm": 3.161594008815882e-08,
      "learning_rate": 0.05645284688110766,
      "loss": 0.0,
      "num_input_tokens_seen": 16233848,
      "step": 28575
    },
    {
      "epoch": 501.4070796460177,
      "grad_norm": 2.6499064276208628e-08,
      "learning_rate": 0.05640680783780532,
      "loss": 0.0,
      "num_input_tokens_seen": 16236648,
      "step": 28580
    },
    {
      "epoch": 501.49557522123894,
      "grad_norm": 1.4709315543370849e-08,
      "learning_rate": 0.056360783227749324,
      "loss": 0.0,
      "num_input_tokens_seen": 16239512,
      "step": 28585
    },
    {
      "epoch": 501.5840707964602,
      "grad_norm": 7.689310166369978e-08,
      "learning_rate": 0.05631477305803728,
      "loss": 0.0,
      "num_input_tokens_seen": 16242424,
      "step": 28590
    },
    {
      "epoch": 501.6725663716814,
      "grad_norm": 2.3818163441546858e-08,
      "learning_rate": 0.05626877733576462,
      "loss": 0.0,
      "num_input_tokens_seen": 16245352,
      "step": 28595
    },
    {
      "epoch": 501.76106194690266,
      "grad_norm": 2.467539594874779e-08,
      "learning_rate": 0.05622279606802435,
      "loss": 0.0,
      "num_input_tokens_seen": 16248200,
      "step": 28600
    },
    {
      "epoch": 501.76106194690266,
      "eval_loss": 0.818134605884552,
      "eval_runtime": 0.9336,
      "eval_samples_per_second": 26.778,
      "eval_steps_per_second": 13.924,
      "num_input_tokens_seen": 16248200,
      "step": 28600
    },
    {
      "epoch": 501.8495575221239,
      "grad_norm": 1.694440321386992e-08,
      "learning_rate": 0.05617682926190744,
      "loss": 0.0,
      "num_input_tokens_seen": 16251432,
      "step": 28605
    },
    {
      "epoch": 501.9380530973451,
      "grad_norm": 3.18505790630752e-08,
      "learning_rate": 0.05613087692450248,
      "loss": 0.0,
      "num_input_tokens_seen": 16254008,
      "step": 28610
    },
    {
      "epoch": 502.01769911504425,
      "grad_norm": 2.2534404564567012e-08,
      "learning_rate": 0.05608493906289592,
      "loss": 0.0,
      "num_input_tokens_seen": 16256376,
      "step": 28615
    },
    {
      "epoch": 502.1061946902655,
      "grad_norm": 2.131108445269092e-08,
      "learning_rate": 0.05603901568417201,
      "loss": 0.0,
      "num_input_tokens_seen": 16258984,
      "step": 28620
    },
    {
      "epoch": 502.1946902654867,
      "grad_norm": 4.512608242634997e-08,
      "learning_rate": 0.055993106795412625,
      "loss": 0.0,
      "num_input_tokens_seen": 16261784,
      "step": 28625
    },
    {
      "epoch": 502.283185840708,
      "grad_norm": 2.373664464982994e-08,
      "learning_rate": 0.05594721240369759,
      "loss": 0.0,
      "num_input_tokens_seen": 16264840,
      "step": 28630
    },
    {
      "epoch": 502.37168141592923,
      "grad_norm": 3.599945586074682e-08,
      "learning_rate": 0.055901332516104296,
      "loss": 0.0,
      "num_input_tokens_seen": 16267768,
      "step": 28635
    },
    {
      "epoch": 502.46017699115043,
      "grad_norm": 2.127608134117054e-08,
      "learning_rate": 0.05585546713970804,
      "loss": 0.0,
      "num_input_tokens_seen": 16270488,
      "step": 28640
    },
    {
      "epoch": 502.5486725663717,
      "grad_norm": 2.104763119348263e-08,
      "learning_rate": 0.05580961628158189,
      "loss": 0.0,
      "num_input_tokens_seen": 16273240,
      "step": 28645
    },
    {
      "epoch": 502.6371681415929,
      "grad_norm": 3.63764414146317e-08,
      "learning_rate": 0.05576377994879659,
      "loss": 0.0,
      "num_input_tokens_seen": 16276024,
      "step": 28650
    },
    {
      "epoch": 502.72566371681415,
      "grad_norm": 7.436902205171236e-09,
      "learning_rate": 0.05571795814842063,
      "loss": 0.0,
      "num_input_tokens_seen": 16278952,
      "step": 28655
    },
    {
      "epoch": 502.8141592920354,
      "grad_norm": 3.817187987920079e-08,
      "learning_rate": 0.05567215088752037,
      "loss": 0.0,
      "num_input_tokens_seen": 16282248,
      "step": 28660
    },
    {
      "epoch": 502.9026548672566,
      "grad_norm": 4.1582907783777046e-08,
      "learning_rate": 0.05562635817315981,
      "loss": 0.0,
      "num_input_tokens_seen": 16285032,
      "step": 28665
    },
    {
      "epoch": 502.9911504424779,
      "grad_norm": 2.5961027105836365e-08,
      "learning_rate": 0.05558058001240083,
      "loss": 0.0,
      "num_input_tokens_seen": 16287864,
      "step": 28670
    },
    {
      "epoch": 503.070796460177,
      "grad_norm": 1.8941047841281033e-08,
      "learning_rate": 0.055534816412302915,
      "loss": 0.0,
      "num_input_tokens_seen": 16290344,
      "step": 28675
    },
    {
      "epoch": 503.1592920353982,
      "grad_norm": 2.0407469492056407e-08,
      "learning_rate": 0.055489067379923436,
      "loss": 0.0,
      "num_input_tokens_seen": 16292936,
      "step": 28680
    },
    {
      "epoch": 503.24778761061947,
      "grad_norm": 4.07561024928782e-08,
      "learning_rate": 0.055443332922317505,
      "loss": 0.0,
      "num_input_tokens_seen": 16296264,
      "step": 28685
    },
    {
      "epoch": 503.3362831858407,
      "grad_norm": 4.481668725020427e-08,
      "learning_rate": 0.055397613046537876,
      "loss": 0.0,
      "num_input_tokens_seen": 16298648,
      "step": 28690
    },
    {
      "epoch": 503.42477876106193,
      "grad_norm": 3.546748672533795e-08,
      "learning_rate": 0.055351907759635145,
      "loss": 0.0,
      "num_input_tokens_seen": 16301512,
      "step": 28695
    },
    {
      "epoch": 503.5132743362832,
      "grad_norm": 1.4936768266693434e-08,
      "learning_rate": 0.05530621706865772,
      "loss": 0.0,
      "num_input_tokens_seen": 16304440,
      "step": 28700
    },
    {
      "epoch": 503.60176991150445,
      "grad_norm": 1.9022813546598627e-08,
      "learning_rate": 0.055260540980651564,
      "loss": 0.0,
      "num_input_tokens_seen": 16307096,
      "step": 28705
    },
    {
      "epoch": 503.69026548672565,
      "grad_norm": 3.676198190305513e-08,
      "learning_rate": 0.05521487950266062,
      "loss": 0.0,
      "num_input_tokens_seen": 16310216,
      "step": 28710
    },
    {
      "epoch": 503.7787610619469,
      "grad_norm": 1.7432252974458606e-08,
      "learning_rate": 0.055169232641726344,
      "loss": 0.0,
      "num_input_tokens_seen": 16312904,
      "step": 28715
    },
    {
      "epoch": 503.86725663716817,
      "grad_norm": 4.344085979823831e-08,
      "learning_rate": 0.055123600404888166,
      "loss": 0.0,
      "num_input_tokens_seen": 16316040,
      "step": 28720
    },
    {
      "epoch": 503.95575221238937,
      "grad_norm": 1.5790016405503593e-08,
      "learning_rate": 0.05507798279918309,
      "loss": 0.0,
      "num_input_tokens_seen": 16319224,
      "step": 28725
    },
    {
      "epoch": 504.0353982300885,
      "grad_norm": 2.304961732590982e-08,
      "learning_rate": 0.0550323798316459,
      "loss": 0.0,
      "num_input_tokens_seen": 16321928,
      "step": 28730
    },
    {
      "epoch": 504.12389380530976,
      "grad_norm": 1.9296637177035336e-08,
      "learning_rate": 0.05498679150930916,
      "loss": 0.0,
      "num_input_tokens_seen": 16324424,
      "step": 28735
    },
    {
      "epoch": 504.21238938053096,
      "grad_norm": 1.720329123600095e-08,
      "learning_rate": 0.05494121783920323,
      "loss": 0.0,
      "num_input_tokens_seen": 16327096,
      "step": 28740
    },
    {
      "epoch": 504.3008849557522,
      "grad_norm": 3.848321483701511e-08,
      "learning_rate": 0.05489565882835605,
      "loss": 0.0,
      "num_input_tokens_seen": 16329848,
      "step": 28745
    },
    {
      "epoch": 504.3893805309734,
      "grad_norm": 1.5898670824299188e-08,
      "learning_rate": 0.05485011448379348,
      "loss": 0.0,
      "num_input_tokens_seen": 16332984,
      "step": 28750
    },
    {
      "epoch": 504.4778761061947,
      "grad_norm": 4.4323787307121165e-08,
      "learning_rate": 0.05480458481253893,
      "loss": 0.0,
      "num_input_tokens_seen": 16335496,
      "step": 28755
    },
    {
      "epoch": 504.56637168141594,
      "grad_norm": 9.92130289034776e-09,
      "learning_rate": 0.054759069821613715,
      "loss": 0.0,
      "num_input_tokens_seen": 16338520,
      "step": 28760
    },
    {
      "epoch": 504.65486725663715,
      "grad_norm": 2.261002229886344e-08,
      "learning_rate": 0.05471356951803683,
      "loss": 0.0,
      "num_input_tokens_seen": 16341640,
      "step": 28765
    },
    {
      "epoch": 504.7433628318584,
      "grad_norm": 1.652521319783773e-08,
      "learning_rate": 0.054668083908824945,
      "loss": 0.0,
      "num_input_tokens_seen": 16344840,
      "step": 28770
    },
    {
      "epoch": 504.83185840707966,
      "grad_norm": 2.936182141866084e-08,
      "learning_rate": 0.054622613000992526,
      "loss": 0.0,
      "num_input_tokens_seen": 16347896,
      "step": 28775
    },
    {
      "epoch": 504.92035398230087,
      "grad_norm": 2.476798321993101e-08,
      "learning_rate": 0.05457715680155182,
      "loss": 0.0,
      "num_input_tokens_seen": 16350840,
      "step": 28780
    },
    {
      "epoch": 505.0,
      "grad_norm": 1.8739035567705287e-07,
      "learning_rate": 0.05453171531751265,
      "loss": 0.0,
      "num_input_tokens_seen": 16353448,
      "step": 28785
    },
    {
      "epoch": 505.08849557522126,
      "grad_norm": 6.131436691703129e-09,
      "learning_rate": 0.05448628855588276,
      "loss": 0.0,
      "num_input_tokens_seen": 16356024,
      "step": 28790
    },
    {
      "epoch": 505.17699115044246,
      "grad_norm": 4.4934754583891845e-08,
      "learning_rate": 0.05444087652366746,
      "loss": 0.0,
      "num_input_tokens_seen": 16359016,
      "step": 28795
    },
    {
      "epoch": 505.2654867256637,
      "grad_norm": 1.5114716589437194e-08,
      "learning_rate": 0.05439547922786984,
      "loss": 0.0,
      "num_input_tokens_seen": 16361560,
      "step": 28800
    },
    {
      "epoch": 505.2654867256637,
      "eval_loss": 0.8467451333999634,
      "eval_runtime": 0.9417,
      "eval_samples_per_second": 26.549,
      "eval_steps_per_second": 13.805,
      "num_input_tokens_seen": 16361560,
      "step": 28800
    },
    {
      "epoch": 505.353982300885,
      "grad_norm": 2.672634558109621e-08,
      "learning_rate": 0.0543500966754908,
      "loss": 0.0,
      "num_input_tokens_seen": 16364184,
      "step": 28805
    },
    {
      "epoch": 505.4424778761062,
      "grad_norm": 3.228395684118368e-08,
      "learning_rate": 0.05430472887352882,
      "loss": 0.0,
      "num_input_tokens_seen": 16367160,
      "step": 28810
    },
    {
      "epoch": 505.53097345132744,
      "grad_norm": 2.6152303433946145e-08,
      "learning_rate": 0.05425937582898023,
      "loss": 0.0,
      "num_input_tokens_seen": 16370264,
      "step": 28815
    },
    {
      "epoch": 505.6194690265487,
      "grad_norm": 2.8168180676857446e-08,
      "learning_rate": 0.054214037548839085,
      "loss": 0.0,
      "num_input_tokens_seen": 16373224,
      "step": 28820
    },
    {
      "epoch": 505.7079646017699,
      "grad_norm": 3.929902092636439e-08,
      "learning_rate": 0.05416871404009703,
      "loss": 0.0,
      "num_input_tokens_seen": 16375992,
      "step": 28825
    },
    {
      "epoch": 505.79646017699116,
      "grad_norm": 2.5480760612595077e-08,
      "learning_rate": 0.054123405309743605,
      "loss": 0.0,
      "num_input_tokens_seen": 16379000,
      "step": 28830
    },
    {
      "epoch": 505.88495575221236,
      "grad_norm": 2.3334717269563043e-08,
      "learning_rate": 0.0540781113647659,
      "loss": 0.0,
      "num_input_tokens_seen": 16381640,
      "step": 28835
    },
    {
      "epoch": 505.9734513274336,
      "grad_norm": 1.1558757506691109e-08,
      "learning_rate": 0.054032832212148836,
      "loss": 0.0,
      "num_input_tokens_seen": 16384680,
      "step": 28840
    },
    {
      "epoch": 506.05309734513276,
      "grad_norm": 3.1090152674551064e-08,
      "learning_rate": 0.0539875678588751,
      "loss": 0.0,
      "num_input_tokens_seen": 16387568,
      "step": 28845
    },
    {
      "epoch": 506.14159292035396,
      "grad_norm": 2.8511927041563467e-08,
      "learning_rate": 0.05394231831192492,
      "loss": 0.0,
      "num_input_tokens_seen": 16390688,
      "step": 28850
    },
    {
      "epoch": 506.2300884955752,
      "grad_norm": 8.384094485336391e-08,
      "learning_rate": 0.05389708357827639,
      "loss": 0.0,
      "num_input_tokens_seen": 16393184,
      "step": 28855
    },
    {
      "epoch": 506.3185840707965,
      "grad_norm": 2.509328922428722e-08,
      "learning_rate": 0.05385186366490533,
      "loss": 0.0,
      "num_input_tokens_seen": 16396352,
      "step": 28860
    },
    {
      "epoch": 506.4070796460177,
      "grad_norm": 2.4643224350029413e-08,
      "learning_rate": 0.053806658578785166,
      "loss": 0.0,
      "num_input_tokens_seen": 16399376,
      "step": 28865
    },
    {
      "epoch": 506.49557522123894,
      "grad_norm": 2.6407469988498633e-08,
      "learning_rate": 0.05376146832688705,
      "loss": 0.0,
      "num_input_tokens_seen": 16402032,
      "step": 28870
    },
    {
      "epoch": 506.5840707964602,
      "grad_norm": 2.2452860903854344e-08,
      "learning_rate": 0.053716292916179964,
      "loss": 0.0,
      "num_input_tokens_seen": 16405216,
      "step": 28875
    },
    {
      "epoch": 506.6725663716814,
      "grad_norm": 4.3088057566365023e-08,
      "learning_rate": 0.05367113235363045,
      "loss": 0.0,
      "num_input_tokens_seen": 16408240,
      "step": 28880
    },
    {
      "epoch": 506.76106194690266,
      "grad_norm": 3.5525879127362714e-08,
      "learning_rate": 0.05362598664620289,
      "loss": 0.0,
      "num_input_tokens_seen": 16410608,
      "step": 28885
    },
    {
      "epoch": 506.8495575221239,
      "grad_norm": 3.1250678489413986e-08,
      "learning_rate": 0.053580855800859285,
      "loss": 0.0,
      "num_input_tokens_seen": 16413552,
      "step": 28890
    },
    {
      "epoch": 506.9380530973451,
      "grad_norm": 1.1197639260274173e-08,
      "learning_rate": 0.05353573982455938,
      "loss": 0.0,
      "num_input_tokens_seen": 16416544,
      "step": 28895
    },
    {
      "epoch": 507.01769911504425,
      "grad_norm": 1.572987429199202e-08,
      "learning_rate": 0.053490638724260686,
      "loss": 0.0,
      "num_input_tokens_seen": 16419008,
      "step": 28900
    },
    {
      "epoch": 507.1061946902655,
      "grad_norm": 3.269613557677076e-08,
      "learning_rate": 0.05344555250691827,
      "loss": 0.0,
      "num_input_tokens_seen": 16422096,
      "step": 28905
    },
    {
      "epoch": 507.1946902654867,
      "grad_norm": 2.189130654528526e-08,
      "learning_rate": 0.053400481179485086,
      "loss": 0.0,
      "num_input_tokens_seen": 16424464,
      "step": 28910
    },
    {
      "epoch": 507.283185840708,
      "grad_norm": 2.6447834144960325e-08,
      "learning_rate": 0.05335542474891159,
      "loss": 0.0,
      "num_input_tokens_seen": 16427456,
      "step": 28915
    },
    {
      "epoch": 507.37168141592923,
      "grad_norm": 2.484727268381448e-08,
      "learning_rate": 0.053310383222146124,
      "loss": 0.0,
      "num_input_tokens_seen": 16429840,
      "step": 28920
    },
    {
      "epoch": 507.46017699115043,
      "grad_norm": 2.6976909595077814e-08,
      "learning_rate": 0.053265356606134684,
      "loss": 0.0,
      "num_input_tokens_seen": 16432976,
      "step": 28925
    },
    {
      "epoch": 507.5486725663717,
      "grad_norm": 2.7755039155863415e-08,
      "learning_rate": 0.053220344907820856,
      "loss": 0.0,
      "num_input_tokens_seen": 16435584,
      "step": 28930
    },
    {
      "epoch": 507.6371681415929,
      "grad_norm": 2.863412440490265e-08,
      "learning_rate": 0.05317534813414608,
      "loss": 0.0,
      "num_input_tokens_seen": 16438512,
      "step": 28935
    },
    {
      "epoch": 507.72566371681415,
      "grad_norm": 2.9745237384304346e-08,
      "learning_rate": 0.05313036629204942,
      "loss": 0.0,
      "num_input_tokens_seen": 16441184,
      "step": 28940
    },
    {
      "epoch": 507.8141592920354,
      "grad_norm": 2.8400426899111153e-08,
      "learning_rate": 0.05308539938846756,
      "loss": 0.0,
      "num_input_tokens_seen": 16443776,
      "step": 28945
    },
    {
      "epoch": 507.9026548672566,
      "grad_norm": 8.082785996066377e-09,
      "learning_rate": 0.05304044743033507,
      "loss": 0.0,
      "num_input_tokens_seen": 16446544,
      "step": 28950
    },
    {
      "epoch": 507.9911504424779,
      "grad_norm": 1.980573394178009e-08,
      "learning_rate": 0.05299551042458401,
      "loss": 0.0,
      "num_input_tokens_seen": 16450080,
      "step": 28955
    },
    {
      "epoch": 508.070796460177,
      "grad_norm": 1.5059377744819358e-08,
      "learning_rate": 0.052950588378144266,
      "loss": 0.0,
      "num_input_tokens_seen": 16452536,
      "step": 28960
    },
    {
      "epoch": 508.1592920353982,
      "grad_norm": 4.387280228002055e-08,
      "learning_rate": 0.052905681297943465,
      "loss": 0.0,
      "num_input_tokens_seen": 16455144,
      "step": 28965
    },
    {
      "epoch": 508.24778761061947,
      "grad_norm": 3.6386680335454e-08,
      "learning_rate": 0.0528607891909067,
      "loss": 0.0,
      "num_input_tokens_seen": 16458168,
      "step": 28970
    },
    {
      "epoch": 508.3362831858407,
      "grad_norm": 2.744015503708397e-08,
      "learning_rate": 0.05281591206395697,
      "loss": 0.0,
      "num_input_tokens_seen": 16461112,
      "step": 28975
    },
    {
      "epoch": 508.42477876106193,
      "grad_norm": 2.9299224379997213e-08,
      "learning_rate": 0.05277104992401496,
      "loss": 0.0,
      "num_input_tokens_seen": 16464376,
      "step": 28980
    },
    {
      "epoch": 508.5132743362832,
      "grad_norm": 3.4909103163727195e-08,
      "learning_rate": 0.05272620277799884,
      "loss": 0.0,
      "num_input_tokens_seen": 16467480,
      "step": 28985
    },
    {
      "epoch": 508.60176991150445,
      "grad_norm": 3.998922082359968e-08,
      "learning_rate": 0.05268137063282473,
      "loss": 0.0,
      "num_input_tokens_seen": 16470344,
      "step": 28990
    },
    {
      "epoch": 508.69026548672565,
      "grad_norm": 4.239520379201167e-08,
      "learning_rate": 0.0526365534954062,
      "loss": 0.0,
      "num_input_tokens_seen": 16472792,
      "step": 28995
    },
    {
      "epoch": 508.7787610619469,
      "grad_norm": 3.674533743946995e-08,
      "learning_rate": 0.052591751372654656,
      "loss": 0.0,
      "num_input_tokens_seen": 16475624,
      "step": 29000
    },
    {
      "epoch": 508.7787610619469,
      "eval_loss": 0.844280481338501,
      "eval_runtime": 0.9382,
      "eval_samples_per_second": 26.647,
      "eval_steps_per_second": 13.857,
      "num_input_tokens_seen": 16475624,
      "step": 29000
    },
    {
      "epoch": 508.86725663716817,
      "grad_norm": 2.8284812714218788e-08,
      "learning_rate": 0.05254696427147921,
      "loss": 0.0,
      "num_input_tokens_seen": 16478728,
      "step": 29005
    },
    {
      "epoch": 508.95575221238937,
      "grad_norm": 2.916135777297768e-08,
      "learning_rate": 0.052502192198786546,
      "loss": 0.0,
      "num_input_tokens_seen": 16481352,
      "step": 29010
    },
    {
      "epoch": 509.0353982300885,
      "grad_norm": 2.9801244139093797e-08,
      "learning_rate": 0.05245743516148103,
      "loss": 0.0,
      "num_input_tokens_seen": 16483560,
      "step": 29015
    },
    {
      "epoch": 509.12389380530976,
      "grad_norm": 1.0416578710703561e-08,
      "learning_rate": 0.05241269316646486,
      "loss": 0.0,
      "num_input_tokens_seen": 16486200,
      "step": 29020
    },
    {
      "epoch": 509.21238938053096,
      "grad_norm": 2.9459432226985882e-08,
      "learning_rate": 0.052367966220637725,
      "loss": 0.0,
      "num_input_tokens_seen": 16488856,
      "step": 29025
    },
    {
      "epoch": 509.3008849557522,
      "grad_norm": 3.927716818452609e-08,
      "learning_rate": 0.05232325433089716,
      "loss": 0.0,
      "num_input_tokens_seen": 16491672,
      "step": 29030
    },
    {
      "epoch": 509.3893805309734,
      "grad_norm": 8.482283675448343e-08,
      "learning_rate": 0.052278557504138214,
      "loss": 0.0,
      "num_input_tokens_seen": 16494904,
      "step": 29035
    },
    {
      "epoch": 509.4778761061947,
      "grad_norm": 4.7782826584352733e-08,
      "learning_rate": 0.05223387574725372,
      "loss": 0.0,
      "num_input_tokens_seen": 16497816,
      "step": 29040
    },
    {
      "epoch": 509.56637168141594,
      "grad_norm": 3.986857777249497e-08,
      "learning_rate": 0.05218920906713428,
      "loss": 0.0,
      "num_input_tokens_seen": 16500984,
      "step": 29045
    },
    {
      "epoch": 509.65486725663715,
      "grad_norm": 2.0131782463295167e-08,
      "learning_rate": 0.05214455747066789,
      "loss": 0.0,
      "num_input_tokens_seen": 16503864,
      "step": 29050
    },
    {
      "epoch": 509.7433628318584,
      "grad_norm": 1.9117724292527782e-08,
      "learning_rate": 0.05209992096474048,
      "loss": 0.0,
      "num_input_tokens_seen": 16507080,
      "step": 29055
    },
    {
      "epoch": 509.83185840707966,
      "grad_norm": 2.2873281935176237e-08,
      "learning_rate": 0.05205529955623559,
      "loss": 0.0,
      "num_input_tokens_seen": 16509768,
      "step": 29060
    },
    {
      "epoch": 509.92035398230087,
      "grad_norm": 3.72949067184436e-08,
      "learning_rate": 0.052010693252034314,
      "loss": 0.0,
      "num_input_tokens_seen": 16512296,
      "step": 29065
    },
    {
      "epoch": 510.0,
      "grad_norm": 8.723119293563286e-08,
      "learning_rate": 0.0519661020590156,
      "loss": 0.0,
      "num_input_tokens_seen": 16514688,
      "step": 29070
    },
    {
      "epoch": 510.08849557522126,
      "grad_norm": 1.6141571634875618e-08,
      "learning_rate": 0.05192152598405586,
      "loss": 0.0,
      "num_input_tokens_seen": 16517760,
      "step": 29075
    },
    {
      "epoch": 510.17699115044246,
      "grad_norm": 2.649319696956809e-08,
      "learning_rate": 0.05187696503402941,
      "loss": 0.0,
      "num_input_tokens_seen": 16520672,
      "step": 29080
    },
    {
      "epoch": 510.2654867256637,
      "grad_norm": 2.7812822267492265e-08,
      "learning_rate": 0.05183241921580798,
      "loss": 0.0,
      "num_input_tokens_seen": 16523504,
      "step": 29085
    },
    {
      "epoch": 510.353982300885,
      "grad_norm": 4.02806143995349e-08,
      "learning_rate": 0.051787888536261206,
      "loss": 0.0,
      "num_input_tokens_seen": 16526496,
      "step": 29090
    },
    {
      "epoch": 510.4424778761062,
      "grad_norm": 2.1580598641435245e-08,
      "learning_rate": 0.051743373002256184,
      "loss": 0.0,
      "num_input_tokens_seen": 16529376,
      "step": 29095
    },
    {
      "epoch": 510.53097345132744,
      "grad_norm": 1.849710784540548e-08,
      "learning_rate": 0.05169887262065787,
      "loss": 0.0,
      "num_input_tokens_seen": 16532480,
      "step": 29100
    },
    {
      "epoch": 510.6194690265487,
      "grad_norm": 2.0600948502647043e-08,
      "learning_rate": 0.051654387398328665,
      "loss": 0.0,
      "num_input_tokens_seen": 16535088,
      "step": 29105
    },
    {
      "epoch": 510.7079646017699,
      "grad_norm": 1.8335377660605445e-08,
      "learning_rate": 0.05160991734212888,
      "loss": 0.0,
      "num_input_tokens_seen": 16538384,
      "step": 29110
    },
    {
      "epoch": 510.79646017699116,
      "grad_norm": 2.7240655953164605e-08,
      "learning_rate": 0.051565462458916224,
      "loss": 0.0,
      "num_input_tokens_seen": 16541168,
      "step": 29115
    },
    {
      "epoch": 510.88495575221236,
      "grad_norm": 8.116147753867153e-09,
      "learning_rate": 0.05152102275554627,
      "loss": 0.0,
      "num_input_tokens_seen": 16543808,
      "step": 29120
    },
    {
      "epoch": 510.9734513274336,
      "grad_norm": 1.8128849532672575e-08,
      "learning_rate": 0.05147659823887222,
      "loss": 0.0,
      "num_input_tokens_seen": 16546624,
      "step": 29125
    },
    {
      "epoch": 511.05309734513276,
      "grad_norm": 3.000089066063083e-08,
      "learning_rate": 0.05143218891574479,
      "loss": 0.0,
      "num_input_tokens_seen": 16548824,
      "step": 29130
    },
    {
      "epoch": 511.14159292035396,
      "grad_norm": 2.809193588859671e-08,
      "learning_rate": 0.0513877947930125,
      "loss": 0.0,
      "num_input_tokens_seen": 16551416,
      "step": 29135
    },
    {
      "epoch": 511.2300884955752,
      "grad_norm": 5.4165720797527683e-08,
      "learning_rate": 0.051343415877521566,
      "loss": 0.0,
      "num_input_tokens_seen": 16554536,
      "step": 29140
    },
    {
      "epoch": 511.3185840707965,
      "grad_norm": 3.113231272777739e-08,
      "learning_rate": 0.051299052176115634,
      "loss": 0.0,
      "num_input_tokens_seen": 16557032,
      "step": 29145
    },
    {
      "epoch": 511.4070796460177,
      "grad_norm": 1.5424014065956726e-08,
      "learning_rate": 0.051254703695636256,
      "loss": 0.0,
      "num_input_tokens_seen": 16559880,
      "step": 29150
    },
    {
      "epoch": 511.49557522123894,
      "grad_norm": 2.0386742960454285e-08,
      "learning_rate": 0.05121037044292249,
      "loss": 0.0,
      "num_input_tokens_seen": 16562824,
      "step": 29155
    },
    {
      "epoch": 511.5840707964602,
      "grad_norm": 2.165664270137313e-08,
      "learning_rate": 0.05116605242481101,
      "loss": 0.0,
      "num_input_tokens_seen": 16565800,
      "step": 29160
    },
    {
      "epoch": 511.6725663716814,
      "grad_norm": 3.212280930142697e-08,
      "learning_rate": 0.05112174964813634,
      "loss": 0.0,
      "num_input_tokens_seen": 16568408,
      "step": 29165
    },
    {
      "epoch": 511.76106194690266,
      "grad_norm": 2.8603329482734807e-08,
      "learning_rate": 0.05107746211973038,
      "loss": 0.0,
      "num_input_tokens_seen": 16571496,
      "step": 29170
    },
    {
      "epoch": 511.8495575221239,
      "grad_norm": 3.3619109274241055e-08,
      "learning_rate": 0.05103318984642291,
      "loss": 0.0,
      "num_input_tokens_seen": 16575064,
      "step": 29175
    },
    {
      "epoch": 511.9380530973451,
      "grad_norm": 1.4348926491436487e-08,
      "learning_rate": 0.05098893283504131,
      "loss": 0.0,
      "num_input_tokens_seen": 16577848,
      "step": 29180
    },
    {
      "epoch": 512.0176991150443,
      "grad_norm": 3.472871767939978e-08,
      "learning_rate": 0.050944691092410475,
      "loss": 0.0,
      "num_input_tokens_seen": 16580344,
      "step": 29185
    },
    {
      "epoch": 512.1061946902655,
      "grad_norm": 2.0656349519754258e-08,
      "learning_rate": 0.05090046462535313,
      "loss": 0.0,
      "num_input_tokens_seen": 16583176,
      "step": 29190
    },
    {
      "epoch": 512.1946902654868,
      "grad_norm": 2.828059741943889e-08,
      "learning_rate": 0.050856253440689454,
      "loss": 0.0,
      "num_input_tokens_seen": 16586280,
      "step": 29195
    },
    {
      "epoch": 512.2831858407079,
      "grad_norm": 1.752897915707763e-08,
      "learning_rate": 0.050812057545237405,
      "loss": 0.0,
      "num_input_tokens_seen": 16588984,
      "step": 29200
    },
    {
      "epoch": 512.2831858407079,
      "eval_loss": 0.8483059406280518,
      "eval_runtime": 0.9324,
      "eval_samples_per_second": 26.812,
      "eval_steps_per_second": 13.942,
      "num_input_tokens_seen": 16588984,
      "step": 29200
    },
    {
      "epoch": 512.3716814159292,
      "grad_norm": 1.4201051001805354e-08,
      "learning_rate": 0.0507678769458126,
      "loss": 0.0,
      "num_input_tokens_seen": 16591880,
      "step": 29205
    },
    {
      "epoch": 512.4601769911504,
      "grad_norm": 5.1089703134721276e-08,
      "learning_rate": 0.050723711649228155,
      "loss": 0.0,
      "num_input_tokens_seen": 16595352,
      "step": 29210
    },
    {
      "epoch": 512.5486725663717,
      "grad_norm": 2.5490205501910168e-08,
      "learning_rate": 0.05067956166229496,
      "loss": 0.0,
      "num_input_tokens_seen": 16597560,
      "step": 29215
    },
    {
      "epoch": 512.637168141593,
      "grad_norm": 3.743438270475963e-08,
      "learning_rate": 0.05063542699182155,
      "loss": 0.0,
      "num_input_tokens_seen": 16600440,
      "step": 29220
    },
    {
      "epoch": 512.7256637168142,
      "grad_norm": 2.9439753745919006e-08,
      "learning_rate": 0.050591307644613996,
      "loss": 0.0,
      "num_input_tokens_seen": 16603464,
      "step": 29225
    },
    {
      "epoch": 512.8141592920354,
      "grad_norm": 2.175112889801767e-08,
      "learning_rate": 0.05054720362747599,
      "loss": 0.0,
      "num_input_tokens_seen": 16606360,
      "step": 29230
    },
    {
      "epoch": 512.9026548672566,
      "grad_norm": 5.2740748657242875e-08,
      "learning_rate": 0.050503114947209035,
      "loss": 0.0,
      "num_input_tokens_seen": 16609224,
      "step": 29235
    },
    {
      "epoch": 512.9911504424779,
      "grad_norm": 3.6781099055360755e-08,
      "learning_rate": 0.05045904161061207,
      "loss": 0.0,
      "num_input_tokens_seen": 16612152,
      "step": 29240
    },
    {
      "epoch": 513.070796460177,
      "grad_norm": 4.0774597920290034e-08,
      "learning_rate": 0.05041498362448185,
      "loss": 0.0,
      "num_input_tokens_seen": 16614920,
      "step": 29245
    },
    {
      "epoch": 513.1592920353983,
      "grad_norm": 3.065213860509175e-08,
      "learning_rate": 0.05037094099561256,
      "loss": 0.0,
      "num_input_tokens_seen": 16617624,
      "step": 29250
    },
    {
      "epoch": 513.2477876106195,
      "grad_norm": 3.9994269229737256e-08,
      "learning_rate": 0.05032691373079624,
      "loss": 0.0,
      "num_input_tokens_seen": 16620296,
      "step": 29255
    },
    {
      "epoch": 513.3362831858407,
      "grad_norm": 3.560701955507284e-08,
      "learning_rate": 0.05028290183682234,
      "loss": 0.0,
      "num_input_tokens_seen": 16623352,
      "step": 29260
    },
    {
      "epoch": 513.4247787610619,
      "grad_norm": 3.003927773193027e-08,
      "learning_rate": 0.050238905320478096,
      "loss": 0.0,
      "num_input_tokens_seen": 16626120,
      "step": 29265
    },
    {
      "epoch": 513.5132743362832,
      "grad_norm": 5.296265470633443e-08,
      "learning_rate": 0.05019492418854838,
      "loss": 0.0,
      "num_input_tokens_seen": 16629112,
      "step": 29270
    },
    {
      "epoch": 513.6017699115044,
      "grad_norm": 1.1777527397782706e-08,
      "learning_rate": 0.05015095844781554,
      "loss": 0.0,
      "num_input_tokens_seen": 16631800,
      "step": 29275
    },
    {
      "epoch": 513.6902654867257,
      "grad_norm": 1.597134691166957e-08,
      "learning_rate": 0.05010700810505968,
      "loss": 0.0,
      "num_input_tokens_seen": 16634680,
      "step": 29280
    },
    {
      "epoch": 513.7787610619469,
      "grad_norm": 1.0867720057206043e-08,
      "learning_rate": 0.05006307316705856,
      "loss": 0.0,
      "num_input_tokens_seen": 16637736,
      "step": 29285
    },
    {
      "epoch": 513.8672566371681,
      "grad_norm": 2.098987117449269e-08,
      "learning_rate": 0.0500191536405874,
      "loss": 0.0,
      "num_input_tokens_seen": 16640120,
      "step": 29290
    },
    {
      "epoch": 513.9557522123894,
      "grad_norm": 3.402622894554952e-08,
      "learning_rate": 0.04997524953241922,
      "loss": 0.0,
      "num_input_tokens_seen": 16643400,
      "step": 29295
    },
    {
      "epoch": 514.0353982300885,
      "grad_norm": 2.429042211815613e-08,
      "learning_rate": 0.049931360849324556,
      "loss": 0.0,
      "num_input_tokens_seen": 16645576,
      "step": 29300
    },
    {
      "epoch": 514.1238938053098,
      "grad_norm": 3.951849336658597e-08,
      "learning_rate": 0.04988748759807155,
      "loss": 0.0,
      "num_input_tokens_seen": 16648184,
      "step": 29305
    },
    {
      "epoch": 514.212389380531,
      "grad_norm": 2.726404346731215e-08,
      "learning_rate": 0.0498436297854261,
      "loss": 0.0,
      "num_input_tokens_seen": 16650536,
      "step": 29310
    },
    {
      "epoch": 514.3008849557522,
      "grad_norm": 2.015318401049626e-08,
      "learning_rate": 0.04979978741815152,
      "loss": 0.0,
      "num_input_tokens_seen": 16653352,
      "step": 29315
    },
    {
      "epoch": 514.3893805309734,
      "grad_norm": 4.053950775073645e-08,
      "learning_rate": 0.04975596050300891,
      "loss": 0.0,
      "num_input_tokens_seen": 16656040,
      "step": 29320
    },
    {
      "epoch": 514.4778761061947,
      "grad_norm": 5.401480507316592e-08,
      "learning_rate": 0.049712149046757005,
      "loss": 0.0,
      "num_input_tokens_seen": 16659160,
      "step": 29325
    },
    {
      "epoch": 514.566371681416,
      "grad_norm": 3.0904171666179536e-08,
      "learning_rate": 0.04966835305615194,
      "loss": 0.0,
      "num_input_tokens_seen": 16662312,
      "step": 29330
    },
    {
      "epoch": 514.6548672566372,
      "grad_norm": 3.263906123152083e-08,
      "learning_rate": 0.049624572537947755,
      "loss": 0.0,
      "num_input_tokens_seen": 16665432,
      "step": 29335
    },
    {
      "epoch": 514.7433628318585,
      "grad_norm": 5.911403277991667e-08,
      "learning_rate": 0.04958080749889582,
      "loss": 0.0,
      "num_input_tokens_seen": 16668440,
      "step": 29340
    },
    {
      "epoch": 514.8318584070796,
      "grad_norm": 3.7967637922520225e-08,
      "learning_rate": 0.049537057945745304,
      "loss": 0.0,
      "num_input_tokens_seen": 16671416,
      "step": 29345
    },
    {
      "epoch": 514.9203539823009,
      "grad_norm": 4.673431419632834e-08,
      "learning_rate": 0.049493323885243,
      "loss": 0.0,
      "num_input_tokens_seen": 16674584,
      "step": 29350
    },
    {
      "epoch": 515.0,
      "grad_norm": 1.0012173845552752e-07,
      "learning_rate": 0.04944960532413318,
      "loss": 0.0,
      "num_input_tokens_seen": 16676800,
      "step": 29355
    },
    {
      "epoch": 515.0884955752213,
      "grad_norm": 2.979025026661475e-08,
      "learning_rate": 0.049405902269157774,
      "loss": 0.0,
      "num_input_tokens_seen": 16679280,
      "step": 29360
    },
    {
      "epoch": 515.1769911504425,
      "grad_norm": 1.896069257156796e-08,
      "learning_rate": 0.04936221472705646,
      "loss": 0.0,
      "num_input_tokens_seen": 16681936,
      "step": 29365
    },
    {
      "epoch": 515.2654867256637,
      "grad_norm": 7.168839744053912e-09,
      "learning_rate": 0.04931854270456632,
      "loss": 0.0,
      "num_input_tokens_seen": 16684608,
      "step": 29370
    },
    {
      "epoch": 515.3539823008849,
      "grad_norm": 3.334028875201511e-08,
      "learning_rate": 0.049274886208422075,
      "loss": 0.0,
      "num_input_tokens_seen": 16688048,
      "step": 29375
    },
    {
      "epoch": 515.4424778761062,
      "grad_norm": 3.951340232788425e-08,
      "learning_rate": 0.049231245245356235,
      "loss": 0.0,
      "num_input_tokens_seen": 16691232,
      "step": 29380
    },
    {
      "epoch": 515.5309734513274,
      "grad_norm": 6.65333246274713e-08,
      "learning_rate": 0.049187619822098655,
      "loss": 0.0,
      "num_input_tokens_seen": 16693840,
      "step": 29385
    },
    {
      "epoch": 515.6194690265487,
      "grad_norm": 2.4580064206247698e-08,
      "learning_rate": 0.04914400994537705,
      "loss": 0.0,
      "num_input_tokens_seen": 16696720,
      "step": 29390
    },
    {
      "epoch": 515.70796460177,
      "grad_norm": 1.3508534735251487e-08,
      "learning_rate": 0.049100415621916485,
      "loss": 0.0,
      "num_input_tokens_seen": 16699984,
      "step": 29395
    },
    {
      "epoch": 515.7964601769911,
      "grad_norm": 3.430937667303624e-08,
      "learning_rate": 0.04905683685843981,
      "loss": 0.0,
      "num_input_tokens_seen": 16702496,
      "step": 29400
    },
    {
      "epoch": 515.7964601769911,
      "eval_loss": 0.8532394766807556,
      "eval_runtime": 0.9436,
      "eval_samples_per_second": 26.494,
      "eval_steps_per_second": 13.777,
      "num_input_tokens_seen": 16702496,
      "step": 29400
    },
    {
      "epoch": 515.8849557522124,
      "grad_norm": 4.999023417440185e-08,
      "learning_rate": 0.049013273661667495,
      "loss": 0.0,
      "num_input_tokens_seen": 16705664,
      "step": 29405
    },
    {
      "epoch": 515.9734513274336,
      "grad_norm": 4.204308012845104e-08,
      "learning_rate": 0.048969726038317396,
      "loss": 0.0,
      "num_input_tokens_seen": 16708528,
      "step": 29410
    },
    {
      "epoch": 516.0530973451328,
      "grad_norm": 8.744005697280954e-08,
      "learning_rate": 0.048926193995105206,
      "loss": 0.0,
      "num_input_tokens_seen": 16710584,
      "step": 29415
    },
    {
      "epoch": 516.141592920354,
      "grad_norm": 1.793705450836569e-08,
      "learning_rate": 0.048882677538744035,
      "loss": 0.0,
      "num_input_tokens_seen": 16713224,
      "step": 29420
    },
    {
      "epoch": 516.2300884955753,
      "grad_norm": 5.983815754007082e-08,
      "learning_rate": 0.048839176675944715,
      "loss": 0.0,
      "num_input_tokens_seen": 16716648,
      "step": 29425
    },
    {
      "epoch": 516.3185840707964,
      "grad_norm": 6.715271183566074e-08,
      "learning_rate": 0.04879569141341566,
      "loss": 0.0,
      "num_input_tokens_seen": 16719608,
      "step": 29430
    },
    {
      "epoch": 516.4070796460177,
      "grad_norm": 2.9746280105769074e-08,
      "learning_rate": 0.04875222175786274,
      "loss": 0.0,
      "num_input_tokens_seen": 16722216,
      "step": 29435
    },
    {
      "epoch": 516.4955752212389,
      "grad_norm": 6.658923012992091e-08,
      "learning_rate": 0.04870876771598966,
      "loss": 0.0,
      "num_input_tokens_seen": 16725176,
      "step": 29440
    },
    {
      "epoch": 516.5840707964602,
      "grad_norm": 2.517070640806196e-08,
      "learning_rate": 0.04866532929449744,
      "loss": 0.0,
      "num_input_tokens_seen": 16727832,
      "step": 29445
    },
    {
      "epoch": 516.6725663716815,
      "grad_norm": 2.5854749452491887e-08,
      "learning_rate": 0.048621906500084945,
      "loss": 0.0,
      "num_input_tokens_seen": 16730760,
      "step": 29450
    },
    {
      "epoch": 516.7610619469026,
      "grad_norm": 2.742712013059645e-08,
      "learning_rate": 0.04857849933944845,
      "loss": 0.0,
      "num_input_tokens_seen": 16733416,
      "step": 29455
    },
    {
      "epoch": 516.8495575221239,
      "grad_norm": 1.4542333559575127e-08,
      "learning_rate": 0.048535107819281866,
      "loss": 0.0,
      "num_input_tokens_seen": 16736632,
      "step": 29460
    },
    {
      "epoch": 516.9380530973451,
      "grad_norm": 3.3786982100991736e-08,
      "learning_rate": 0.04849173194627675,
      "loss": 0.0,
      "num_input_tokens_seen": 16739592,
      "step": 29465
    },
    {
      "epoch": 517.0176991150443,
      "grad_norm": 2.862393344571501e-08,
      "learning_rate": 0.04844837172712223,
      "loss": 0.0,
      "num_input_tokens_seen": 16741720,
      "step": 29470
    },
    {
      "epoch": 517.1061946902655,
      "grad_norm": 4.980795509368363e-08,
      "learning_rate": 0.04840502716850494,
      "loss": 0.0,
      "num_input_tokens_seen": 16744920,
      "step": 29475
    },
    {
      "epoch": 517.1946902654868,
      "grad_norm": 3.2123029569675055e-08,
      "learning_rate": 0.04836169827710916,
      "loss": 0.0,
      "num_input_tokens_seen": 16747704,
      "step": 29480
    },
    {
      "epoch": 517.2831858407079,
      "grad_norm": 4.1118287441577195e-08,
      "learning_rate": 0.04831838505961684,
      "loss": 0.0,
      "num_input_tokens_seen": 16750760,
      "step": 29485
    },
    {
      "epoch": 517.3716814159292,
      "grad_norm": 5.207387943073627e-08,
      "learning_rate": 0.048275087522707295,
      "loss": 0.0,
      "num_input_tokens_seen": 16753672,
      "step": 29490
    },
    {
      "epoch": 517.4601769911504,
      "grad_norm": 5.071173703186105e-08,
      "learning_rate": 0.04823180567305766,
      "loss": 0.0,
      "num_input_tokens_seen": 16756328,
      "step": 29495
    },
    {
      "epoch": 517.5486725663717,
      "grad_norm": 2.4260797815145452e-08,
      "learning_rate": 0.04818853951734244,
      "loss": 0.0,
      "num_input_tokens_seen": 16758728,
      "step": 29500
    },
    {
      "epoch": 517.637168141593,
      "grad_norm": 2.6837739142138162e-08,
      "learning_rate": 0.04814528906223387,
      "loss": 0.0,
      "num_input_tokens_seen": 16761656,
      "step": 29505
    },
    {
      "epoch": 517.7256637168142,
      "grad_norm": 4.9602871143861194e-08,
      "learning_rate": 0.04810205431440177,
      "loss": 0.0,
      "num_input_tokens_seen": 16764280,
      "step": 29510
    },
    {
      "epoch": 517.8141592920354,
      "grad_norm": 2.4037381862740403e-08,
      "learning_rate": 0.04805883528051341,
      "loss": 0.0,
      "num_input_tokens_seen": 16767160,
      "step": 29515
    },
    {
      "epoch": 517.9026548672566,
      "grad_norm": 3.178397278702505e-08,
      "learning_rate": 0.048015631967233685,
      "loss": 0.0,
      "num_input_tokens_seen": 16770168,
      "step": 29520
    },
    {
      "epoch": 517.9911504424779,
      "grad_norm": 2.292821577043469e-08,
      "learning_rate": 0.04797244438122517,
      "loss": 0.0,
      "num_input_tokens_seen": 16773192,
      "step": 29525
    },
    {
      "epoch": 518.070796460177,
      "grad_norm": 2.4429491318755936e-08,
      "learning_rate": 0.04792927252914784,
      "loss": 0.0,
      "num_input_tokens_seen": 16775816,
      "step": 29530
    },
    {
      "epoch": 518.1592920353983,
      "grad_norm": 4.6681964960271216e-08,
      "learning_rate": 0.04788611641765944,
      "loss": 0.0,
      "num_input_tokens_seen": 16778696,
      "step": 29535
    },
    {
      "epoch": 518.2477876106195,
      "grad_norm": 1.8904907861383435e-08,
      "learning_rate": 0.04784297605341508,
      "loss": 0.0,
      "num_input_tokens_seen": 16781720,
      "step": 29540
    },
    {
      "epoch": 518.3362831858407,
      "grad_norm": 4.786094720543588e-08,
      "learning_rate": 0.04779985144306761,
      "loss": 0.0,
      "num_input_tokens_seen": 16785320,
      "step": 29545
    },
    {
      "epoch": 518.4247787610619,
      "grad_norm": 3.780314727919176e-08,
      "learning_rate": 0.047756742593267405,
      "loss": 0.0,
      "num_input_tokens_seen": 16788184,
      "step": 29550
    },
    {
      "epoch": 518.5132743362832,
      "grad_norm": 2.8370763516250008e-08,
      "learning_rate": 0.047713649510662315,
      "loss": 0.0,
      "num_input_tokens_seen": 16790488,
      "step": 29555
    },
    {
      "epoch": 518.6017699115044,
      "grad_norm": 1.6876551711675347e-08,
      "learning_rate": 0.04767057220189789,
      "loss": 0.0,
      "num_input_tokens_seen": 16793080,
      "step": 29560
    },
    {
      "epoch": 518.6902654867257,
      "grad_norm": 9.829182090470567e-08,
      "learning_rate": 0.04762751067361722,
      "loss": 0.0,
      "num_input_tokens_seen": 16796024,
      "step": 29565
    },
    {
      "epoch": 518.7787610619469,
      "grad_norm": 4.740511627687738e-08,
      "learning_rate": 0.04758446493246086,
      "loss": 0.0,
      "num_input_tokens_seen": 16799032,
      "step": 29570
    },
    {
      "epoch": 518.8672566371681,
      "grad_norm": 3.457608599433115e-08,
      "learning_rate": 0.047541434985067084,
      "loss": 0.0,
      "num_input_tokens_seen": 16802152,
      "step": 29575
    },
    {
      "epoch": 518.9557522123894,
      "grad_norm": 4.6436429812501956e-08,
      "learning_rate": 0.047498420838071556,
      "loss": 0.0,
      "num_input_tokens_seen": 16804712,
      "step": 29580
    },
    {
      "epoch": 519.0353982300885,
      "grad_norm": 2.8709404631399593e-08,
      "learning_rate": 0.04745542249810772,
      "loss": 0.0,
      "num_input_tokens_seen": 16806992,
      "step": 29585
    },
    {
      "epoch": 519.1238938053098,
      "grad_norm": 3.0813335882839965e-08,
      "learning_rate": 0.047412439971806324,
      "loss": 0.0,
      "num_input_tokens_seen": 16810432,
      "step": 29590
    },
    {
      "epoch": 519.212389380531,
      "grad_norm": 4.479647586208557e-08,
      "learning_rate": 0.04736947326579592,
      "loss": 0.0,
      "num_input_tokens_seen": 16813504,
      "step": 29595
    },
    {
      "epoch": 519.3008849557522,
      "grad_norm": 5.5190778169844634e-08,
      "learning_rate": 0.04732652238670245,
      "loss": 0.0,
      "num_input_tokens_seen": 16816272,
      "step": 29600
    },
    {
      "epoch": 519.3008849557522,
      "eval_loss": 0.8267478346824646,
      "eval_runtime": 0.9426,
      "eval_samples_per_second": 26.522,
      "eval_steps_per_second": 13.792,
      "num_input_tokens_seen": 16816272,
      "step": 29600
    },
    {
      "epoch": 519.3893805309734,
      "grad_norm": 3.44584840661355e-08,
      "learning_rate": 0.04728358734114952,
      "loss": 0.0,
      "num_input_tokens_seen": 16818864,
      "step": 29605
    },
    {
      "epoch": 519.4778761061947,
      "grad_norm": 5.3015135392797674e-08,
      "learning_rate": 0.04724066813575821,
      "loss": 0.0,
      "num_input_tokens_seen": 16821888,
      "step": 29610
    },
    {
      "epoch": 519.566371681416,
      "grad_norm": 3.833364203842393e-08,
      "learning_rate": 0.04719776477714729,
      "loss": 0.0,
      "num_input_tokens_seen": 16824464,
      "step": 29615
    },
    {
      "epoch": 519.6548672566372,
      "grad_norm": 2.3377213054232016e-08,
      "learning_rate": 0.047154877271932856,
      "loss": 0.0,
      "num_input_tokens_seen": 16827456,
      "step": 29620
    },
    {
      "epoch": 519.7433628318585,
      "grad_norm": 1.2596831133748765e-08,
      "learning_rate": 0.0471120056267288,
      "loss": 0.0,
      "num_input_tokens_seen": 16830080,
      "step": 29625
    },
    {
      "epoch": 519.8318584070796,
      "grad_norm": 2.1195917909722084e-08,
      "learning_rate": 0.047069149848146495,
      "loss": 0.0,
      "num_input_tokens_seen": 16832912,
      "step": 29630
    },
    {
      "epoch": 519.9203539823009,
      "grad_norm": 3.0343471735250205e-08,
      "learning_rate": 0.04702630994279473,
      "loss": 0.0,
      "num_input_tokens_seen": 16835984,
      "step": 29635
    },
    {
      "epoch": 520.0,
      "grad_norm": 1.8847362781571064e-08,
      "learning_rate": 0.046983485917280035,
      "loss": 0.0,
      "num_input_tokens_seen": 16838416,
      "step": 29640
    },
    {
      "epoch": 520.0884955752213,
      "grad_norm": 2.3111866198632924e-08,
      "learning_rate": 0.04694067777820644,
      "loss": 0.0,
      "num_input_tokens_seen": 16841680,
      "step": 29645
    },
    {
      "epoch": 520.1769911504425,
      "grad_norm": 3.839771878233478e-08,
      "learning_rate": 0.046897885532175415,
      "loss": 0.0,
      "num_input_tokens_seen": 16844688,
      "step": 29650
    },
    {
      "epoch": 520.2654867256637,
      "grad_norm": 2.0054155669413376e-08,
      "learning_rate": 0.04685510918578613,
      "loss": 0.0,
      "num_input_tokens_seen": 16847440,
      "step": 29655
    },
    {
      "epoch": 520.3539823008849,
      "grad_norm": 2.5513697821111236e-08,
      "learning_rate": 0.04681234874563519,
      "loss": 0.0,
      "num_input_tokens_seen": 16850000,
      "step": 29660
    },
    {
      "epoch": 520.4424778761062,
      "grad_norm": 6.973971267143497e-08,
      "learning_rate": 0.046769604218316836,
      "loss": 0.0,
      "num_input_tokens_seen": 16852880,
      "step": 29665
    },
    {
      "epoch": 520.5309734513274,
      "grad_norm": 3.7972913702333244e-08,
      "learning_rate": 0.04672687561042279,
      "loss": 0.0,
      "num_input_tokens_seen": 16855152,
      "step": 29670
    },
    {
      "epoch": 520.6194690265487,
      "grad_norm": 2.3902821055798995e-08,
      "learning_rate": 0.046684162928542286,
      "loss": 0.0,
      "num_input_tokens_seen": 16857888,
      "step": 29675
    },
    {
      "epoch": 520.70796460177,
      "grad_norm": 3.1049424364937295e-08,
      "learning_rate": 0.04664146617926222,
      "loss": 0.0,
      "num_input_tokens_seen": 16860784,
      "step": 29680
    },
    {
      "epoch": 520.7964601769911,
      "grad_norm": 5.769940614186453e-08,
      "learning_rate": 0.046598785369167,
      "loss": 0.0,
      "num_input_tokens_seen": 16863808,
      "step": 29685
    },
    {
      "epoch": 520.8849557522124,
      "grad_norm": 2.6933346219948362e-08,
      "learning_rate": 0.046556120504838434,
      "loss": 0.0,
      "num_input_tokens_seen": 16866656,
      "step": 29690
    },
    {
      "epoch": 520.9734513274336,
      "grad_norm": 5.3680867750927064e-08,
      "learning_rate": 0.04651347159285609,
      "loss": 0.0,
      "num_input_tokens_seen": 16869920,
      "step": 29695
    },
    {
      "epoch": 521.0530973451328,
      "grad_norm": 7.420442216243828e-08,
      "learning_rate": 0.04647083863979688,
      "loss": 0.0,
      "num_input_tokens_seen": 16872592,
      "step": 29700
    },
    {
      "epoch": 521.141592920354,
      "grad_norm": 2.2767252971789276e-08,
      "learning_rate": 0.04642822165223538,
      "loss": 0.0,
      "num_input_tokens_seen": 16875248,
      "step": 29705
    },
    {
      "epoch": 521.2300884955753,
      "grad_norm": 2.310086166801284e-08,
      "learning_rate": 0.046385620636743716,
      "loss": 0.0,
      "num_input_tokens_seen": 16877952,
      "step": 29710
    },
    {
      "epoch": 521.3185840707964,
      "grad_norm": 4.741790249340738e-08,
      "learning_rate": 0.04634303559989141,
      "loss": 0.0,
      "num_input_tokens_seen": 16880848,
      "step": 29715
    },
    {
      "epoch": 521.4070796460177,
      "grad_norm": 3.698026063148063e-08,
      "learning_rate": 0.046300466548245635,
      "loss": 0.0,
      "num_input_tokens_seen": 16883664,
      "step": 29720
    },
    {
      "epoch": 521.4955752212389,
      "grad_norm": 1.3696088707604304e-08,
      "learning_rate": 0.04625791348837114,
      "loss": 0.0,
      "num_input_tokens_seen": 16886560,
      "step": 29725
    },
    {
      "epoch": 521.5840707964602,
      "grad_norm": 1.4152764293839937e-08,
      "learning_rate": 0.046215376426830095,
      "loss": 0.0,
      "num_input_tokens_seen": 16889408,
      "step": 29730
    },
    {
      "epoch": 521.6725663716815,
      "grad_norm": 3.6289883098561404e-08,
      "learning_rate": 0.04617285537018219,
      "loss": 0.0,
      "num_input_tokens_seen": 16892192,
      "step": 29735
    },
    {
      "epoch": 521.7610619469026,
      "grad_norm": 3.5760123751060746e-08,
      "learning_rate": 0.046130350324984803,
      "loss": 0.0,
      "num_input_tokens_seen": 16894688,
      "step": 29740
    },
    {
      "epoch": 521.8495575221239,
      "grad_norm": 3.4232353840479846e-08,
      "learning_rate": 0.046087861297792666,
      "loss": 0.0,
      "num_input_tokens_seen": 16897888,
      "step": 29745
    },
    {
      "epoch": 521.9380530973451,
      "grad_norm": 3.545966364981723e-08,
      "learning_rate": 0.0460453882951582,
      "loss": 0.0,
      "num_input_tokens_seen": 16901152,
      "step": 29750
    },
    {
      "epoch": 522.0176991150443,
      "grad_norm": 3.7371506778072217e-08,
      "learning_rate": 0.04600293132363119,
      "loss": 0.0,
      "num_input_tokens_seen": 16903504,
      "step": 29755
    },
    {
      "epoch": 522.1061946902655,
      "grad_norm": 5.1345608653718955e-08,
      "learning_rate": 0.045960490389759086,
      "loss": 0.0,
      "num_input_tokens_seen": 16905968,
      "step": 29760
    },
    {
      "epoch": 522.1946902654868,
      "grad_norm": 1.703772944949833e-08,
      "learning_rate": 0.04591806550008685,
      "loss": 0.0,
      "num_input_tokens_seen": 16908688,
      "step": 29765
    },
    {
      "epoch": 522.2831858407079,
      "grad_norm": 3.310366736286596e-08,
      "learning_rate": 0.045875656661156825,
      "loss": 0.0,
      "num_input_tokens_seen": 16911840,
      "step": 29770
    },
    {
      "epoch": 522.3716814159292,
      "grad_norm": 2.2744281125142152e-08,
      "learning_rate": 0.04583326387950911,
      "loss": 0.0,
      "num_input_tokens_seen": 16915072,
      "step": 29775
    },
    {
      "epoch": 522.4601769911504,
      "grad_norm": 3.007609095107e-08,
      "learning_rate": 0.0457908871616811,
      "loss": 0.0,
      "num_input_tokens_seen": 16917680,
      "step": 29780
    },
    {
      "epoch": 522.5486725663717,
      "grad_norm": 2.8211539770950367e-08,
      "learning_rate": 0.04574852651420786,
      "loss": 0.0,
      "num_input_tokens_seen": 16920240,
      "step": 29785
    },
    {
      "epoch": 522.637168141593,
      "grad_norm": 1.0891797685985694e-08,
      "learning_rate": 0.045706181943621985,
      "loss": 0.0,
      "num_input_tokens_seen": 16922992,
      "step": 29790
    },
    {
      "epoch": 522.7256637168142,
      "grad_norm": 3.354238842234736e-08,
      "learning_rate": 0.04566385345645344,
      "loss": 0.0,
      "num_input_tokens_seen": 16925824,
      "step": 29795
    },
    {
      "epoch": 522.8141592920354,
      "grad_norm": 3.038427820456491e-08,
      "learning_rate": 0.04562154105922993,
      "loss": 0.0,
      "num_input_tokens_seen": 16929072,
      "step": 29800
    },
    {
      "epoch": 522.8141592920354,
      "eval_loss": 0.8346766233444214,
      "eval_runtime": 0.9421,
      "eval_samples_per_second": 26.536,
      "eval_steps_per_second": 13.799,
      "num_input_tokens_seen": 16929072,
      "step": 29800
    },
    {
      "epoch": 522.9026548672566,
      "grad_norm": 5.133327363182616e-08,
      "learning_rate": 0.04557924475847642,
      "loss": 0.0,
      "num_input_tokens_seen": 16932176,
      "step": 29805
    },
    {
      "epoch": 522.9911504424779,
      "grad_norm": 2.6894479532302284e-08,
      "learning_rate": 0.04553696456071567,
      "loss": 0.0,
      "num_input_tokens_seen": 16934736,
      "step": 29810
    },
    {
      "epoch": 523.070796460177,
      "grad_norm": 2.3539461935229156e-08,
      "learning_rate": 0.045494700472467724,
      "loss": 0.0,
      "num_input_tokens_seen": 16937232,
      "step": 29815
    },
    {
      "epoch": 523.1592920353983,
      "grad_norm": 2.3968286910758252e-08,
      "learning_rate": 0.04545245250025024,
      "loss": 0.0,
      "num_input_tokens_seen": 16940544,
      "step": 29820
    },
    {
      "epoch": 523.2477876106195,
      "grad_norm": 3.3039601277096153e-08,
      "learning_rate": 0.045410220650578384,
      "loss": 0.0,
      "num_input_tokens_seen": 16943328,
      "step": 29825
    },
    {
      "epoch": 523.3362831858407,
      "grad_norm": 3.189406072579004e-08,
      "learning_rate": 0.04536800492996492,
      "loss": 0.0,
      "num_input_tokens_seen": 16946032,
      "step": 29830
    },
    {
      "epoch": 523.4247787610619,
      "grad_norm": 4.444512669010692e-08,
      "learning_rate": 0.04532580534491994,
      "loss": 0.0,
      "num_input_tokens_seen": 16949232,
      "step": 29835
    },
    {
      "epoch": 523.5132743362832,
      "grad_norm": 2.9004690205169936e-08,
      "learning_rate": 0.045283621901951183,
      "loss": 0.0,
      "num_input_tokens_seen": 16952080,
      "step": 29840
    },
    {
      "epoch": 523.6017699115044,
      "grad_norm": 2.6571099098759987e-08,
      "learning_rate": 0.04524145460756393,
      "loss": 0.0,
      "num_input_tokens_seen": 16955328,
      "step": 29845
    },
    {
      "epoch": 523.6902654867257,
      "grad_norm": 2.2561501111795224e-08,
      "learning_rate": 0.045199303468260794,
      "loss": 0.0,
      "num_input_tokens_seen": 16957632,
      "step": 29850
    },
    {
      "epoch": 523.7787610619469,
      "grad_norm": 2.216491878925808e-08,
      "learning_rate": 0.04515716849054214,
      "loss": 0.0,
      "num_input_tokens_seen": 16960688,
      "step": 29855
    },
    {
      "epoch": 523.8672566371681,
      "grad_norm": 3.598429998419306e-08,
      "learning_rate": 0.04511504968090558,
      "loss": 0.0,
      "num_input_tokens_seen": 16963424,
      "step": 29860
    },
    {
      "epoch": 523.9557522123894,
      "grad_norm": 3.7788421280993134e-08,
      "learning_rate": 0.04507294704584644,
      "loss": 0.0,
      "num_input_tokens_seen": 16966576,
      "step": 29865
    },
    {
      "epoch": 524.0353982300885,
      "grad_norm": 3.0785965776658486e-08,
      "learning_rate": 0.04503086059185749,
      "loss": 0.0,
      "num_input_tokens_seen": 16968544,
      "step": 29870
    },
    {
      "epoch": 524.1238938053098,
      "grad_norm": 3.671055992526817e-08,
      "learning_rate": 0.04498879032542893,
      "loss": 0.0,
      "num_input_tokens_seen": 16971152,
      "step": 29875
    },
    {
      "epoch": 524.212389380531,
      "grad_norm": 4.923289509406459e-08,
      "learning_rate": 0.0449467362530486,
      "loss": 0.0,
      "num_input_tokens_seen": 16973984,
      "step": 29880
    },
    {
      "epoch": 524.3008849557522,
      "grad_norm": 2.1601596955633795e-08,
      "learning_rate": 0.04490469838120171,
      "loss": 0.0,
      "num_input_tokens_seen": 16977440,
      "step": 29885
    },
    {
      "epoch": 524.3893805309734,
      "grad_norm": 5.584973550298855e-08,
      "learning_rate": 0.04486267671637101,
      "loss": 0.0,
      "num_input_tokens_seen": 16980368,
      "step": 29890
    },
    {
      "epoch": 524.4778761061947,
      "grad_norm": 3.744687759876797e-08,
      "learning_rate": 0.04482067126503683,
      "loss": 0.0,
      "num_input_tokens_seen": 16982768,
      "step": 29895
    },
    {
      "epoch": 524.566371681416,
      "grad_norm": 3.5290170785629016e-08,
      "learning_rate": 0.04477868203367687,
      "loss": 0.0,
      "num_input_tokens_seen": 16985440,
      "step": 29900
    },
    {
      "epoch": 524.6548672566372,
      "grad_norm": 2.7590207452021787e-08,
      "learning_rate": 0.044736709028766426,
      "loss": 0.0,
      "num_input_tokens_seen": 16988608,
      "step": 29905
    },
    {
      "epoch": 524.7433628318585,
      "grad_norm": 2.639009366589562e-08,
      "learning_rate": 0.04469475225677832,
      "loss": 0.0,
      "num_input_tokens_seen": 16991488,
      "step": 29910
    },
    {
      "epoch": 524.8318584070796,
      "grad_norm": 1.6971421601397196e-08,
      "learning_rate": 0.04465281172418273,
      "loss": 0.0,
      "num_input_tokens_seen": 16994688,
      "step": 29915
    },
    {
      "epoch": 524.9203539823009,
      "grad_norm": 1.9176601639969704e-08,
      "learning_rate": 0.044610887437447476,
      "loss": 0.0,
      "num_input_tokens_seen": 16997584,
      "step": 29920
    },
    {
      "epoch": 525.0,
      "grad_norm": 1.1036419067522729e-07,
      "learning_rate": 0.044568979403037744,
      "loss": 0.0,
      "num_input_tokens_seen": 16999896,
      "step": 29925
    },
    {
      "epoch": 525.0884955752213,
      "grad_norm": 1.431159812881333e-08,
      "learning_rate": 0.04452708762741631,
      "loss": 0.0,
      "num_input_tokens_seen": 17002184,
      "step": 29930
    },
    {
      "epoch": 525.1769911504425,
      "grad_norm": 6.116609796436023e-08,
      "learning_rate": 0.044485212117043475,
      "loss": 0.0,
      "num_input_tokens_seen": 17004472,
      "step": 29935
    },
    {
      "epoch": 525.2654867256637,
      "grad_norm": 2.2071755978458896e-08,
      "learning_rate": 0.04444335287837687,
      "loss": 0.0,
      "num_input_tokens_seen": 17007528,
      "step": 29940
    },
    {
      "epoch": 525.3539823008849,
      "grad_norm": 4.1353096946750156e-08,
      "learning_rate": 0.04440150991787179,
      "loss": 0.0,
      "num_input_tokens_seen": 17010824,
      "step": 29945
    },
    {
      "epoch": 525.4424778761062,
      "grad_norm": 3.4516663305339534e-08,
      "learning_rate": 0.04435968324198088,
      "loss": 0.0,
      "num_input_tokens_seen": 17013688,
      "step": 29950
    },
    {
      "epoch": 525.5309734513274,
      "grad_norm": 3.3870481530584584e-08,
      "learning_rate": 0.04431787285715442,
      "loss": 0.0,
      "num_input_tokens_seen": 17016264,
      "step": 29955
    },
    {
      "epoch": 525.6194690265487,
      "grad_norm": 3.81201274990417e-08,
      "learning_rate": 0.04427607876984004,
      "loss": 0.0,
      "num_input_tokens_seen": 17019576,
      "step": 29960
    },
    {
      "epoch": 525.70796460177,
      "grad_norm": 2.1900358859738844e-08,
      "learning_rate": 0.044234300986482886,
      "loss": 0.0,
      "num_input_tokens_seen": 17022472,
      "step": 29965
    },
    {
      "epoch": 525.7964601769911,
      "grad_norm": 2.1081550727330978e-08,
      "learning_rate": 0.04419253951352566,
      "loss": 0.0,
      "num_input_tokens_seen": 17025880,
      "step": 29970
    },
    {
      "epoch": 525.8849557522124,
      "grad_norm": 7.032826943031978e-08,
      "learning_rate": 0.044150794357408533,
      "loss": 0.0,
      "num_input_tokens_seen": 17028856,
      "step": 29975
    },
    {
      "epoch": 525.9734513274336,
      "grad_norm": 1.0127313210261946e-08,
      "learning_rate": 0.044109065524569065,
      "loss": 0.0,
      "num_input_tokens_seen": 17031848,
      "step": 29980
    },
    {
      "epoch": 526.0530973451328,
      "grad_norm": 1.469257604469476e-08,
      "learning_rate": 0.0440673530214424,
      "loss": 0.0,
      "num_input_tokens_seen": 17034336,
      "step": 29985
    },
    {
      "epoch": 526.141592920354,
      "grad_norm": 2.4382469376860172e-08,
      "learning_rate": 0.04402565685446117,
      "loss": 0.0,
      "num_input_tokens_seen": 17036992,
      "step": 29990
    },
    {
      "epoch": 526.2300884955753,
      "grad_norm": 1.2174481867077702e-08,
      "learning_rate": 0.04398397703005536,
      "loss": 0.0,
      "num_input_tokens_seen": 17039888,
      "step": 29995
    },
    {
      "epoch": 526.3185840707964,
      "grad_norm": 3.4689634276219294e-08,
      "learning_rate": 0.043942313554652626,
      "loss": 0.0,
      "num_input_tokens_seen": 17043120,
      "step": 30000
    },
    {
      "epoch": 526.3185840707964,
      "eval_loss": 0.8453069925308228,
      "eval_runtime": 0.9342,
      "eval_samples_per_second": 26.761,
      "eval_steps_per_second": 13.916,
      "num_input_tokens_seen": 17043120,
      "step": 30000
    },
    {
      "epoch": 526.4070796460177,
      "grad_norm": 2.2309293967737176e-08,
      "learning_rate": 0.0439006664346779,
      "loss": 0.0,
      "num_input_tokens_seen": 17046224,
      "step": 30005
    },
    {
      "epoch": 526.4955752212389,
      "grad_norm": 4.186296820307689e-08,
      "learning_rate": 0.043859035676553755,
      "loss": 0.0,
      "num_input_tokens_seen": 17049360,
      "step": 30010
    },
    {
      "epoch": 526.5840707964602,
      "grad_norm": 6.36867127923324e-08,
      "learning_rate": 0.043817421286700194,
      "loss": 0.0,
      "num_input_tokens_seen": 17052352,
      "step": 30015
    },
    {
      "epoch": 526.6725663716815,
      "grad_norm": 2.9082153574222502e-08,
      "learning_rate": 0.043775823271534585,
      "loss": 0.0,
      "num_input_tokens_seen": 17055312,
      "step": 30020
    },
    {
      "epoch": 526.7610619469026,
      "grad_norm": 1.735475940733977e-08,
      "learning_rate": 0.04373424163747197,
      "loss": 0.0,
      "num_input_tokens_seen": 17057760,
      "step": 30025
    },
    {
      "epoch": 526.8495575221239,
      "grad_norm": 1.9995599842559386e-08,
      "learning_rate": 0.04369267639092473,
      "loss": 0.0,
      "num_input_tokens_seen": 17060432,
      "step": 30030
    },
    {
      "epoch": 526.9380530973451,
      "grad_norm": 4.241191575715675e-08,
      "learning_rate": 0.04365112753830268,
      "loss": 0.0,
      "num_input_tokens_seen": 17063104,
      "step": 30035
    },
    {
      "epoch": 527.0176991150443,
      "grad_norm": 2.6602270608577783e-08,
      "learning_rate": 0.04360959508601327,
      "loss": 0.0,
      "num_input_tokens_seen": 17065312,
      "step": 30040
    },
    {
      "epoch": 527.1061946902655,
      "grad_norm": 2.829878376076067e-08,
      "learning_rate": 0.04356807904046123,
      "loss": 0.0,
      "num_input_tokens_seen": 17068096,
      "step": 30045
    },
    {
      "epoch": 527.1946902654868,
      "grad_norm": 5.735458685762751e-08,
      "learning_rate": 0.04352657940804892,
      "loss": 0.0,
      "num_input_tokens_seen": 17070832,
      "step": 30050
    },
    {
      "epoch": 527.2831858407079,
      "grad_norm": 3.031236417427863e-08,
      "learning_rate": 0.04348509619517613,
      "loss": 0.0,
      "num_input_tokens_seen": 17073680,
      "step": 30055
    },
    {
      "epoch": 527.3716814159292,
      "grad_norm": 3.452461783126637e-08,
      "learning_rate": 0.04344362940824002,
      "loss": 0.0,
      "num_input_tokens_seen": 17076512,
      "step": 30060
    },
    {
      "epoch": 527.4601769911504,
      "grad_norm": 5.480072218233545e-08,
      "learning_rate": 0.04340217905363533,
      "loss": 0.0,
      "num_input_tokens_seen": 17079632,
      "step": 30065
    },
    {
      "epoch": 527.5486725663717,
      "grad_norm": 1.3834167589266144e-08,
      "learning_rate": 0.04336074513775425,
      "loss": 0.0,
      "num_input_tokens_seen": 17082176,
      "step": 30070
    },
    {
      "epoch": 527.637168141593,
      "grad_norm": 1.8299987303294074e-08,
      "learning_rate": 0.04331932766698636,
      "loss": 0.0,
      "num_input_tokens_seen": 17085872,
      "step": 30075
    },
    {
      "epoch": 527.7256637168142,
      "grad_norm": 2.9905233844829127e-08,
      "learning_rate": 0.0432779266477188,
      "loss": 0.0,
      "num_input_tokens_seen": 17089216,
      "step": 30080
    },
    {
      "epoch": 527.8141592920354,
      "grad_norm": 2.292353862287655e-08,
      "learning_rate": 0.04323654208633607,
      "loss": 0.0,
      "num_input_tokens_seen": 17091968,
      "step": 30085
    },
    {
      "epoch": 527.9026548672566,
      "grad_norm": 1.1361504625995167e-08,
      "learning_rate": 0.04319517398922024,
      "loss": 0.0,
      "num_input_tokens_seen": 17094224,
      "step": 30090
    },
    {
      "epoch": 527.9911504424779,
      "grad_norm": 3.015043859022626e-08,
      "learning_rate": 0.04315382236275079,
      "loss": 0.0,
      "num_input_tokens_seen": 17097264,
      "step": 30095
    },
    {
      "epoch": 528.070796460177,
      "grad_norm": 1.498019663870309e-08,
      "learning_rate": 0.043112487213304664,
      "loss": 0.0,
      "num_input_tokens_seen": 17099152,
      "step": 30100
    },
    {
      "epoch": 528.1592920353983,
      "grad_norm": 3.257978420379004e-08,
      "learning_rate": 0.04307116854725618,
      "loss": 0.0,
      "num_input_tokens_seen": 17101872,
      "step": 30105
    },
    {
      "epoch": 528.2477876106195,
      "grad_norm": 3.181616747838234e-08,
      "learning_rate": 0.043029866370977325,
      "loss": 0.0,
      "num_input_tokens_seen": 17105024,
      "step": 30110
    },
    {
      "epoch": 528.3362831858407,
      "grad_norm": 3.9364557835597225e-08,
      "learning_rate": 0.04298858069083728,
      "loss": 0.0,
      "num_input_tokens_seen": 17107536,
      "step": 30115
    },
    {
      "epoch": 528.4247787610619,
      "grad_norm": 2.7584645678757624e-08,
      "learning_rate": 0.04294731151320295,
      "loss": 0.0,
      "num_input_tokens_seen": 17111008,
      "step": 30120
    },
    {
      "epoch": 528.5132743362832,
      "grad_norm": 4.6382641727404916e-08,
      "learning_rate": 0.04290605884443841,
      "loss": 0.0,
      "num_input_tokens_seen": 17113520,
      "step": 30125
    },
    {
      "epoch": 528.6017699115044,
      "grad_norm": 1.3384648056558035e-08,
      "learning_rate": 0.04286482269090545,
      "loss": 0.0,
      "num_input_tokens_seen": 17116144,
      "step": 30130
    },
    {
      "epoch": 528.6902654867257,
      "grad_norm": 5.289134463737355e-08,
      "learning_rate": 0.04282360305896323,
      "loss": 0.0,
      "num_input_tokens_seen": 17118912,
      "step": 30135
    },
    {
      "epoch": 528.7787610619469,
      "grad_norm": 3.90367489444543e-08,
      "learning_rate": 0.04278239995496822,
      "loss": 0.0,
      "num_input_tokens_seen": 17121584,
      "step": 30140
    },
    {
      "epoch": 528.8672566371681,
      "grad_norm": 2.5517637780581026e-08,
      "learning_rate": 0.042741213385274514,
      "loss": 0.0,
      "num_input_tokens_seen": 17124800,
      "step": 30145
    },
    {
      "epoch": 528.9557522123894,
      "grad_norm": 5.5791467445942544e-08,
      "learning_rate": 0.04270004335623366,
      "loss": 0.0,
      "num_input_tokens_seen": 17127600,
      "step": 30150
    },
    {
      "epoch": 529.0353982300885,
      "grad_norm": 3.1629291186163755e-08,
      "learning_rate": 0.04265888987419448,
      "loss": 0.0,
      "num_input_tokens_seen": 17130488,
      "step": 30155
    },
    {
      "epoch": 529.1238938053098,
      "grad_norm": 1.7164248689027772e-08,
      "learning_rate": 0.04261775294550346,
      "loss": 0.0,
      "num_input_tokens_seen": 17133512,
      "step": 30160
    },
    {
      "epoch": 529.212389380531,
      "grad_norm": 4.892986282811762e-08,
      "learning_rate": 0.042576632576504354,
      "loss": 0.0,
      "num_input_tokens_seen": 17135688,
      "step": 30165
    },
    {
      "epoch": 529.3008849557522,
      "grad_norm": 4.623865024200313e-08,
      "learning_rate": 0.0425355287735385,
      "loss": 0.0,
      "num_input_tokens_seen": 17138888,
      "step": 30170
    },
    {
      "epoch": 529.3893805309734,
      "grad_norm": 4.397288222435236e-08,
      "learning_rate": 0.0424944415429446,
      "loss": 0.0,
      "num_input_tokens_seen": 17141656,
      "step": 30175
    },
    {
      "epoch": 529.4778761061947,
      "grad_norm": 2.931733256161806e-08,
      "learning_rate": 0.04245337089105877,
      "loss": 0.0,
      "num_input_tokens_seen": 17144344,
      "step": 30180
    },
    {
      "epoch": 529.566371681416,
      "grad_norm": 6.252415118979116e-08,
      "learning_rate": 0.04241231682421467,
      "loss": 0.0,
      "num_input_tokens_seen": 17147272,
      "step": 30185
    },
    {
      "epoch": 529.6548672566372,
      "grad_norm": 3.012245386457835e-08,
      "learning_rate": 0.04237127934874337,
      "loss": 0.0,
      "num_input_tokens_seen": 17150568,
      "step": 30190
    },
    {
      "epoch": 529.7433628318585,
      "grad_norm": 2.9811292989734284e-08,
      "learning_rate": 0.042330258470973305,
      "loss": 0.0,
      "num_input_tokens_seen": 17153800,
      "step": 30195
    },
    {
      "epoch": 529.8318584070796,
      "grad_norm": 3.261952841171478e-08,
      "learning_rate": 0.042289254197230515,
      "loss": 0.0,
      "num_input_tokens_seen": 17156344,
      "step": 30200
    },
    {
      "epoch": 529.8318584070796,
      "eval_loss": 0.8593949675559998,
      "eval_runtime": 0.9405,
      "eval_samples_per_second": 26.581,
      "eval_steps_per_second": 13.822,
      "num_input_tokens_seen": 17156344,
      "step": 30200
    },
    {
      "epoch": 529.9203539823009,
      "grad_norm": 2.3842584795374933e-08,
      "learning_rate": 0.04224826653383823,
      "loss": 0.0,
      "num_input_tokens_seen": 17159240,
      "step": 30205
    },
    {
      "epoch": 530.0,
      "grad_norm": 1.9002403917056654e-07,
      "learning_rate": 0.04220729548711735,
      "loss": 0.0,
      "num_input_tokens_seen": 17161240,
      "step": 30210
    },
    {
      "epoch": 530.0884955752213,
      "grad_norm": 2.3626236966833858e-08,
      "learning_rate": 0.04216634106338616,
      "loss": 0.0,
      "num_input_tokens_seen": 17164104,
      "step": 30215
    },
    {
      "epoch": 530.1769911504425,
      "grad_norm": 2.656368458531233e-08,
      "learning_rate": 0.04212540326896025,
      "loss": 0.0,
      "num_input_tokens_seen": 17167256,
      "step": 30220
    },
    {
      "epoch": 530.2654867256637,
      "grad_norm": 2.9524969136218715e-08,
      "learning_rate": 0.0420844821101528,
      "loss": 0.0,
      "num_input_tokens_seen": 17170360,
      "step": 30225
    },
    {
      "epoch": 530.3539823008849,
      "grad_norm": 2.0288666746637318e-08,
      "learning_rate": 0.04204357759327441,
      "loss": 0.0,
      "num_input_tokens_seen": 17173128,
      "step": 30230
    },
    {
      "epoch": 530.4424778761062,
      "grad_norm": 4.009054777043275e-08,
      "learning_rate": 0.042002689724632954,
      "loss": 0.0,
      "num_input_tokens_seen": 17176024,
      "step": 30235
    },
    {
      "epoch": 530.5309734513274,
      "grad_norm": 5.783320844443551e-08,
      "learning_rate": 0.04196181851053398,
      "loss": 0.0,
      "num_input_tokens_seen": 17178920,
      "step": 30240
    },
    {
      "epoch": 530.6194690265487,
      "grad_norm": 1.742279742700248e-08,
      "learning_rate": 0.041920963957280295,
      "loss": 0.0,
      "num_input_tokens_seen": 17181768,
      "step": 30245
    },
    {
      "epoch": 530.70796460177,
      "grad_norm": 2.8896593562421913e-08,
      "learning_rate": 0.04188012607117212,
      "loss": 0.0,
      "num_input_tokens_seen": 17184536,
      "step": 30250
    },
    {
      "epoch": 530.7964601769911,
      "grad_norm": 2.117886488406384e-08,
      "learning_rate": 0.04183930485850725,
      "loss": 0.0,
      "num_input_tokens_seen": 17187192,
      "step": 30255
    },
    {
      "epoch": 530.8849557522124,
      "grad_norm": 3.528489145310232e-08,
      "learning_rate": 0.04179850032558078,
      "loss": 0.0,
      "num_input_tokens_seen": 17190344,
      "step": 30260
    },
    {
      "epoch": 530.9734513274336,
      "grad_norm": 5.74377274631388e-08,
      "learning_rate": 0.041757712478685295,
      "loss": 0.0,
      "num_input_tokens_seen": 17192936,
      "step": 30265
    },
    {
      "epoch": 531.0530973451328,
      "grad_norm": 4.88250755381614e-08,
      "learning_rate": 0.04171694132411085,
      "loss": 0.0,
      "num_input_tokens_seen": 17195168,
      "step": 30270
    },
    {
      "epoch": 531.141592920354,
      "grad_norm": 2.3258065695586083e-08,
      "learning_rate": 0.04167618686814479,
      "loss": 0.0,
      "num_input_tokens_seen": 17198016,
      "step": 30275
    },
    {
      "epoch": 531.2300884955753,
      "grad_norm": 2.6014577159116925e-08,
      "learning_rate": 0.041635449117072024,
      "loss": 0.0,
      "num_input_tokens_seen": 17200832,
      "step": 30280
    },
    {
      "epoch": 531.3185840707964,
      "grad_norm": 3.183180652399642e-08,
      "learning_rate": 0.04159472807717477,
      "loss": 0.0,
      "num_input_tokens_seen": 17205024,
      "step": 30285
    },
    {
      "epoch": 531.4070796460177,
      "grad_norm": 2.432869017354733e-08,
      "learning_rate": 0.041554023754732744,
      "loss": 0.0,
      "num_input_tokens_seen": 17207392,
      "step": 30290
    },
    {
      "epoch": 531.4955752212389,
      "grad_norm": 3.5500132611332447e-08,
      "learning_rate": 0.04151333615602311,
      "loss": 0.0,
      "num_input_tokens_seen": 17210176,
      "step": 30295
    },
    {
      "epoch": 531.5840707964602,
      "grad_norm": 4.9134960988794774e-08,
      "learning_rate": 0.04147266528732034,
      "loss": 0.0,
      "num_input_tokens_seen": 17212864,
      "step": 30300
    },
    {
      "epoch": 531.6725663716815,
      "grad_norm": 6.704637200982688e-08,
      "learning_rate": 0.0414320111548964,
      "loss": 0.0,
      "num_input_tokens_seen": 17215360,
      "step": 30305
    },
    {
      "epoch": 531.7610619469026,
      "grad_norm": 4.3165769625375106e-08,
      "learning_rate": 0.04139137376502076,
      "loss": 0.0,
      "num_input_tokens_seen": 17218048,
      "step": 30310
    },
    {
      "epoch": 531.8495575221239,
      "grad_norm": 4.6039577483725225e-08,
      "learning_rate": 0.04135075312396014,
      "loss": 0.0,
      "num_input_tokens_seen": 17220688,
      "step": 30315
    },
    {
      "epoch": 531.9380530973451,
      "grad_norm": 3.999831932333109e-08,
      "learning_rate": 0.04131014923797875,
      "loss": 0.0,
      "num_input_tokens_seen": 17224112,
      "step": 30320
    },
    {
      "epoch": 532.0176991150443,
      "grad_norm": 2.9569726223144244e-08,
      "learning_rate": 0.04126956211333819,
      "loss": 0.0,
      "num_input_tokens_seen": 17226920,
      "step": 30325
    },
    {
      "epoch": 532.1061946902655,
      "grad_norm": 1.7049474720920443e-08,
      "learning_rate": 0.041228991756297545,
      "loss": 0.0,
      "num_input_tokens_seen": 17229416,
      "step": 30330
    },
    {
      "epoch": 532.1946902654868,
      "grad_norm": 4.28885016390268e-08,
      "learning_rate": 0.04118843817311332,
      "loss": 0.0,
      "num_input_tokens_seen": 17232344,
      "step": 30335
    },
    {
      "epoch": 532.2831858407079,
      "grad_norm": 3.997123698695759e-08,
      "learning_rate": 0.0411479013700393,
      "loss": 0.0,
      "num_input_tokens_seen": 17235544,
      "step": 30340
    },
    {
      "epoch": 532.3716814159292,
      "grad_norm": 2.3179870467515684e-08,
      "learning_rate": 0.0411073813533268,
      "loss": 0.0,
      "num_input_tokens_seen": 17238456,
      "step": 30345
    },
    {
      "epoch": 532.4601769911504,
      "grad_norm": 2.848982205705397e-08,
      "learning_rate": 0.04106687812922456,
      "loss": 0.0,
      "num_input_tokens_seen": 17241512,
      "step": 30350
    },
    {
      "epoch": 532.5486725663717,
      "grad_norm": 2.9639437570949667e-08,
      "learning_rate": 0.041026391703978635,
      "loss": 0.0,
      "num_input_tokens_seen": 17244456,
      "step": 30355
    },
    {
      "epoch": 532.637168141593,
      "grad_norm": 4.6895252125978004e-08,
      "learning_rate": 0.04098592208383259,
      "loss": 0.0,
      "num_input_tokens_seen": 17246920,
      "step": 30360
    },
    {
      "epoch": 532.7256637168142,
      "grad_norm": 5.388551471696701e-08,
      "learning_rate": 0.040945469275027256,
      "loss": 0.0,
      "num_input_tokens_seen": 17249880,
      "step": 30365
    },
    {
      "epoch": 532.8141592920354,
      "grad_norm": 1.890628986700449e-08,
      "learning_rate": 0.04090503328380104,
      "loss": 0.0,
      "num_input_tokens_seen": 17252920,
      "step": 30370
    },
    {
      "epoch": 532.9026548672566,
      "grad_norm": 3.4279597826980535e-08,
      "learning_rate": 0.04086461411638971,
      "loss": 0.0,
      "num_input_tokens_seen": 17255336,
      "step": 30375
    },
    {
      "epoch": 532.9911504424779,
      "grad_norm": 4.0175688553745204e-08,
      "learning_rate": 0.04082421177902631,
      "loss": 0.0,
      "num_input_tokens_seen": 17258136,
      "step": 30380
    },
    {
      "epoch": 533.070796460177,
      "grad_norm": 2.0590750438032046e-08,
      "learning_rate": 0.04078382627794149,
      "loss": 0.0,
      "num_input_tokens_seen": 17260336,
      "step": 30385
    },
    {
      "epoch": 533.1592920353983,
      "grad_norm": 3.100868894989617e-08,
      "learning_rate": 0.04074345761936316,
      "loss": 0.0,
      "num_input_tokens_seen": 17262688,
      "step": 30390
    },
    {
      "epoch": 533.2477876106195,
      "grad_norm": 3.874368559309005e-08,
      "learning_rate": 0.04070310580951663,
      "loss": 0.0,
      "num_input_tokens_seen": 17266016,
      "step": 30395
    },
    {
      "epoch": 533.3362831858407,
      "grad_norm": 2.2068034510880352e-08,
      "learning_rate": 0.040662770854624726,
      "loss": 0.0,
      "num_input_tokens_seen": 17268656,
      "step": 30400
    },
    {
      "epoch": 533.3362831858407,
      "eval_loss": 0.8479840159416199,
      "eval_runtime": 0.9458,
      "eval_samples_per_second": 26.432,
      "eval_steps_per_second": 13.744,
      "num_input_tokens_seen": 17268656,
      "step": 30400
    },
    {
      "epoch": 533.4247787610619,
      "grad_norm": 3.772976242544246e-08,
      "learning_rate": 0.040622452760907535,
      "loss": 0.0,
      "num_input_tokens_seen": 17271920,
      "step": 30405
    },
    {
      "epoch": 533.5132743362832,
      "grad_norm": 3.317260421908941e-08,
      "learning_rate": 0.04058215153458265,
      "loss": 0.0,
      "num_input_tokens_seen": 17274720,
      "step": 30410
    },
    {
      "epoch": 533.6017699115044,
      "grad_norm": 3.926645320007083e-08,
      "learning_rate": 0.04054186718186507,
      "loss": 0.0,
      "num_input_tokens_seen": 17277680,
      "step": 30415
    },
    {
      "epoch": 533.6902654867257,
      "grad_norm": 1.0326212418476644e-07,
      "learning_rate": 0.04050159970896708,
      "loss": 0.0,
      "num_input_tokens_seen": 17280192,
      "step": 30420
    },
    {
      "epoch": 533.7787610619469,
      "grad_norm": 2.822047129313887e-08,
      "learning_rate": 0.04046134912209843,
      "loss": 0.0,
      "num_input_tokens_seen": 17283296,
      "step": 30425
    },
    {
      "epoch": 533.8672566371681,
      "grad_norm": 5.4026394025186164e-08,
      "learning_rate": 0.040421115427466354,
      "loss": 0.0,
      "num_input_tokens_seen": 17286688,
      "step": 30430
    },
    {
      "epoch": 533.9557522123894,
      "grad_norm": 3.829650907505311e-08,
      "learning_rate": 0.04038089863127529,
      "loss": 0.0,
      "num_input_tokens_seen": 17289952,
      "step": 30435
    },
    {
      "epoch": 534.0353982300885,
      "grad_norm": 2.2439321512024435e-08,
      "learning_rate": 0.04034069873972727,
      "loss": 0.0,
      "num_input_tokens_seen": 17292320,
      "step": 30440
    },
    {
      "epoch": 534.1238938053098,
      "grad_norm": 5.6094929590244647e-08,
      "learning_rate": 0.040300515759021514,
      "loss": 0.0,
      "num_input_tokens_seen": 17294800,
      "step": 30445
    },
    {
      "epoch": 534.212389380531,
      "grad_norm": 3.666077574848714e-08,
      "learning_rate": 0.04026034969535478,
      "loss": 0.0,
      "num_input_tokens_seen": 17297888,
      "step": 30450
    },
    {
      "epoch": 534.3008849557522,
      "grad_norm": 2.8543153618443284e-08,
      "learning_rate": 0.040220200554921266,
      "loss": 0.0,
      "num_input_tokens_seen": 17300704,
      "step": 30455
    },
    {
      "epoch": 534.3893805309734,
      "grad_norm": 3.4917931657219015e-08,
      "learning_rate": 0.0401800683439124,
      "loss": 0.0,
      "num_input_tokens_seen": 17303120,
      "step": 30460
    },
    {
      "epoch": 534.4778761061947,
      "grad_norm": 4.2677367417809364e-08,
      "learning_rate": 0.04013995306851704,
      "loss": 0.0,
      "num_input_tokens_seen": 17305888,
      "step": 30465
    },
    {
      "epoch": 534.566371681416,
      "grad_norm": 2.5832742167608558e-08,
      "learning_rate": 0.040099854734921545,
      "loss": 0.0,
      "num_input_tokens_seen": 17309120,
      "step": 30470
    },
    {
      "epoch": 534.6548672566372,
      "grad_norm": 5.166131700207188e-08,
      "learning_rate": 0.0400597733493095,
      "loss": 0.0,
      "num_input_tokens_seen": 17312464,
      "step": 30475
    },
    {
      "epoch": 534.7433628318585,
      "grad_norm": 4.166736289334949e-08,
      "learning_rate": 0.04001970891786203,
      "loss": 0.0,
      "num_input_tokens_seen": 17315360,
      "step": 30480
    },
    {
      "epoch": 534.8318584070796,
      "grad_norm": 2.3584780350915935e-08,
      "learning_rate": 0.03997966144675752,
      "loss": 0.0,
      "num_input_tokens_seen": 17318048,
      "step": 30485
    },
    {
      "epoch": 534.9203539823009,
      "grad_norm": 4.088485638931161e-08,
      "learning_rate": 0.039939630942171796,
      "loss": 0.0,
      "num_input_tokens_seen": 17320976,
      "step": 30490
    },
    {
      "epoch": 535.0,
      "grad_norm": 1.7726721424082825e-08,
      "learning_rate": 0.03989961741027815,
      "loss": 0.0,
      "num_input_tokens_seen": 17323544,
      "step": 30495
    },
    {
      "epoch": 535.0884955752213,
      "grad_norm": 2.3595703169121407e-08,
      "learning_rate": 0.03985962085724704,
      "loss": 0.0,
      "num_input_tokens_seen": 17326568,
      "step": 30500
    },
    {
      "epoch": 535.1769911504425,
      "grad_norm": 2.625287542912247e-08,
      "learning_rate": 0.03981964128924656,
      "loss": 0.0,
      "num_input_tokens_seen": 17329160,
      "step": 30505
    },
    {
      "epoch": 535.2654867256637,
      "grad_norm": 1.1916505116005283e-08,
      "learning_rate": 0.03977967871244197,
      "loss": 0.0,
      "num_input_tokens_seen": 17331704,
      "step": 30510
    },
    {
      "epoch": 535.3539823008849,
      "grad_norm": 2.4710063328825527e-08,
      "learning_rate": 0.03973973313299602,
      "loss": 0.0,
      "num_input_tokens_seen": 17334616,
      "step": 30515
    },
    {
      "epoch": 535.4424778761062,
      "grad_norm": 1.0128180605306625e-07,
      "learning_rate": 0.0396998045570689,
      "loss": 0.0,
      "num_input_tokens_seen": 17337672,
      "step": 30520
    },
    {
      "epoch": 535.5309734513274,
      "grad_norm": 6.665836593811036e-08,
      "learning_rate": 0.03965989299081798,
      "loss": 0.0,
      "num_input_tokens_seen": 17340328,
      "step": 30525
    },
    {
      "epoch": 535.6194690265487,
      "grad_norm": 1.943642402579826e-08,
      "learning_rate": 0.039619998440398235,
      "loss": 0.0,
      "num_input_tokens_seen": 17343304,
      "step": 30530
    },
    {
      "epoch": 535.70796460177,
      "grad_norm": 2.3033901896951647e-08,
      "learning_rate": 0.03958012091196184,
      "loss": 0.0,
      "num_input_tokens_seen": 17346472,
      "step": 30535
    },
    {
      "epoch": 535.7964601769911,
      "grad_norm": 4.2769926267283154e-08,
      "learning_rate": 0.039540260411658396,
      "loss": 0.0,
      "num_input_tokens_seen": 17349448,
      "step": 30540
    },
    {
      "epoch": 535.8849557522124,
      "grad_norm": 2.2525632914494054e-08,
      "learning_rate": 0.03950041694563496,
      "loss": 0.0,
      "num_input_tokens_seen": 17352248,
      "step": 30545
    },
    {
      "epoch": 535.9734513274336,
      "grad_norm": 3.301075324202429e-08,
      "learning_rate": 0.0394605905200358,
      "loss": 0.0,
      "num_input_tokens_seen": 17355080,
      "step": 30550
    },
    {
      "epoch": 536.0530973451328,
      "grad_norm": 4.365591621535714e-08,
      "learning_rate": 0.03942078114100272,
      "loss": 0.0,
      "num_input_tokens_seen": 17357600,
      "step": 30555
    },
    {
      "epoch": 536.141592920354,
      "grad_norm": 2.629039741464112e-08,
      "learning_rate": 0.03938098881467485,
      "loss": 0.0,
      "num_input_tokens_seen": 17360256,
      "step": 30560
    },
    {
      "epoch": 536.2300884955753,
      "grad_norm": 1.1203712624308082e-08,
      "learning_rate": 0.039341213547188586,
      "loss": 0.0,
      "num_input_tokens_seen": 17363392,
      "step": 30565
    },
    {
      "epoch": 536.3185840707964,
      "grad_norm": 6.203335090049222e-08,
      "learning_rate": 0.03930145534467782,
      "loss": 0.0,
      "num_input_tokens_seen": 17366256,
      "step": 30570
    },
    {
      "epoch": 536.4070796460177,
      "grad_norm": 1.1802469224164724e-08,
      "learning_rate": 0.0392617142132738,
      "loss": 0.0,
      "num_input_tokens_seen": 17369040,
      "step": 30575
    },
    {
      "epoch": 536.4955752212389,
      "grad_norm": 3.533337888939059e-08,
      "learning_rate": 0.03922199015910504,
      "loss": 0.0,
      "num_input_tokens_seen": 17372240,
      "step": 30580
    },
    {
      "epoch": 536.5840707964602,
      "grad_norm": 1.5790133645054993e-08,
      "learning_rate": 0.039182283188297556,
      "loss": 0.0,
      "num_input_tokens_seen": 17375264,
      "step": 30585
    },
    {
      "epoch": 536.6725663716815,
      "grad_norm": 4.344568083070044e-08,
      "learning_rate": 0.039142593306974595,
      "loss": 0.0,
      "num_input_tokens_seen": 17378016,
      "step": 30590
    },
    {
      "epoch": 536.7610619469026,
      "grad_norm": 4.8057675172685776e-08,
      "learning_rate": 0.039102920521256856,
      "loss": 0.0,
      "num_input_tokens_seen": 17380928,
      "step": 30595
    },
    {
      "epoch": 536.8495575221239,
      "grad_norm": 1.8951032743075302e-08,
      "learning_rate": 0.03906326483726243,
      "loss": 0.0,
      "num_input_tokens_seen": 17383696,
      "step": 30600
    },
    {
      "epoch": 536.8495575221239,
      "eval_loss": 0.8396360874176025,
      "eval_runtime": 0.9413,
      "eval_samples_per_second": 26.558,
      "eval_steps_per_second": 13.81,
      "num_input_tokens_seen": 17383696,
      "step": 30600
    },
    {
      "epoch": 536.9380530973451,
      "grad_norm": 3.8259294399267674e-08,
      "learning_rate": 0.039023626261106704,
      "loss": 0.0,
      "num_input_tokens_seen": 17386624,
      "step": 30605
    },
    {
      "epoch": 537.0176991150443,
      "grad_norm": 7.001153790042736e-08,
      "learning_rate": 0.03898400479890237,
      "loss": 0.0,
      "num_input_tokens_seen": 17389112,
      "step": 30610
    },
    {
      "epoch": 537.1061946902655,
      "grad_norm": 1.4266182013500384e-08,
      "learning_rate": 0.038944400456759655,
      "loss": 0.0,
      "num_input_tokens_seen": 17392024,
      "step": 30615
    },
    {
      "epoch": 537.1946902654868,
      "grad_norm": 3.2257016613357337e-08,
      "learning_rate": 0.038904813240785964,
      "loss": 0.0,
      "num_input_tokens_seen": 17394344,
      "step": 30620
    },
    {
      "epoch": 537.2831858407079,
      "grad_norm": 2.9296947090529102e-08,
      "learning_rate": 0.03886524315708621,
      "loss": 0.0,
      "num_input_tokens_seen": 17397416,
      "step": 30625
    },
    {
      "epoch": 537.3716814159292,
      "grad_norm": 5.163908056715627e-08,
      "learning_rate": 0.03882569021176255,
      "loss": 0.0,
      "num_input_tokens_seen": 17400328,
      "step": 30630
    },
    {
      "epoch": 537.4601769911504,
      "grad_norm": 2.241670493674519e-08,
      "learning_rate": 0.038786154410914535,
      "loss": 0.0,
      "num_input_tokens_seen": 17403080,
      "step": 30635
    },
    {
      "epoch": 537.5486725663717,
      "grad_norm": 1.773529589854661e-08,
      "learning_rate": 0.03874663576063917,
      "loss": 0.0,
      "num_input_tokens_seen": 17405448,
      "step": 30640
    },
    {
      "epoch": 537.637168141593,
      "grad_norm": 1.5271179876208407e-08,
      "learning_rate": 0.038707134267030624,
      "loss": 0.0,
      "num_input_tokens_seen": 17408312,
      "step": 30645
    },
    {
      "epoch": 537.7256637168142,
      "grad_norm": 3.644825596893497e-08,
      "learning_rate": 0.038667649936180555,
      "loss": 0.0,
      "num_input_tokens_seen": 17410936,
      "step": 30650
    },
    {
      "epoch": 537.8141592920354,
      "grad_norm": 2.58982524314888e-08,
      "learning_rate": 0.038628182774178,
      "loss": 0.0,
      "num_input_tokens_seen": 17413768,
      "step": 30655
    },
    {
      "epoch": 537.9026548672566,
      "grad_norm": 1.4474243137385656e-08,
      "learning_rate": 0.038588732787109226,
      "loss": 0.0,
      "num_input_tokens_seen": 17416952,
      "step": 30660
    },
    {
      "epoch": 537.9911504424779,
      "grad_norm": 2.943834864765904e-08,
      "learning_rate": 0.03854929998105795,
      "loss": 0.0,
      "num_input_tokens_seen": 17419864,
      "step": 30665
    },
    {
      "epoch": 538.070796460177,
      "grad_norm": 2.1845057318614636e-08,
      "learning_rate": 0.03850988436210518,
      "loss": 0.0,
      "num_input_tokens_seen": 17422224,
      "step": 30670
    },
    {
      "epoch": 538.1592920353983,
      "grad_norm": 3.004060644684614e-08,
      "learning_rate": 0.03847048593632933,
      "loss": 0.0,
      "num_input_tokens_seen": 17424848,
      "step": 30675
    },
    {
      "epoch": 538.2477876106195,
      "grad_norm": 2.9340915475017937e-08,
      "learning_rate": 0.038431104709806096,
      "loss": 0.0,
      "num_input_tokens_seen": 17427600,
      "step": 30680
    },
    {
      "epoch": 538.3362831858407,
      "grad_norm": 2.470131477139148e-08,
      "learning_rate": 0.0383917406886086,
      "loss": 0.0,
      "num_input_tokens_seen": 17429824,
      "step": 30685
    },
    {
      "epoch": 538.4247787610619,
      "grad_norm": 2.6874486636074835e-08,
      "learning_rate": 0.03835239387880722,
      "loss": 0.0,
      "num_input_tokens_seen": 17431984,
      "step": 30690
    },
    {
      "epoch": 538.5132743362832,
      "grad_norm": 1.9907522741391404e-08,
      "learning_rate": 0.03831306428646979,
      "loss": 0.0,
      "num_input_tokens_seen": 17435168,
      "step": 30695
    },
    {
      "epoch": 538.6017699115044,
      "grad_norm": 4.233086770000227e-08,
      "learning_rate": 0.03827375191766135,
      "loss": 0.0,
      "num_input_tokens_seen": 17438368,
      "step": 30700
    },
    {
      "epoch": 538.6902654867257,
      "grad_norm": 3.315192387276511e-08,
      "learning_rate": 0.03823445677844446,
      "loss": 0.0,
      "num_input_tokens_seen": 17441264,
      "step": 30705
    },
    {
      "epoch": 538.7787610619469,
      "grad_norm": 7.578910299343988e-08,
      "learning_rate": 0.03819517887487881,
      "loss": 0.0,
      "num_input_tokens_seen": 17444256,
      "step": 30710
    },
    {
      "epoch": 538.8672566371681,
      "grad_norm": 2.29587566735745e-08,
      "learning_rate": 0.03815591821302161,
      "loss": 0.0,
      "num_input_tokens_seen": 17446960,
      "step": 30715
    },
    {
      "epoch": 538.9557522123894,
      "grad_norm": 5.681452819317201e-08,
      "learning_rate": 0.03811667479892739,
      "loss": 0.0,
      "num_input_tokens_seen": 17450416,
      "step": 30720
    },
    {
      "epoch": 539.0353982300885,
      "grad_norm": 4.069336156931058e-08,
      "learning_rate": 0.03807744863864788,
      "loss": 0.0,
      "num_input_tokens_seen": 17453288,
      "step": 30725
    },
    {
      "epoch": 539.1238938053098,
      "grad_norm": 1.1174649650058655e-08,
      "learning_rate": 0.03803823973823229,
      "loss": 0.0,
      "num_input_tokens_seen": 17456168,
      "step": 30730
    },
    {
      "epoch": 539.212389380531,
      "grad_norm": 7.005844082641488e-08,
      "learning_rate": 0.03799904810372719,
      "loss": 0.0,
      "num_input_tokens_seen": 17459080,
      "step": 30735
    },
    {
      "epoch": 539.3008849557522,
      "grad_norm": 4.437412570723609e-08,
      "learning_rate": 0.03795987374117632,
      "loss": 0.0,
      "num_input_tokens_seen": 17462104,
      "step": 30740
    },
    {
      "epoch": 539.3893805309734,
      "grad_norm": 2.7218057141453755e-08,
      "learning_rate": 0.03792071665662093,
      "loss": 0.0,
      "num_input_tokens_seen": 17464792,
      "step": 30745
    },
    {
      "epoch": 539.4778761061947,
      "grad_norm": 4.5633658629640195e-08,
      "learning_rate": 0.03788157685609952,
      "loss": 0.0,
      "num_input_tokens_seen": 17467960,
      "step": 30750
    },
    {
      "epoch": 539.566371681416,
      "grad_norm": 5.566777261378775e-08,
      "learning_rate": 0.037842454345647876,
      "loss": 0.0,
      "num_input_tokens_seen": 17470616,
      "step": 30755
    },
    {
      "epoch": 539.6548672566372,
      "grad_norm": 1.6966039240173814e-08,
      "learning_rate": 0.03780334913129929,
      "loss": 0.0,
      "num_input_tokens_seen": 17473512,
      "step": 30760
    },
    {
      "epoch": 539.7433628318585,
      "grad_norm": 3.429098782703477e-08,
      "learning_rate": 0.037764261219084175,
      "loss": 0.0,
      "num_input_tokens_seen": 17475992,
      "step": 30765
    },
    {
      "epoch": 539.8318584070796,
      "grad_norm": 3.7960621313004594e-08,
      "learning_rate": 0.037725190615030414,
      "loss": 0.0,
      "num_input_tokens_seen": 17479144,
      "step": 30770
    },
    {
      "epoch": 539.9203539823009,
      "grad_norm": 7.69148513768414e-08,
      "learning_rate": 0.037686137325163224,
      "loss": 0.0,
      "num_input_tokens_seen": 17481544,
      "step": 30775
    },
    {
      "epoch": 540.0,
      "grad_norm": 2.1673645278497133e-07,
      "learning_rate": 0.037647101355505065,
      "loss": 0.0,
      "num_input_tokens_seen": 17483856,
      "step": 30780
    },
    {
      "epoch": 540.0884955752213,
      "grad_norm": 3.602072951025548e-08,
      "learning_rate": 0.03760808271207581,
      "loss": 0.0,
      "num_input_tokens_seen": 17486816,
      "step": 30785
    },
    {
      "epoch": 540.1769911504425,
      "grad_norm": 2.9547752689040863e-08,
      "learning_rate": 0.03756908140089258,
      "loss": 0.0,
      "num_input_tokens_seen": 17489920,
      "step": 30790
    },
    {
      "epoch": 540.2654867256637,
      "grad_norm": 2.885857064427455e-08,
      "learning_rate": 0.03753009742796989,
      "loss": 0.0,
      "num_input_tokens_seen": 17493120,
      "step": 30795
    },
    {
      "epoch": 540.3539823008849,
      "grad_norm": 9.006566870084498e-08,
      "learning_rate": 0.037491130799319615,
      "loss": 0.0,
      "num_input_tokens_seen": 17495648,
      "step": 30800
    },
    {
      "epoch": 540.3539823008849,
      "eval_loss": 0.8238071203231812,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.588,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 17495648,
      "step": 30800
    },
    {
      "epoch": 540.4424778761062,
      "grad_norm": 3.9104893545527375e-08,
      "learning_rate": 0.03745218152095079,
      "loss": 0.0,
      "num_input_tokens_seen": 17498352,
      "step": 30805
    },
    {
      "epoch": 540.5309734513274,
      "grad_norm": 2.3799181292361027e-08,
      "learning_rate": 0.037413249598869935,
      "loss": 0.0,
      "num_input_tokens_seen": 17501120,
      "step": 30810
    },
    {
      "epoch": 540.6194690265487,
      "grad_norm": 1.5952405618691046e-08,
      "learning_rate": 0.037374335039080886,
      "loss": 0.0,
      "num_input_tokens_seen": 17504240,
      "step": 30815
    },
    {
      "epoch": 540.70796460177,
      "grad_norm": 6.995137624699055e-08,
      "learning_rate": 0.037335437847584724,
      "loss": 0.0,
      "num_input_tokens_seen": 17507136,
      "step": 30820
    },
    {
      "epoch": 540.7964601769911,
      "grad_norm": 5.477207665194328e-08,
      "learning_rate": 0.03729655803037983,
      "loss": 0.0,
      "num_input_tokens_seen": 17509952,
      "step": 30825
    },
    {
      "epoch": 540.8849557522124,
      "grad_norm": 2.7077252440221855e-08,
      "learning_rate": 0.03725769559346207,
      "loss": 0.0,
      "num_input_tokens_seen": 17512608,
      "step": 30830
    },
    {
      "epoch": 540.9734513274336,
      "grad_norm": 4.5790756075803074e-08,
      "learning_rate": 0.03721885054282439,
      "loss": 0.0,
      "num_input_tokens_seen": 17515712,
      "step": 30835
    },
    {
      "epoch": 541.0530973451328,
      "grad_norm": 4.63393234895193e-08,
      "learning_rate": 0.03718002288445731,
      "loss": 0.0,
      "num_input_tokens_seen": 17518920,
      "step": 30840
    },
    {
      "epoch": 541.141592920354,
      "grad_norm": 4.0495507391824503e-08,
      "learning_rate": 0.03714121262434844,
      "loss": 0.0,
      "num_input_tokens_seen": 17521688,
      "step": 30845
    },
    {
      "epoch": 541.2300884955753,
      "grad_norm": 1.2810676075503125e-08,
      "learning_rate": 0.037102419768482844,
      "loss": 0.0,
      "num_input_tokens_seen": 17524584,
      "step": 30850
    },
    {
      "epoch": 541.3185840707964,
      "grad_norm": 3.165702011642679e-08,
      "learning_rate": 0.03706364432284293,
      "loss": 0.0,
      "num_input_tokens_seen": 17527096,
      "step": 30855
    },
    {
      "epoch": 541.4070796460177,
      "grad_norm": 1.6560621318717494e-08,
      "learning_rate": 0.03702488629340828,
      "loss": 0.0,
      "num_input_tokens_seen": 17530072,
      "step": 30860
    },
    {
      "epoch": 541.4955752212389,
      "grad_norm": 4.631999317439295e-08,
      "learning_rate": 0.036986145686155915,
      "loss": 0.0,
      "num_input_tokens_seen": 17532920,
      "step": 30865
    },
    {
      "epoch": 541.5840707964602,
      "grad_norm": 1.845766028907292e-08,
      "learning_rate": 0.036947422507060075,
      "loss": 0.0,
      "num_input_tokens_seen": 17535288,
      "step": 30870
    },
    {
      "epoch": 541.6725663716815,
      "grad_norm": 1.466239840652861e-08,
      "learning_rate": 0.0369087167620924,
      "loss": 0.0,
      "num_input_tokens_seen": 17538088,
      "step": 30875
    },
    {
      "epoch": 541.7610619469026,
      "grad_norm": 8.869184142668018e-08,
      "learning_rate": 0.03687002845722183,
      "loss": 0.0,
      "num_input_tokens_seen": 17541112,
      "step": 30880
    },
    {
      "epoch": 541.8495575221239,
      "grad_norm": 2.774811846961711e-08,
      "learning_rate": 0.03683135759841451,
      "loss": 0.0,
      "num_input_tokens_seen": 17543800,
      "step": 30885
    },
    {
      "epoch": 541.9380530973451,
      "grad_norm": 3.294646333529272e-08,
      "learning_rate": 0.03679270419163406,
      "loss": 0.0,
      "num_input_tokens_seen": 17546808,
      "step": 30890
    },
    {
      "epoch": 542.0176991150443,
      "grad_norm": 2.7817140590968847e-08,
      "learning_rate": 0.03675406824284127,
      "loss": 0.0,
      "num_input_tokens_seen": 17549200,
      "step": 30895
    },
    {
      "epoch": 542.1061946902655,
      "grad_norm": 2.4936021247867757e-08,
      "learning_rate": 0.03671544975799425,
      "loss": 0.0,
      "num_input_tokens_seen": 17551856,
      "step": 30900
    },
    {
      "epoch": 542.1946902654868,
      "grad_norm": 4.340290615800768e-08,
      "learning_rate": 0.03667684874304854,
      "loss": 0.0,
      "num_input_tokens_seen": 17554480,
      "step": 30905
    },
    {
      "epoch": 542.2831858407079,
      "grad_norm": 4.062257374926048e-08,
      "learning_rate": 0.03663826520395683,
      "loss": 0.0,
      "num_input_tokens_seen": 17556944,
      "step": 30910
    },
    {
      "epoch": 542.3716814159292,
      "grad_norm": 3.032761597410172e-08,
      "learning_rate": 0.03659969914666922,
      "loss": 0.0,
      "num_input_tokens_seen": 17559408,
      "step": 30915
    },
    {
      "epoch": 542.4601769911504,
      "grad_norm": 2.90691239968055e-08,
      "learning_rate": 0.036561150577133106,
      "loss": 0.0,
      "num_input_tokens_seen": 17562704,
      "step": 30920
    },
    {
      "epoch": 542.5486725663717,
      "grad_norm": 3.3326333692684784e-08,
      "learning_rate": 0.036522619501293103,
      "loss": 0.0,
      "num_input_tokens_seen": 17566128,
      "step": 30925
    },
    {
      "epoch": 542.637168141593,
      "grad_norm": 3.8087360820782123e-08,
      "learning_rate": 0.03648410592509122,
      "loss": 0.0,
      "num_input_tokens_seen": 17569056,
      "step": 30930
    },
    {
      "epoch": 542.7256637168142,
      "grad_norm": 3.946432158841162e-08,
      "learning_rate": 0.03644560985446676,
      "loss": 0.0,
      "num_input_tokens_seen": 17571920,
      "step": 30935
    },
    {
      "epoch": 542.8141592920354,
      "grad_norm": 4.466085457011104e-08,
      "learning_rate": 0.036407131295356256,
      "loss": 0.0,
      "num_input_tokens_seen": 17574656,
      "step": 30940
    },
    {
      "epoch": 542.9026548672566,
      "grad_norm": 2.172793323040878e-08,
      "learning_rate": 0.03636867025369362,
      "loss": 0.0,
      "num_input_tokens_seen": 17577936,
      "step": 30945
    },
    {
      "epoch": 542.9911504424779,
      "grad_norm": 2.2984563585737305e-08,
      "learning_rate": 0.03633022673540999,
      "loss": 0.0,
      "num_input_tokens_seen": 17580992,
      "step": 30950
    },
    {
      "epoch": 543.070796460177,
      "grad_norm": 4.3926217330181316e-08,
      "learning_rate": 0.03629180074643385,
      "loss": 0.0,
      "num_input_tokens_seen": 17583632,
      "step": 30955
    },
    {
      "epoch": 543.1592920353983,
      "grad_norm": 8.652971672518106e-08,
      "learning_rate": 0.03625339229269102,
      "loss": 0.0,
      "num_input_tokens_seen": 17586720,
      "step": 30960
    },
    {
      "epoch": 543.2477876106195,
      "grad_norm": 1.3439332313680552e-08,
      "learning_rate": 0.036215001380104535,
      "loss": 0.0,
      "num_input_tokens_seen": 17589520,
      "step": 30965
    },
    {
      "epoch": 543.3362831858407,
      "grad_norm": 3.953411464863166e-08,
      "learning_rate": 0.03617662801459471,
      "loss": 0.0,
      "num_input_tokens_seen": 17592240,
      "step": 30970
    },
    {
      "epoch": 543.4247787610619,
      "grad_norm": 2.1940850913892973e-08,
      "learning_rate": 0.036138272202079276,
      "loss": 0.0,
      "num_input_tokens_seen": 17595472,
      "step": 30975
    },
    {
      "epoch": 543.5132743362832,
      "grad_norm": 3.5493293637500756e-08,
      "learning_rate": 0.036099933948473106,
      "loss": 0.0,
      "num_input_tokens_seen": 17598656,
      "step": 30980
    },
    {
      "epoch": 543.6017699115044,
      "grad_norm": 5.849438977634236e-08,
      "learning_rate": 0.03606161325968851,
      "loss": 0.0,
      "num_input_tokens_seen": 17601600,
      "step": 30985
    },
    {
      "epoch": 543.6902654867257,
      "grad_norm": 4.781225371175424e-08,
      "learning_rate": 0.03602331014163496,
      "loss": 0.0,
      "num_input_tokens_seen": 17604112,
      "step": 30990
    },
    {
      "epoch": 543.7787610619469,
      "grad_norm": 5.9467808455337945e-08,
      "learning_rate": 0.035985024600219295,
      "loss": 0.0,
      "num_input_tokens_seen": 17606848,
      "step": 30995
    },
    {
      "epoch": 543.8672566371681,
      "grad_norm": 2.6160455135482152e-08,
      "learning_rate": 0.03594675664134569,
      "loss": 0.0,
      "num_input_tokens_seen": 17609616,
      "step": 31000
    },
    {
      "epoch": 543.8672566371681,
      "eval_loss": 0.8566185832023621,
      "eval_runtime": 0.9274,
      "eval_samples_per_second": 26.958,
      "eval_steps_per_second": 14.018,
      "num_input_tokens_seen": 17609616,
      "step": 31000
    },
    {
      "epoch": 543.9557522123894,
      "grad_norm": 1.3883992622254482e-08,
      "learning_rate": 0.03590850627091545,
      "loss": 0.0,
      "num_input_tokens_seen": 17612352,
      "step": 31005
    },
    {
      "epoch": 544.0353982300885,
      "grad_norm": 3.7168774724705145e-08,
      "learning_rate": 0.03587027349482731,
      "loss": 0.0,
      "num_input_tokens_seen": 17614896,
      "step": 31010
    },
    {
      "epoch": 544.1238938053098,
      "grad_norm": 3.182884000807462e-08,
      "learning_rate": 0.035832058318977275,
      "loss": 0.0,
      "num_input_tokens_seen": 17617664,
      "step": 31015
    },
    {
      "epoch": 544.212389380531,
      "grad_norm": 1.7536212482127667e-08,
      "learning_rate": 0.03579386074925853,
      "loss": 0.0,
      "num_input_tokens_seen": 17620752,
      "step": 31020
    },
    {
      "epoch": 544.3008849557522,
      "grad_norm": 9.159756331200697e-08,
      "learning_rate": 0.035755680791561696,
      "loss": 0.0,
      "num_input_tokens_seen": 17623200,
      "step": 31025
    },
    {
      "epoch": 544.3893805309734,
      "grad_norm": 2.5169116568690697e-08,
      "learning_rate": 0.03571751845177454,
      "loss": 0.0,
      "num_input_tokens_seen": 17625744,
      "step": 31030
    },
    {
      "epoch": 544.4778761061947,
      "grad_norm": 4.8361922466710894e-08,
      "learning_rate": 0.03567937373578225,
      "loss": 0.0,
      "num_input_tokens_seen": 17628736,
      "step": 31035
    },
    {
      "epoch": 544.566371681416,
      "grad_norm": 4.049661583849229e-08,
      "learning_rate": 0.03564124664946711,
      "loss": 0.0,
      "num_input_tokens_seen": 17631328,
      "step": 31040
    },
    {
      "epoch": 544.6548672566372,
      "grad_norm": 9.25178067490151e-09,
      "learning_rate": 0.035603137198708924,
      "loss": 0.0,
      "num_input_tokens_seen": 17634928,
      "step": 31045
    },
    {
      "epoch": 544.7433628318585,
      "grad_norm": 2.7220419696050158e-08,
      "learning_rate": 0.035565045389384514,
      "loss": 0.0,
      "num_input_tokens_seen": 17637808,
      "step": 31050
    },
    {
      "epoch": 544.8318584070796,
      "grad_norm": 3.681022775481324e-08,
      "learning_rate": 0.03552697122736823,
      "loss": 0.0,
      "num_input_tokens_seen": 17640384,
      "step": 31055
    },
    {
      "epoch": 544.9203539823009,
      "grad_norm": 5.8117507251154166e-08,
      "learning_rate": 0.03548891471853153,
      "loss": 0.0,
      "num_input_tokens_seen": 17643376,
      "step": 31060
    },
    {
      "epoch": 545.0,
      "grad_norm": 5.037058414814055e-08,
      "learning_rate": 0.03545087586874322,
      "loss": 0.0,
      "num_input_tokens_seen": 17646072,
      "step": 31065
    },
    {
      "epoch": 545.0884955752213,
      "grad_norm": 3.628938571864637e-08,
      "learning_rate": 0.03541285468386935,
      "loss": 0.0,
      "num_input_tokens_seen": 17648888,
      "step": 31070
    },
    {
      "epoch": 545.1769911504425,
      "grad_norm": 3.014700666881254e-08,
      "learning_rate": 0.03537485116977327,
      "loss": 0.0,
      "num_input_tokens_seen": 17651992,
      "step": 31075
    },
    {
      "epoch": 545.2654867256637,
      "grad_norm": 1.078429257006519e-08,
      "learning_rate": 0.03533686533231565,
      "loss": 0.0,
      "num_input_tokens_seen": 17654792,
      "step": 31080
    },
    {
      "epoch": 545.3539823008849,
      "grad_norm": 1.9376816595695345e-08,
      "learning_rate": 0.0352988971773543,
      "loss": 0.0,
      "num_input_tokens_seen": 17657592,
      "step": 31085
    },
    {
      "epoch": 545.4424778761062,
      "grad_norm": 2.490453177017571e-08,
      "learning_rate": 0.03526094671074443,
      "loss": 0.0,
      "num_input_tokens_seen": 17660344,
      "step": 31090
    },
    {
      "epoch": 545.5309734513274,
      "grad_norm": 3.357365940814816e-08,
      "learning_rate": 0.03522301393833852,
      "loss": 0.0,
      "num_input_tokens_seen": 17663032,
      "step": 31095
    },
    {
      "epoch": 545.6194690265487,
      "grad_norm": 3.5713895840672194e-08,
      "learning_rate": 0.035185098865986204,
      "loss": 0.0,
      "num_input_tokens_seen": 17666024,
      "step": 31100
    },
    {
      "epoch": 545.70796460177,
      "grad_norm": 3.3485139994127167e-08,
      "learning_rate": 0.03514720149953453,
      "loss": 0.0,
      "num_input_tokens_seen": 17669080,
      "step": 31105
    },
    {
      "epoch": 545.7964601769911,
      "grad_norm": 5.826250060181337e-08,
      "learning_rate": 0.03510932184482773,
      "loss": 0.0,
      "num_input_tokens_seen": 17672040,
      "step": 31110
    },
    {
      "epoch": 545.8849557522124,
      "grad_norm": 5.163633431948256e-08,
      "learning_rate": 0.03507145990770724,
      "loss": 0.0,
      "num_input_tokens_seen": 17675288,
      "step": 31115
    },
    {
      "epoch": 545.9734513274336,
      "grad_norm": 9.21094045480686e-09,
      "learning_rate": 0.035033615694011984,
      "loss": 0.0,
      "num_input_tokens_seen": 17678200,
      "step": 31120
    },
    {
      "epoch": 546.0530973451328,
      "grad_norm": 1.5970718081348423e-08,
      "learning_rate": 0.03499578920957788,
      "loss": 0.0,
      "num_input_tokens_seen": 17680416,
      "step": 31125
    },
    {
      "epoch": 546.141592920354,
      "grad_norm": 3.033219897474737e-08,
      "learning_rate": 0.034957980460238375,
      "loss": 0.0,
      "num_input_tokens_seen": 17683328,
      "step": 31130
    },
    {
      "epoch": 546.2300884955753,
      "grad_norm": 3.6943855974413964e-08,
      "learning_rate": 0.03492018945182393,
      "loss": 0.0,
      "num_input_tokens_seen": 17686576,
      "step": 31135
    },
    {
      "epoch": 546.3185840707964,
      "grad_norm": 3.6947497505934734e-08,
      "learning_rate": 0.03488241619016247,
      "loss": 0.0,
      "num_input_tokens_seen": 17689184,
      "step": 31140
    },
    {
      "epoch": 546.4070796460177,
      "grad_norm": 4.0759790209676794e-08,
      "learning_rate": 0.03484466068107913,
      "loss": 0.0,
      "num_input_tokens_seen": 17691776,
      "step": 31145
    },
    {
      "epoch": 546.4955752212389,
      "grad_norm": 2.872407911524988e-08,
      "learning_rate": 0.034806922930396195,
      "loss": 0.0,
      "num_input_tokens_seen": 17694752,
      "step": 31150
    },
    {
      "epoch": 546.5840707964602,
      "grad_norm": 2.5599362629691313e-08,
      "learning_rate": 0.03476920294393337,
      "loss": 0.0,
      "num_input_tokens_seen": 17697488,
      "step": 31155
    },
    {
      "epoch": 546.6725663716815,
      "grad_norm": 2.376644481216772e-08,
      "learning_rate": 0.03473150072750755,
      "loss": 0.0,
      "num_input_tokens_seen": 17700560,
      "step": 31160
    },
    {
      "epoch": 546.7610619469026,
      "grad_norm": 1.8179228788994806e-08,
      "learning_rate": 0.03469381628693284,
      "loss": 0.0,
      "num_input_tokens_seen": 17703584,
      "step": 31165
    },
    {
      "epoch": 546.8495575221239,
      "grad_norm": 2.8041927890853913e-08,
      "learning_rate": 0.03465614962802072,
      "loss": 0.0,
      "num_input_tokens_seen": 17706320,
      "step": 31170
    },
    {
      "epoch": 546.9380530973451,
      "grad_norm": 9.38844024744867e-09,
      "learning_rate": 0.0346185007565798,
      "loss": 0.0,
      "num_input_tokens_seen": 17709136,
      "step": 31175
    },
    {
      "epoch": 547.0176991150443,
      "grad_norm": 5.596477237190811e-08,
      "learning_rate": 0.03458086967841609,
      "loss": 0.0,
      "num_input_tokens_seen": 17712368,
      "step": 31180
    },
    {
      "epoch": 547.1061946902655,
      "grad_norm": 1.981118202820653e-08,
      "learning_rate": 0.03454325639933266,
      "loss": 0.0,
      "num_input_tokens_seen": 17715392,
      "step": 31185
    },
    {
      "epoch": 547.1946902654868,
      "grad_norm": 3.531403081069584e-08,
      "learning_rate": 0.03450566092513007,
      "loss": 0.0,
      "num_input_tokens_seen": 17718640,
      "step": 31190
    },
    {
      "epoch": 547.2831858407079,
      "grad_norm": 1.440098795768563e-08,
      "learning_rate": 0.034468083261605914,
      "loss": 0.0,
      "num_input_tokens_seen": 17721200,
      "step": 31195
    },
    {
      "epoch": 547.3716814159292,
      "grad_norm": 5.371366285089607e-08,
      "learning_rate": 0.03443052341455522,
      "loss": 0.0,
      "num_input_tokens_seen": 17723600,
      "step": 31200
    },
    {
      "epoch": 547.3716814159292,
      "eval_loss": 0.8399250507354736,
      "eval_runtime": 0.9391,
      "eval_samples_per_second": 26.622,
      "eval_steps_per_second": 13.844,
      "num_input_tokens_seen": 17723600,
      "step": 31200
    },
    {
      "epoch": 547.4601769911504,
      "grad_norm": 4.252002483440265e-08,
      "learning_rate": 0.0343929813897701,
      "loss": 0.0,
      "num_input_tokens_seen": 17726800,
      "step": 31205
    },
    {
      "epoch": 547.5486725663717,
      "grad_norm": 2.8403734475546116e-08,
      "learning_rate": 0.034355457193040125,
      "loss": 0.0,
      "num_input_tokens_seen": 17729728,
      "step": 31210
    },
    {
      "epoch": 547.637168141593,
      "grad_norm": 5.0726054467986614e-08,
      "learning_rate": 0.03431795083015186,
      "loss": 0.0,
      "num_input_tokens_seen": 17732624,
      "step": 31215
    },
    {
      "epoch": 547.7256637168142,
      "grad_norm": 1.86413533498353e-08,
      "learning_rate": 0.03428046230688936,
      "loss": 0.0,
      "num_input_tokens_seen": 17735488,
      "step": 31220
    },
    {
      "epoch": 547.8141592920354,
      "grad_norm": 5.762649024632083e-08,
      "learning_rate": 0.034242991629033805,
      "loss": 0.0,
      "num_input_tokens_seen": 17737824,
      "step": 31225
    },
    {
      "epoch": 547.9026548672566,
      "grad_norm": 3.433035189459588e-08,
      "learning_rate": 0.03420553880236362,
      "loss": 0.0,
      "num_input_tokens_seen": 17740432,
      "step": 31230
    },
    {
      "epoch": 547.9911504424779,
      "grad_norm": 3.1796215438362196e-08,
      "learning_rate": 0.03416810383265449,
      "loss": 0.0,
      "num_input_tokens_seen": 17743584,
      "step": 31235
    },
    {
      "epoch": 548.070796460177,
      "grad_norm": 2.5893761801398796e-08,
      "learning_rate": 0.03413068672567944,
      "loss": 0.0,
      "num_input_tokens_seen": 17746264,
      "step": 31240
    },
    {
      "epoch": 548.1592920353983,
      "grad_norm": 3.165169459862227e-08,
      "learning_rate": 0.034093287487208565,
      "loss": 0.0,
      "num_input_tokens_seen": 17748776,
      "step": 31245
    },
    {
      "epoch": 548.2477876106195,
      "grad_norm": 1.7191043255593286e-08,
      "learning_rate": 0.03405590612300937,
      "loss": 0.0,
      "num_input_tokens_seen": 17751256,
      "step": 31250
    },
    {
      "epoch": 548.3362831858407,
      "grad_norm": 3.7597761348706626e-08,
      "learning_rate": 0.03401854263884646,
      "loss": 0.0,
      "num_input_tokens_seen": 17754136,
      "step": 31255
    },
    {
      "epoch": 548.4247787610619,
      "grad_norm": 2.8538433838320998e-08,
      "learning_rate": 0.033981197040481824,
      "loss": 0.0,
      "num_input_tokens_seen": 17757288,
      "step": 31260
    },
    {
      "epoch": 548.5132743362832,
      "grad_norm": 3.107983204131415e-08,
      "learning_rate": 0.03394386933367459,
      "loss": 0.0,
      "num_input_tokens_seen": 17760312,
      "step": 31265
    },
    {
      "epoch": 548.6017699115044,
      "grad_norm": 4.832480016148111e-08,
      "learning_rate": 0.033906559524181104,
      "loss": 0.0,
      "num_input_tokens_seen": 17762920,
      "step": 31270
    },
    {
      "epoch": 548.6902654867257,
      "grad_norm": 2.2438721103412718e-08,
      "learning_rate": 0.033869267617755085,
      "loss": 0.0,
      "num_input_tokens_seen": 17765608,
      "step": 31275
    },
    {
      "epoch": 548.7787610619469,
      "grad_norm": 3.2907532698800424e-08,
      "learning_rate": 0.0338319936201474,
      "loss": 0.0,
      "num_input_tokens_seen": 17768024,
      "step": 31280
    },
    {
      "epoch": 548.8672566371681,
      "grad_norm": 8.947360186084552e-09,
      "learning_rate": 0.033794737537106136,
      "loss": 0.0,
      "num_input_tokens_seen": 17771368,
      "step": 31285
    },
    {
      "epoch": 548.9557522123894,
      "grad_norm": 2.2511027708560505e-08,
      "learning_rate": 0.03375749937437671,
      "loss": 0.0,
      "num_input_tokens_seen": 17774648,
      "step": 31290
    },
    {
      "epoch": 549.0353982300885,
      "grad_norm": 2.7076419328864176e-08,
      "learning_rate": 0.033720279137701634,
      "loss": 0.0,
      "num_input_tokens_seen": 17776816,
      "step": 31295
    },
    {
      "epoch": 549.1238938053098,
      "grad_norm": 2.6090201998840712e-08,
      "learning_rate": 0.03368307683282078,
      "loss": 0.0,
      "num_input_tokens_seen": 17779472,
      "step": 31300
    },
    {
      "epoch": 549.212389380531,
      "grad_norm": 1.8019884251430085e-08,
      "learning_rate": 0.033645892465471235,
      "loss": 0.0,
      "num_input_tokens_seen": 17782336,
      "step": 31305
    },
    {
      "epoch": 549.3008849557522,
      "grad_norm": 1.9626821057272537e-08,
      "learning_rate": 0.03360872604138724,
      "loss": 0.0,
      "num_input_tokens_seen": 17784976,
      "step": 31310
    },
    {
      "epoch": 549.3893805309734,
      "grad_norm": 4.0561754843793096e-08,
      "learning_rate": 0.03357157756630034,
      "loss": 0.0,
      "num_input_tokens_seen": 17787536,
      "step": 31315
    },
    {
      "epoch": 549.4778761061947,
      "grad_norm": 1.499696899998071e-08,
      "learning_rate": 0.033534447045939365,
      "loss": 0.0,
      "num_input_tokens_seen": 17790512,
      "step": 31320
    },
    {
      "epoch": 549.566371681416,
      "grad_norm": 1.836124852161447e-08,
      "learning_rate": 0.03349733448603026,
      "loss": 0.0,
      "num_input_tokens_seen": 17793200,
      "step": 31325
    },
    {
      "epoch": 549.6548672566372,
      "grad_norm": 4.2785885057128326e-08,
      "learning_rate": 0.03346023989229619,
      "loss": 0.0,
      "num_input_tokens_seen": 17796080,
      "step": 31330
    },
    {
      "epoch": 549.7433628318585,
      "grad_norm": 4.649458773542392e-08,
      "learning_rate": 0.03342316327045769,
      "loss": 0.0,
      "num_input_tokens_seen": 17798960,
      "step": 31335
    },
    {
      "epoch": 549.8318584070796,
      "grad_norm": 3.0656714500310045e-08,
      "learning_rate": 0.033386104626232385,
      "loss": 0.0,
      "num_input_tokens_seen": 17801680,
      "step": 31340
    },
    {
      "epoch": 549.9203539823009,
      "grad_norm": 3.9996077560999765e-08,
      "learning_rate": 0.03334906396533525,
      "loss": 0.0,
      "num_input_tokens_seen": 17805264,
      "step": 31345
    },
    {
      "epoch": 550.0,
      "grad_norm": 2.4193161252128448e-08,
      "learning_rate": 0.033312041293478326,
      "loss": 0.0,
      "num_input_tokens_seen": 17807696,
      "step": 31350
    },
    {
      "epoch": 550.0884955752213,
      "grad_norm": 5.8012279424701774e-08,
      "learning_rate": 0.03327503661637103,
      "loss": 0.0,
      "num_input_tokens_seen": 17810464,
      "step": 31355
    },
    {
      "epoch": 550.1769911504425,
      "grad_norm": 2.2500366014810425e-08,
      "learning_rate": 0.03323804993971998,
      "loss": 0.0,
      "num_input_tokens_seen": 17813504,
      "step": 31360
    },
    {
      "epoch": 550.2654867256637,
      "grad_norm": 1.3214218164137037e-08,
      "learning_rate": 0.033201081269228924,
      "loss": 0.0,
      "num_input_tokens_seen": 17816384,
      "step": 31365
    },
    {
      "epoch": 550.3539823008849,
      "grad_norm": 5.556846360832424e-08,
      "learning_rate": 0.03316413061059895,
      "loss": 0.0,
      "num_input_tokens_seen": 17819248,
      "step": 31370
    },
    {
      "epoch": 550.4424778761062,
      "grad_norm": 3.774916024212871e-08,
      "learning_rate": 0.03312719796952827,
      "loss": 0.0,
      "num_input_tokens_seen": 17822096,
      "step": 31375
    },
    {
      "epoch": 550.5309734513274,
      "grad_norm": 5.7359073935003835e-08,
      "learning_rate": 0.03309028335171236,
      "loss": 0.0,
      "num_input_tokens_seen": 17825216,
      "step": 31380
    },
    {
      "epoch": 550.6194690265487,
      "grad_norm": 2.7910477484738294e-08,
      "learning_rate": 0.03305338676284398,
      "loss": 0.0,
      "num_input_tokens_seen": 17827856,
      "step": 31385
    },
    {
      "epoch": 550.70796460177,
      "grad_norm": 7.590455908257354e-08,
      "learning_rate": 0.03301650820861296,
      "loss": 0.0,
      "num_input_tokens_seen": 17830752,
      "step": 31390
    },
    {
      "epoch": 550.7964601769911,
      "grad_norm": 2.4801002140861783e-08,
      "learning_rate": 0.03297964769470652,
      "loss": 0.0,
      "num_input_tokens_seen": 17833584,
      "step": 31395
    },
    {
      "epoch": 550.8849557522124,
      "grad_norm": 4.9061831930430344e-08,
      "learning_rate": 0.032942805226808945,
      "loss": 0.0,
      "num_input_tokens_seen": 17836576,
      "step": 31400
    },
    {
      "epoch": 550.8849557522124,
      "eval_loss": 0.8476734757423401,
      "eval_runtime": 0.9389,
      "eval_samples_per_second": 26.627,
      "eval_steps_per_second": 13.846,
      "num_input_tokens_seen": 17836576,
      "step": 31400
    },
    {
      "epoch": 550.9734513274336,
      "grad_norm": 2.5127473435304637e-08,
      "learning_rate": 0.03290598081060187,
      "loss": 0.0,
      "num_input_tokens_seen": 17839360,
      "step": 31405
    },
    {
      "epoch": 551.0530973451328,
      "grad_norm": 7.082100950128734e-08,
      "learning_rate": 0.03286917445176407,
      "loss": 0.0,
      "num_input_tokens_seen": 17841592,
      "step": 31410
    },
    {
      "epoch": 551.141592920354,
      "grad_norm": 2.289892364615298e-08,
      "learning_rate": 0.032832386155971456,
      "loss": 0.0,
      "num_input_tokens_seen": 17844552,
      "step": 31415
    },
    {
      "epoch": 551.2300884955753,
      "grad_norm": 1.6438567840282303e-08,
      "learning_rate": 0.032795615928897334,
      "loss": 0.0,
      "num_input_tokens_seen": 17847512,
      "step": 31420
    },
    {
      "epoch": 551.3185840707964,
      "grad_norm": 1.1411323441734567e-07,
      "learning_rate": 0.03275886377621215,
      "loss": 0.0,
      "num_input_tokens_seen": 17850616,
      "step": 31425
    },
    {
      "epoch": 551.4070796460177,
      "grad_norm": 3.739004128533452e-08,
      "learning_rate": 0.03272212970358348,
      "loss": 0.0,
      "num_input_tokens_seen": 17853432,
      "step": 31430
    },
    {
      "epoch": 551.4955752212389,
      "grad_norm": 3.041462193209554e-08,
      "learning_rate": 0.032685413716676215,
      "loss": 0.0,
      "num_input_tokens_seen": 17856856,
      "step": 31435
    },
    {
      "epoch": 551.5840707964602,
      "grad_norm": 3.998993136633544e-08,
      "learning_rate": 0.032648715821152474,
      "loss": 0.0,
      "num_input_tokens_seen": 17860120,
      "step": 31440
    },
    {
      "epoch": 551.6725663716815,
      "grad_norm": 1.9666817507868473e-08,
      "learning_rate": 0.03261203602267143,
      "loss": 0.0,
      "num_input_tokens_seen": 17862840,
      "step": 31445
    },
    {
      "epoch": 551.7610619469026,
      "grad_norm": 1.1997927096274452e-08,
      "learning_rate": 0.03257537432688966,
      "loss": 0.0,
      "num_input_tokens_seen": 17865560,
      "step": 31450
    },
    {
      "epoch": 551.8495575221239,
      "grad_norm": 6.111061878755208e-08,
      "learning_rate": 0.03253873073946077,
      "loss": 0.0,
      "num_input_tokens_seen": 17868056,
      "step": 31455
    },
    {
      "epoch": 551.9380530973451,
      "grad_norm": 3.2120215820441445e-08,
      "learning_rate": 0.03250210526603572,
      "loss": 0.0,
      "num_input_tokens_seen": 17870984,
      "step": 31460
    },
    {
      "epoch": 552.0176991150443,
      "grad_norm": 3.304077367261016e-08,
      "learning_rate": 0.03246549791226266,
      "loss": 0.0,
      "num_input_tokens_seen": 17873320,
      "step": 31465
    },
    {
      "epoch": 552.1061946902655,
      "grad_norm": 3.412790405832311e-08,
      "learning_rate": 0.03242890868378679,
      "loss": 0.0,
      "num_input_tokens_seen": 17876056,
      "step": 31470
    },
    {
      "epoch": 552.1946902654868,
      "grad_norm": 3.385157398838601e-08,
      "learning_rate": 0.03239233758625074,
      "loss": 0.0,
      "num_input_tokens_seen": 17879000,
      "step": 31475
    },
    {
      "epoch": 552.2831858407079,
      "grad_norm": 6.846718036968014e-08,
      "learning_rate": 0.032355784625294204,
      "loss": 0.0,
      "num_input_tokens_seen": 17881992,
      "step": 31480
    },
    {
      "epoch": 552.3716814159292,
      "grad_norm": 8.533422146683733e-08,
      "learning_rate": 0.03231924980655402,
      "loss": 0.0,
      "num_input_tokens_seen": 17884488,
      "step": 31485
    },
    {
      "epoch": 552.4601769911504,
      "grad_norm": 4.2557292800893265e-08,
      "learning_rate": 0.032282733135664446,
      "loss": 0.0,
      "num_input_tokens_seen": 17887144,
      "step": 31490
    },
    {
      "epoch": 552.5486725663717,
      "grad_norm": 2.732546633410493e-08,
      "learning_rate": 0.03224623461825669,
      "loss": 0.0,
      "num_input_tokens_seen": 17889704,
      "step": 31495
    },
    {
      "epoch": 552.637168141593,
      "grad_norm": 3.731884135049768e-08,
      "learning_rate": 0.03220975425995937,
      "loss": 0.0,
      "num_input_tokens_seen": 17892952,
      "step": 31500
    },
    {
      "epoch": 552.7256637168142,
      "grad_norm": 1.95855847096027e-08,
      "learning_rate": 0.032173292066398206,
      "loss": 0.0,
      "num_input_tokens_seen": 17895880,
      "step": 31505
    },
    {
      "epoch": 552.8141592920354,
      "grad_norm": 3.9044206090466105e-08,
      "learning_rate": 0.03213684804319606,
      "loss": 0.0,
      "num_input_tokens_seen": 17898680,
      "step": 31510
    },
    {
      "epoch": 552.9026548672566,
      "grad_norm": 1.7305790578348024e-08,
      "learning_rate": 0.03210042219597312,
      "loss": 0.0,
      "num_input_tokens_seen": 17902360,
      "step": 31515
    },
    {
      "epoch": 552.9911504424779,
      "grad_norm": 2.646485164348178e-08,
      "learning_rate": 0.03206401453034675,
      "loss": 0.0,
      "num_input_tokens_seen": 17904840,
      "step": 31520
    },
    {
      "epoch": 553.070796460177,
      "grad_norm": 2.8611649938170558e-08,
      "learning_rate": 0.03202762505193136,
      "loss": 0.0,
      "num_input_tokens_seen": 17907416,
      "step": 31525
    },
    {
      "epoch": 553.1592920353983,
      "grad_norm": 2.5846130569107117e-08,
      "learning_rate": 0.031991253766338754,
      "loss": 0.0,
      "num_input_tokens_seen": 17910024,
      "step": 31530
    },
    {
      "epoch": 553.2477876106195,
      "grad_norm": 2.6244679318665476e-08,
      "learning_rate": 0.03195490067917778,
      "loss": 0.0,
      "num_input_tokens_seen": 17913384,
      "step": 31535
    },
    {
      "epoch": 553.3362831858407,
      "grad_norm": 2.0899900476933908e-08,
      "learning_rate": 0.03191856579605461,
      "loss": 0.0,
      "num_input_tokens_seen": 17916520,
      "step": 31540
    },
    {
      "epoch": 553.4247787610619,
      "grad_norm": 5.597342322971599e-08,
      "learning_rate": 0.031882249122572454,
      "loss": 0.0,
      "num_input_tokens_seen": 17919768,
      "step": 31545
    },
    {
      "epoch": 553.5132743362832,
      "grad_norm": 5.620238141545997e-08,
      "learning_rate": 0.03184595066433188,
      "loss": 0.0,
      "num_input_tokens_seen": 17922728,
      "step": 31550
    },
    {
      "epoch": 553.6017699115044,
      "grad_norm": 2.35721504537878e-08,
      "learning_rate": 0.03180967042693049,
      "loss": 0.0,
      "num_input_tokens_seen": 17925240,
      "step": 31555
    },
    {
      "epoch": 553.6902654867257,
      "grad_norm": 3.47364057518007e-08,
      "learning_rate": 0.03177340841596323,
      "loss": 0.0,
      "num_input_tokens_seen": 17928088,
      "step": 31560
    },
    {
      "epoch": 553.7787610619469,
      "grad_norm": 2.63867789840333e-08,
      "learning_rate": 0.03173716463702209,
      "loss": 0.0,
      "num_input_tokens_seen": 17930728,
      "step": 31565
    },
    {
      "epoch": 553.8672566371681,
      "grad_norm": 4.45282850591866e-08,
      "learning_rate": 0.03170093909569638,
      "loss": 0.0,
      "num_input_tokens_seen": 17933432,
      "step": 31570
    },
    {
      "epoch": 553.9557522123894,
      "grad_norm": 1.686256112520823e-08,
      "learning_rate": 0.03166473179757246,
      "loss": 0.0,
      "num_input_tokens_seen": 17936280,
      "step": 31575
    },
    {
      "epoch": 554.0353982300885,
      "grad_norm": 2.6738042890883662e-08,
      "learning_rate": 0.031628542748234005,
      "loss": 0.0,
      "num_input_tokens_seen": 17938712,
      "step": 31580
    },
    {
      "epoch": 554.1238938053098,
      "grad_norm": 5.158285887318925e-08,
      "learning_rate": 0.03159237195326184,
      "loss": 0.0,
      "num_input_tokens_seen": 17941640,
      "step": 31585
    },
    {
      "epoch": 554.212389380531,
      "grad_norm": 2.9914776433770385e-08,
      "learning_rate": 0.031556219418233875,
      "loss": 0.0,
      "num_input_tokens_seen": 17944392,
      "step": 31590
    },
    {
      "epoch": 554.3008849557522,
      "grad_norm": 2.835413859259006e-08,
      "learning_rate": 0.03152008514872533,
      "loss": 0.0,
      "num_input_tokens_seen": 17947416,
      "step": 31595
    },
    {
      "epoch": 554.3893805309734,
      "grad_norm": 2.5087224742037506e-08,
      "learning_rate": 0.03148396915030862,
      "loss": 0.0,
      "num_input_tokens_seen": 17949928,
      "step": 31600
    },
    {
      "epoch": 554.3893805309734,
      "eval_loss": 0.8595802783966064,
      "eval_runtime": 0.9324,
      "eval_samples_per_second": 26.812,
      "eval_steps_per_second": 13.942,
      "num_input_tokens_seen": 17949928,
      "step": 31600
    },
    {
      "epoch": 554.4778761061947,
      "grad_norm": 6.19829449988174e-08,
      "learning_rate": 0.03144787142855318,
      "loss": 0.0,
      "num_input_tokens_seen": 17952648,
      "step": 31605
    },
    {
      "epoch": 554.566371681416,
      "grad_norm": 6.947912112309496e-08,
      "learning_rate": 0.031411791989025835,
      "loss": 0.0,
      "num_input_tokens_seen": 17955656,
      "step": 31610
    },
    {
      "epoch": 554.6548672566372,
      "grad_norm": 6.083919856791908e-08,
      "learning_rate": 0.031375730837290394,
      "loss": 0.0,
      "num_input_tokens_seen": 17958504,
      "step": 31615
    },
    {
      "epoch": 554.7433628318585,
      "grad_norm": 5.6194210173998727e-08,
      "learning_rate": 0.031339687978908015,
      "loss": 0.0,
      "num_input_tokens_seen": 17961272,
      "step": 31620
    },
    {
      "epoch": 554.8318584070796,
      "grad_norm": 3.8240870026129414e-08,
      "learning_rate": 0.03130366341943694,
      "loss": 0.0,
      "num_input_tokens_seen": 17964104,
      "step": 31625
    },
    {
      "epoch": 554.9203539823009,
      "grad_norm": 3.0447075971551385e-08,
      "learning_rate": 0.031267657164432555,
      "loss": 0.0,
      "num_input_tokens_seen": 17967416,
      "step": 31630
    },
    {
      "epoch": 555.0,
      "grad_norm": 2.56572736390126e-08,
      "learning_rate": 0.03123166921944752,
      "loss": 0.0,
      "num_input_tokens_seen": 17969928,
      "step": 31635
    },
    {
      "epoch": 555.0884955752213,
      "grad_norm": 1.8560474046580566e-08,
      "learning_rate": 0.031195699590031666,
      "loss": 0.0,
      "num_input_tokens_seen": 17973176,
      "step": 31640
    },
    {
      "epoch": 555.1769911504425,
      "grad_norm": 4.826016564152269e-08,
      "learning_rate": 0.031159748281731885,
      "loss": 0.0,
      "num_input_tokens_seen": 17976056,
      "step": 31645
    },
    {
      "epoch": 555.2654867256637,
      "grad_norm": 1.0812028605755586e-08,
      "learning_rate": 0.031123815300092394,
      "loss": 0.0,
      "num_input_tokens_seen": 17979336,
      "step": 31650
    },
    {
      "epoch": 555.3539823008849,
      "grad_norm": 1.869250354502583e-08,
      "learning_rate": 0.031087900650654424,
      "loss": 0.0,
      "num_input_tokens_seen": 17982168,
      "step": 31655
    },
    {
      "epoch": 555.4424778761062,
      "grad_norm": 2.573244906045602e-08,
      "learning_rate": 0.031052004338956534,
      "loss": 0.0,
      "num_input_tokens_seen": 17985144,
      "step": 31660
    },
    {
      "epoch": 555.5309734513274,
      "grad_norm": 2.3535612569958175e-08,
      "learning_rate": 0.031016126370534407,
      "loss": 0.0,
      "num_input_tokens_seen": 17988200,
      "step": 31665
    },
    {
      "epoch": 555.6194690265487,
      "grad_norm": 2.6288189403089746e-08,
      "learning_rate": 0.030980266750920804,
      "loss": 0.0,
      "num_input_tokens_seen": 17990808,
      "step": 31670
    },
    {
      "epoch": 555.70796460177,
      "grad_norm": 2.3900422974065805e-08,
      "learning_rate": 0.030944425485645747,
      "loss": 0.0,
      "num_input_tokens_seen": 17993320,
      "step": 31675
    },
    {
      "epoch": 555.7964601769911,
      "grad_norm": 2.417526268061465e-08,
      "learning_rate": 0.03090860258023647,
      "loss": 0.0,
      "num_input_tokens_seen": 17995992,
      "step": 31680
    },
    {
      "epoch": 555.8849557522124,
      "grad_norm": 2.234373219778263e-08,
      "learning_rate": 0.030872798040217236,
      "loss": 0.0,
      "num_input_tokens_seen": 17998856,
      "step": 31685
    },
    {
      "epoch": 555.9734513274336,
      "grad_norm": 6.266418495215476e-08,
      "learning_rate": 0.03083701187110964,
      "loss": 0.0,
      "num_input_tokens_seen": 18001784,
      "step": 31690
    },
    {
      "epoch": 556.0530973451328,
      "grad_norm": 3.0253762162146813e-08,
      "learning_rate": 0.030801244078432294,
      "loss": 0.0,
      "num_input_tokens_seen": 18003784,
      "step": 31695
    },
    {
      "epoch": 556.141592920354,
      "grad_norm": 4.093229932777831e-08,
      "learning_rate": 0.030765494667701024,
      "loss": 0.0,
      "num_input_tokens_seen": 18007000,
      "step": 31700
    },
    {
      "epoch": 556.2300884955753,
      "grad_norm": 2.7375516964411872e-08,
      "learning_rate": 0.030729763644428913,
      "loss": 0.0,
      "num_input_tokens_seen": 18009592,
      "step": 31705
    },
    {
      "epoch": 556.3185840707964,
      "grad_norm": 8.043864596629646e-08,
      "learning_rate": 0.030694051014126048,
      "loss": 0.0,
      "num_input_tokens_seen": 18012728,
      "step": 31710
    },
    {
      "epoch": 556.4070796460177,
      "grad_norm": 1.828237294887458e-08,
      "learning_rate": 0.030658356782299792,
      "loss": 0.0,
      "num_input_tokens_seen": 18015144,
      "step": 31715
    },
    {
      "epoch": 556.4955752212389,
      "grad_norm": 3.643975787781528e-08,
      "learning_rate": 0.030622680954454726,
      "loss": 0.0,
      "num_input_tokens_seen": 18018264,
      "step": 31720
    },
    {
      "epoch": 556.5840707964602,
      "grad_norm": 2.6426654642364156e-08,
      "learning_rate": 0.030587023536092398,
      "loss": 0.0,
      "num_input_tokens_seen": 18020776,
      "step": 31725
    },
    {
      "epoch": 556.6725663716815,
      "grad_norm": 3.458106689890883e-08,
      "learning_rate": 0.03055138453271171,
      "loss": 0.0,
      "num_input_tokens_seen": 18024152,
      "step": 31730
    },
    {
      "epoch": 556.7610619469026,
      "grad_norm": 3.006467963473369e-08,
      "learning_rate": 0.03051576394980858,
      "loss": 0.0,
      "num_input_tokens_seen": 18026984,
      "step": 31735
    },
    {
      "epoch": 556.8495575221239,
      "grad_norm": 2.9623114627952418e-08,
      "learning_rate": 0.030480161792876187,
      "loss": 0.0,
      "num_input_tokens_seen": 18029896,
      "step": 31740
    },
    {
      "epoch": 556.9380530973451,
      "grad_norm": 2.5514179213814714e-08,
      "learning_rate": 0.030444578067404846,
      "loss": 0.0,
      "num_input_tokens_seen": 18033000,
      "step": 31745
    },
    {
      "epoch": 557.0176991150443,
      "grad_norm": 2.4994012193246817e-08,
      "learning_rate": 0.030409012778881975,
      "loss": 0.0,
      "num_input_tokens_seen": 18035296,
      "step": 31750
    },
    {
      "epoch": 557.1061946902655,
      "grad_norm": 6.215282866151028e-08,
      "learning_rate": 0.030373465932792235,
      "loss": 0.0,
      "num_input_tokens_seen": 18038320,
      "step": 31755
    },
    {
      "epoch": 557.1946902654868,
      "grad_norm": 2.907481899683262e-08,
      "learning_rate": 0.030337937534617342,
      "loss": 0.0,
      "num_input_tokens_seen": 18041040,
      "step": 31760
    },
    {
      "epoch": 557.2831858407079,
      "grad_norm": 1.764807144866154e-08,
      "learning_rate": 0.030302427589836277,
      "loss": 0.0,
      "num_input_tokens_seen": 18044080,
      "step": 31765
    },
    {
      "epoch": 557.3716814159292,
      "grad_norm": 3.404825932307176e-08,
      "learning_rate": 0.030266936103925095,
      "loss": 0.0,
      "num_input_tokens_seen": 18046656,
      "step": 31770
    },
    {
      "epoch": 557.4601769911504,
      "grad_norm": 2.076536631534509e-08,
      "learning_rate": 0.030231463082356982,
      "loss": 0.0,
      "num_input_tokens_seen": 18049968,
      "step": 31775
    },
    {
      "epoch": 557.5486725663717,
      "grad_norm": 5.153796678314393e-08,
      "learning_rate": 0.030196008530602367,
      "loss": 0.0,
      "num_input_tokens_seen": 18052640,
      "step": 31780
    },
    {
      "epoch": 557.637168141593,
      "grad_norm": 2.960424083653379e-08,
      "learning_rate": 0.030160572454128842,
      "loss": 0.0,
      "num_input_tokens_seen": 18055680,
      "step": 31785
    },
    {
      "epoch": 557.7256637168142,
      "grad_norm": 2.4410988785916743e-08,
      "learning_rate": 0.03012515485840098,
      "loss": 0.0,
      "num_input_tokens_seen": 18058944,
      "step": 31790
    },
    {
      "epoch": 557.8141592920354,
      "grad_norm": 4.005080711522169e-08,
      "learning_rate": 0.030089755748880734,
      "loss": 0.0,
      "num_input_tokens_seen": 18062080,
      "step": 31795
    },
    {
      "epoch": 557.9026548672566,
      "grad_norm": 2.4467023962415624e-08,
      "learning_rate": 0.030054375131027003,
      "loss": 0.0,
      "num_input_tokens_seen": 18064576,
      "step": 31800
    },
    {
      "epoch": 557.9026548672566,
      "eval_loss": 0.8298357725143433,
      "eval_runtime": 0.942,
      "eval_samples_per_second": 26.539,
      "eval_steps_per_second": 13.8,
      "num_input_tokens_seen": 18064576,
      "step": 31800
    },
    {
      "epoch": 557.9911504424779,
      "grad_norm": 2.7345832265268655e-08,
      "learning_rate": 0.030019013010295942,
      "loss": 0.0,
      "num_input_tokens_seen": 18067184,
      "step": 31805
    },
    {
      "epoch": 558.070796460177,
      "grad_norm": 6.256162521367514e-08,
      "learning_rate": 0.029983669392140897,
      "loss": 0.0,
      "num_input_tokens_seen": 18069656,
      "step": 31810
    },
    {
      "epoch": 558.1592920353983,
      "grad_norm": 3.381957824899473e-08,
      "learning_rate": 0.029948344282012217,
      "loss": 0.0,
      "num_input_tokens_seen": 18072440,
      "step": 31815
    },
    {
      "epoch": 558.2477876106195,
      "grad_norm": 3.403372517141179e-08,
      "learning_rate": 0.029913037685357507,
      "loss": 0.0,
      "num_input_tokens_seen": 18075960,
      "step": 31820
    },
    {
      "epoch": 558.3362831858407,
      "grad_norm": 3.2477558420396235e-08,
      "learning_rate": 0.029877749607621528,
      "loss": 0.0,
      "num_input_tokens_seen": 18079208,
      "step": 31825
    },
    {
      "epoch": 558.4247787610619,
      "grad_norm": 1.88536581902099e-08,
      "learning_rate": 0.029842480054246077,
      "loss": 0.0,
      "num_input_tokens_seen": 18082024,
      "step": 31830
    },
    {
      "epoch": 558.5132743362832,
      "grad_norm": 2.714562263861353e-08,
      "learning_rate": 0.02980722903067022,
      "loss": 0.0,
      "num_input_tokens_seen": 18084456,
      "step": 31835
    },
    {
      "epoch": 558.6017699115044,
      "grad_norm": 5.2803123651301576e-08,
      "learning_rate": 0.029771996542330113,
      "loss": 0.0,
      "num_input_tokens_seen": 18087352,
      "step": 31840
    },
    {
      "epoch": 558.6902654867257,
      "grad_norm": 3.9160958920092526e-08,
      "learning_rate": 0.029736782594658954,
      "loss": 0.0,
      "num_input_tokens_seen": 18090072,
      "step": 31845
    },
    {
      "epoch": 558.7787610619469,
      "grad_norm": 2.982779179205863e-08,
      "learning_rate": 0.029701587193087284,
      "loss": 0.0,
      "num_input_tokens_seen": 18093160,
      "step": 31850
    },
    {
      "epoch": 558.8672566371681,
      "grad_norm": 2.153304734520134e-08,
      "learning_rate": 0.0296664103430426,
      "loss": 0.0,
      "num_input_tokens_seen": 18095976,
      "step": 31855
    },
    {
      "epoch": 558.9557522123894,
      "grad_norm": 1.8007920488116724e-08,
      "learning_rate": 0.029631252049949652,
      "loss": 0.0,
      "num_input_tokens_seen": 18098824,
      "step": 31860
    },
    {
      "epoch": 559.0353982300885,
      "grad_norm": 3.839394935312157e-08,
      "learning_rate": 0.02959611231923031,
      "loss": 0.0,
      "num_input_tokens_seen": 18101160,
      "step": 31865
    },
    {
      "epoch": 559.1238938053098,
      "grad_norm": 1.2693112338979518e-08,
      "learning_rate": 0.029560991156303507,
      "loss": 0.0,
      "num_input_tokens_seen": 18103528,
      "step": 31870
    },
    {
      "epoch": 559.212389380531,
      "grad_norm": 2.7216882969582912e-08,
      "learning_rate": 0.02952588856658544,
      "loss": 0.0,
      "num_input_tokens_seen": 18106248,
      "step": 31875
    },
    {
      "epoch": 559.3008849557522,
      "grad_norm": 2.5439712558750216e-08,
      "learning_rate": 0.029490804555489296,
      "loss": 0.0,
      "num_input_tokens_seen": 18109000,
      "step": 31880
    },
    {
      "epoch": 559.3893805309734,
      "grad_norm": 4.557321986453644e-08,
      "learning_rate": 0.029455739128425484,
      "loss": 0.0,
      "num_input_tokens_seen": 18112120,
      "step": 31885
    },
    {
      "epoch": 559.4778761061947,
      "grad_norm": 3.218249133851714e-08,
      "learning_rate": 0.029420692290801607,
      "loss": 0.0,
      "num_input_tokens_seen": 18115448,
      "step": 31890
    },
    {
      "epoch": 559.566371681416,
      "grad_norm": 2.5356014177191355e-08,
      "learning_rate": 0.02938566404802223,
      "loss": 0.0,
      "num_input_tokens_seen": 18118536,
      "step": 31895
    },
    {
      "epoch": 559.6548672566372,
      "grad_norm": 2.936116061391658e-08,
      "learning_rate": 0.029350654405489195,
      "loss": 0.0,
      "num_input_tokens_seen": 18121528,
      "step": 31900
    },
    {
      "epoch": 559.7433628318585,
      "grad_norm": 2.6928661966962864e-08,
      "learning_rate": 0.02931566336860145,
      "loss": 0.0,
      "num_input_tokens_seen": 18124504,
      "step": 31905
    },
    {
      "epoch": 559.8318584070796,
      "grad_norm": 1.3868465487121284e-08,
      "learning_rate": 0.02928069094275505,
      "loss": 0.0,
      "num_input_tokens_seen": 18127512,
      "step": 31910
    },
    {
      "epoch": 559.9203539823009,
      "grad_norm": 2.4330798709115697e-08,
      "learning_rate": 0.02924573713334314,
      "loss": 0.0,
      "num_input_tokens_seen": 18130072,
      "step": 31915
    },
    {
      "epoch": 560.0,
      "grad_norm": 3.2770607560905773e-08,
      "learning_rate": 0.02921080194575603,
      "loss": 0.0,
      "num_input_tokens_seen": 18132392,
      "step": 31920
    },
    {
      "epoch": 560.0884955752213,
      "grad_norm": 2.9676643364950905e-08,
      "learning_rate": 0.029175885385381177,
      "loss": 0.0,
      "num_input_tokens_seen": 18134904,
      "step": 31925
    },
    {
      "epoch": 560.1769911504425,
      "grad_norm": 3.825673289270526e-08,
      "learning_rate": 0.029140987457603223,
      "loss": 0.0,
      "num_input_tokens_seen": 18138536,
      "step": 31930
    },
    {
      "epoch": 560.2654867256637,
      "grad_norm": 2.9683103974775804e-08,
      "learning_rate": 0.029106108167803763,
      "loss": 0.0,
      "num_input_tokens_seen": 18141096,
      "step": 31935
    },
    {
      "epoch": 560.3539823008849,
      "grad_norm": 4.779053597303573e-08,
      "learning_rate": 0.029071247521361674,
      "loss": 0.0,
      "num_input_tokens_seen": 18143880,
      "step": 31940
    },
    {
      "epoch": 560.4424778761062,
      "grad_norm": 4.9870823914943685e-08,
      "learning_rate": 0.029036405523652945,
      "loss": 0.0,
      "num_input_tokens_seen": 18147144,
      "step": 31945
    },
    {
      "epoch": 560.5309734513274,
      "grad_norm": 3.9890583281021463e-08,
      "learning_rate": 0.029001582180050577,
      "loss": 0.0,
      "num_input_tokens_seen": 18149880,
      "step": 31950
    },
    {
      "epoch": 560.6194690265487,
      "grad_norm": 2.5825666938317227e-08,
      "learning_rate": 0.02896677749592482,
      "loss": 0.0,
      "num_input_tokens_seen": 18153160,
      "step": 31955
    },
    {
      "epoch": 560.70796460177,
      "grad_norm": 1.9866073230900838e-08,
      "learning_rate": 0.028931991476642938,
      "loss": 0.0,
      "num_input_tokens_seen": 18155784,
      "step": 31960
    },
    {
      "epoch": 560.7964601769911,
      "grad_norm": 3.027640005370813e-08,
      "learning_rate": 0.028897224127569412,
      "loss": 0.0,
      "num_input_tokens_seen": 18158344,
      "step": 31965
    },
    {
      "epoch": 560.8849557522124,
      "grad_norm": 2.915602870245948e-08,
      "learning_rate": 0.028862475454065832,
      "loss": 0.0,
      "num_input_tokens_seen": 18161032,
      "step": 31970
    },
    {
      "epoch": 560.9734513274336,
      "grad_norm": 2.4197587933372233e-08,
      "learning_rate": 0.028827745461490806,
      "loss": 0.0,
      "num_input_tokens_seen": 18163976,
      "step": 31975
    },
    {
      "epoch": 561.0530973451328,
      "grad_norm": 2.728687853448264e-08,
      "learning_rate": 0.028793034155200212,
      "loss": 0.0,
      "num_input_tokens_seen": 18166088,
      "step": 31980
    },
    {
      "epoch": 561.141592920354,
      "grad_norm": 9.924986166254257e-09,
      "learning_rate": 0.028758341540546944,
      "loss": 0.0,
      "num_input_tokens_seen": 18168776,
      "step": 31985
    },
    {
      "epoch": 561.2300884955753,
      "grad_norm": 3.079158261698467e-08,
      "learning_rate": 0.02872366762288098,
      "loss": 0.0,
      "num_input_tokens_seen": 18171368,
      "step": 31990
    },
    {
      "epoch": 561.3185840707964,
      "grad_norm": 3.646580282179457e-08,
      "learning_rate": 0.028689012407549567,
      "loss": 0.0,
      "num_input_tokens_seen": 18174280,
      "step": 31995
    },
    {
      "epoch": 561.4070796460177,
      "grad_norm": 2.737489879223176e-08,
      "learning_rate": 0.028654375899896892,
      "loss": 0.0,
      "num_input_tokens_seen": 18177096,
      "step": 32000
    },
    {
      "epoch": 561.4070796460177,
      "eval_loss": 0.8073485493659973,
      "eval_runtime": 0.9421,
      "eval_samples_per_second": 26.535,
      "eval_steps_per_second": 13.798,
      "num_input_tokens_seen": 18177096,
      "step": 32000
    },
    {
      "epoch": 561.4955752212389,
      "grad_norm": 6.593278101263422e-08,
      "learning_rate": 0.02861975810526437,
      "loss": 0.0,
      "num_input_tokens_seen": 18180024,
      "step": 32005
    },
    {
      "epoch": 561.5840707964602,
      "grad_norm": 3.013640181848132e-08,
      "learning_rate": 0.02858515902899056,
      "loss": 0.0,
      "num_input_tokens_seen": 18182696,
      "step": 32010
    },
    {
      "epoch": 561.6725663716815,
      "grad_norm": 4.189232427620482e-08,
      "learning_rate": 0.028550578676410976,
      "loss": 0.0,
      "num_input_tokens_seen": 18185784,
      "step": 32015
    },
    {
      "epoch": 561.7610619469026,
      "grad_norm": 3.2263297811141456e-08,
      "learning_rate": 0.02851601705285837,
      "loss": 0.0,
      "num_input_tokens_seen": 18188264,
      "step": 32020
    },
    {
      "epoch": 561.8495575221239,
      "grad_norm": 3.2265280225374227e-08,
      "learning_rate": 0.028481474163662666,
      "loss": 0.0,
      "num_input_tokens_seen": 18191592,
      "step": 32025
    },
    {
      "epoch": 561.9380530973451,
      "grad_norm": 2.8984558753109013e-08,
      "learning_rate": 0.028446950014150683,
      "loss": 0.0,
      "num_input_tokens_seen": 18194616,
      "step": 32030
    },
    {
      "epoch": 562.0176991150443,
      "grad_norm": 2.788895514527212e-08,
      "learning_rate": 0.028412444609646596,
      "loss": 0.0,
      "num_input_tokens_seen": 18197592,
      "step": 32035
    },
    {
      "epoch": 562.1061946902655,
      "grad_norm": 3.237154899693451e-08,
      "learning_rate": 0.028377957955471465,
      "loss": 0.0,
      "num_input_tokens_seen": 18200152,
      "step": 32040
    },
    {
      "epoch": 562.1946902654868,
      "grad_norm": 5.2415952467299576e-08,
      "learning_rate": 0.0283434900569436,
      "loss": 0.0,
      "num_input_tokens_seen": 18202856,
      "step": 32045
    },
    {
      "epoch": 562.2831858407079,
      "grad_norm": 6.5770898061146e-08,
      "learning_rate": 0.028309040919378456,
      "loss": 0.0,
      "num_input_tokens_seen": 18206424,
      "step": 32050
    },
    {
      "epoch": 562.3716814159292,
      "grad_norm": 4.403535669439407e-08,
      "learning_rate": 0.02827461054808848,
      "loss": 0.0,
      "num_input_tokens_seen": 18208904,
      "step": 32055
    },
    {
      "epoch": 562.4601769911504,
      "grad_norm": 2.3453424091712805e-08,
      "learning_rate": 0.028240198948383186,
      "loss": 0.0,
      "num_input_tokens_seen": 18211496,
      "step": 32060
    },
    {
      "epoch": 562.5486725663717,
      "grad_norm": 2.679263744198579e-08,
      "learning_rate": 0.028205806125569402,
      "loss": 0.0,
      "num_input_tokens_seen": 18214104,
      "step": 32065
    },
    {
      "epoch": 562.637168141593,
      "grad_norm": 4.4978946789342444e-08,
      "learning_rate": 0.028171432084950834,
      "loss": 0.0,
      "num_input_tokens_seen": 18217016,
      "step": 32070
    },
    {
      "epoch": 562.7256637168142,
      "grad_norm": 2.7222410992067125e-08,
      "learning_rate": 0.028137076831828478,
      "loss": 0.0,
      "num_input_tokens_seen": 18219688,
      "step": 32075
    },
    {
      "epoch": 562.8141592920354,
      "grad_norm": 3.742650633853373e-08,
      "learning_rate": 0.028102740371500238,
      "loss": 0.0,
      "num_input_tokens_seen": 18222680,
      "step": 32080
    },
    {
      "epoch": 562.9026548672566,
      "grad_norm": 3.6969943550957396e-08,
      "learning_rate": 0.0280684227092613,
      "loss": 0.0,
      "num_input_tokens_seen": 18226072,
      "step": 32085
    },
    {
      "epoch": 562.9911504424779,
      "grad_norm": 6.437138466708348e-08,
      "learning_rate": 0.02803412385040392,
      "loss": 0.0,
      "num_input_tokens_seen": 18228840,
      "step": 32090
    },
    {
      "epoch": 563.070796460177,
      "grad_norm": 6.781178996106973e-08,
      "learning_rate": 0.027999843800217306,
      "loss": 0.0,
      "num_input_tokens_seen": 18231328,
      "step": 32095
    },
    {
      "epoch": 563.1592920353983,
      "grad_norm": 1.0549150886163261e-08,
      "learning_rate": 0.027965582563987932,
      "loss": 0.0,
      "num_input_tokens_seen": 18234000,
      "step": 32100
    },
    {
      "epoch": 563.2477876106195,
      "grad_norm": 2.5789802293729736e-08,
      "learning_rate": 0.027931340146999346,
      "loss": 0.0,
      "num_input_tokens_seen": 18236352,
      "step": 32105
    },
    {
      "epoch": 563.3362831858407,
      "grad_norm": 3.195966868929645e-08,
      "learning_rate": 0.02789711655453208,
      "loss": 0.0,
      "num_input_tokens_seen": 18239312,
      "step": 32110
    },
    {
      "epoch": 563.4247787610619,
      "grad_norm": 2.4339726678590523e-08,
      "learning_rate": 0.02786291179186392,
      "loss": 0.0,
      "num_input_tokens_seen": 18242208,
      "step": 32115
    },
    {
      "epoch": 563.5132743362832,
      "grad_norm": 2.7263903135121836e-08,
      "learning_rate": 0.02782872586426961,
      "loss": 0.0,
      "num_input_tokens_seen": 18244624,
      "step": 32120
    },
    {
      "epoch": 563.6017699115044,
      "grad_norm": 2.0251736287946187e-08,
      "learning_rate": 0.027794558777021083,
      "loss": 0.0,
      "num_input_tokens_seen": 18247712,
      "step": 32125
    },
    {
      "epoch": 563.6902654867257,
      "grad_norm": 4.2750105677669126e-08,
      "learning_rate": 0.02776041053538734,
      "loss": 0.0,
      "num_input_tokens_seen": 18251168,
      "step": 32130
    },
    {
      "epoch": 563.7787610619469,
      "grad_norm": 2.8182634892459646e-08,
      "learning_rate": 0.027726281144634407,
      "loss": 0.0,
      "num_input_tokens_seen": 18254016,
      "step": 32135
    },
    {
      "epoch": 563.8672566371681,
      "grad_norm": 4.1281658980096836e-08,
      "learning_rate": 0.02769217061002552,
      "loss": 0.0,
      "num_input_tokens_seen": 18257392,
      "step": 32140
    },
    {
      "epoch": 563.9557522123894,
      "grad_norm": 2.923917108432761e-08,
      "learning_rate": 0.027658078936820967,
      "loss": 0.0,
      "num_input_tokens_seen": 18260528,
      "step": 32145
    },
    {
      "epoch": 564.0353982300885,
      "grad_norm": 2.9200583284705317e-08,
      "learning_rate": 0.02762400613027805,
      "loss": 0.0,
      "num_input_tokens_seen": 18263056,
      "step": 32150
    },
    {
      "epoch": 564.1238938053098,
      "grad_norm": 9.65773736538722e-08,
      "learning_rate": 0.027589952195651295,
      "loss": 0.0,
      "num_input_tokens_seen": 18265872,
      "step": 32155
    },
    {
      "epoch": 564.212389380531,
      "grad_norm": 3.699310013871582e-08,
      "learning_rate": 0.027555917138192186,
      "loss": 0.0,
      "num_input_tokens_seen": 18268544,
      "step": 32160
    },
    {
      "epoch": 564.3008849557522,
      "grad_norm": 1.1616115358492607e-07,
      "learning_rate": 0.027521900963149375,
      "loss": 0.0,
      "num_input_tokens_seen": 18271472,
      "step": 32165
    },
    {
      "epoch": 564.3893805309734,
      "grad_norm": 3.3389301989927844e-08,
      "learning_rate": 0.027487903675768633,
      "loss": 0.0,
      "num_input_tokens_seen": 18274480,
      "step": 32170
    },
    {
      "epoch": 564.4778761061947,
      "grad_norm": 1.9287098140807757e-08,
      "learning_rate": 0.027453925281292677,
      "loss": 0.0,
      "num_input_tokens_seen": 18277248,
      "step": 32175
    },
    {
      "epoch": 564.566371681416,
      "grad_norm": 4.8059312973691704e-08,
      "learning_rate": 0.027419965784961475,
      "loss": 0.0,
      "num_input_tokens_seen": 18279808,
      "step": 32180
    },
    {
      "epoch": 564.6548672566372,
      "grad_norm": 2.216758687723086e-08,
      "learning_rate": 0.027386025192012015,
      "loss": 0.0,
      "num_input_tokens_seen": 18282448,
      "step": 32185
    },
    {
      "epoch": 564.7433628318585,
      "grad_norm": 3.0641785997431725e-08,
      "learning_rate": 0.027352103507678277,
      "loss": 0.0,
      "num_input_tokens_seen": 18285472,
      "step": 32190
    },
    {
      "epoch": 564.8318584070796,
      "grad_norm": 5.520942636394466e-08,
      "learning_rate": 0.027318200737191527,
      "loss": 0.0,
      "num_input_tokens_seen": 18287728,
      "step": 32195
    },
    {
      "epoch": 564.9203539823009,
      "grad_norm": 1.777510050260389e-08,
      "learning_rate": 0.027284316885779935,
      "loss": 0.0,
      "num_input_tokens_seen": 18290608,
      "step": 32200
    },
    {
      "epoch": 564.9203539823009,
      "eval_loss": 0.8398646116256714,
      "eval_runtime": 0.9257,
      "eval_samples_per_second": 27.008,
      "eval_steps_per_second": 14.044,
      "num_input_tokens_seen": 18290608,
      "step": 32200
    },
    {
      "epoch": 565.0,
      "grad_norm": 5.1819675661590736e-08,
      "learning_rate": 0.027250451958668785,
      "loss": 0.0,
      "num_input_tokens_seen": 18293712,
      "step": 32205
    },
    {
      "epoch": 565.0884955752213,
      "grad_norm": 2.6454848978119117e-08,
      "learning_rate": 0.027216605961080536,
      "loss": 0.0,
      "num_input_tokens_seen": 18296784,
      "step": 32210
    },
    {
      "epoch": 565.1769911504425,
      "grad_norm": 8.71311954142584e-07,
      "learning_rate": 0.02718277889823461,
      "loss": 0.0,
      "num_input_tokens_seen": 18299248,
      "step": 32215
    },
    {
      "epoch": 565.2654867256637,
      "grad_norm": 8.797027106766109e-08,
      "learning_rate": 0.027148970775347604,
      "loss": 0.0,
      "num_input_tokens_seen": 18302000,
      "step": 32220
    },
    {
      "epoch": 565.3539823008849,
      "grad_norm": 3.618352906187283e-08,
      "learning_rate": 0.027115181597633174,
      "loss": 0.0,
      "num_input_tokens_seen": 18304720,
      "step": 32225
    },
    {
      "epoch": 565.4424778761062,
      "grad_norm": 4.6224425176433215e-08,
      "learning_rate": 0.027081411370301976,
      "loss": 0.0,
      "num_input_tokens_seen": 18307456,
      "step": 32230
    },
    {
      "epoch": 565.5309734513274,
      "grad_norm": 6.115289608032981e-08,
      "learning_rate": 0.027047660098561875,
      "loss": 0.0,
      "num_input_tokens_seen": 18310592,
      "step": 32235
    },
    {
      "epoch": 565.6194690265487,
      "grad_norm": 8.40072900132327e-08,
      "learning_rate": 0.02701392778761766,
      "loss": 0.0,
      "num_input_tokens_seen": 18313648,
      "step": 32240
    },
    {
      "epoch": 565.70796460177,
      "grad_norm": 5.3477950956448694e-08,
      "learning_rate": 0.02698021444267133,
      "loss": 0.0,
      "num_input_tokens_seen": 18316544,
      "step": 32245
    },
    {
      "epoch": 565.7964601769911,
      "grad_norm": 2.821511024819756e-08,
      "learning_rate": 0.026946520068921915,
      "loss": 0.0,
      "num_input_tokens_seen": 18319248,
      "step": 32250
    },
    {
      "epoch": 565.8849557522124,
      "grad_norm": 4.013490695342625e-08,
      "learning_rate": 0.02691284467156547,
      "loss": 0.0,
      "num_input_tokens_seen": 18322176,
      "step": 32255
    },
    {
      "epoch": 565.9734513274336,
      "grad_norm": 1.6650979262067267e-08,
      "learning_rate": 0.026879188255795182,
      "loss": 0.0,
      "num_input_tokens_seen": 18325200,
      "step": 32260
    },
    {
      "epoch": 566.0530973451328,
      "grad_norm": 4.401104192197636e-08,
      "learning_rate": 0.026845550826801328,
      "loss": 0.0,
      "num_input_tokens_seen": 18327664,
      "step": 32265
    },
    {
      "epoch": 566.141592920354,
      "grad_norm": 3.2155664797528516e-08,
      "learning_rate": 0.02681193238977121,
      "loss": 0.0,
      "num_input_tokens_seen": 18330784,
      "step": 32270
    },
    {
      "epoch": 566.2300884955753,
      "grad_norm": 2.5164494488194578e-08,
      "learning_rate": 0.026778332949889145,
      "loss": 0.0,
      "num_input_tokens_seen": 18333424,
      "step": 32275
    },
    {
      "epoch": 566.3185840707964,
      "grad_norm": 3.759303268680014e-08,
      "learning_rate": 0.026744752512336673,
      "loss": 0.0,
      "num_input_tokens_seen": 18336512,
      "step": 32280
    },
    {
      "epoch": 566.4070796460177,
      "grad_norm": 2.6826187493611542e-08,
      "learning_rate": 0.02671119108229225,
      "loss": 0.0,
      "num_input_tokens_seen": 18339616,
      "step": 32285
    },
    {
      "epoch": 566.4955752212389,
      "grad_norm": 4.1719687260410865e-08,
      "learning_rate": 0.026677648664931556,
      "loss": 0.0,
      "num_input_tokens_seen": 18342144,
      "step": 32290
    },
    {
      "epoch": 566.5840707964602,
      "grad_norm": 4.476478210335699e-08,
      "learning_rate": 0.026644125265427154,
      "loss": 0.0,
      "num_input_tokens_seen": 18344944,
      "step": 32295
    },
    {
      "epoch": 566.6725663716815,
      "grad_norm": 2.6721707513388537e-08,
      "learning_rate": 0.026610620888948822,
      "loss": 0.0,
      "num_input_tokens_seen": 18347792,
      "step": 32300
    },
    {
      "epoch": 566.7610619469026,
      "grad_norm": 7.628389653291379e-08,
      "learning_rate": 0.026577135540663408,
      "loss": 0.0,
      "num_input_tokens_seen": 18350816,
      "step": 32305
    },
    {
      "epoch": 566.8495575221239,
      "grad_norm": 1.5132718189647676e-08,
      "learning_rate": 0.026543669225734673,
      "loss": 0.0,
      "num_input_tokens_seen": 18353712,
      "step": 32310
    },
    {
      "epoch": 566.9380530973451,
      "grad_norm": 1.565828711136419e-08,
      "learning_rate": 0.02651022194932363,
      "loss": 0.0,
      "num_input_tokens_seen": 18356640,
      "step": 32315
    },
    {
      "epoch": 567.0176991150443,
      "grad_norm": 3.9371933269194415e-08,
      "learning_rate": 0.026476793716588194,
      "loss": 0.0,
      "num_input_tokens_seen": 18358896,
      "step": 32320
    },
    {
      "epoch": 567.1061946902655,
      "grad_norm": 3.6551973892073875e-08,
      "learning_rate": 0.026443384532683467,
      "loss": 0.0,
      "num_input_tokens_seen": 18361824,
      "step": 32325
    },
    {
      "epoch": 567.1946902654868,
      "grad_norm": 3.1842095182810226e-08,
      "learning_rate": 0.026409994402761584,
      "loss": 0.0,
      "num_input_tokens_seen": 18364736,
      "step": 32330
    },
    {
      "epoch": 567.2831858407079,
      "grad_norm": 1.6364456456585685e-08,
      "learning_rate": 0.026376623331971653,
      "loss": 0.0,
      "num_input_tokens_seen": 18367536,
      "step": 32335
    },
    {
      "epoch": 567.3716814159292,
      "grad_norm": 3.315091134936665e-08,
      "learning_rate": 0.026343271325459997,
      "loss": 0.0,
      "num_input_tokens_seen": 18370400,
      "step": 32340
    },
    {
      "epoch": 567.4601769911504,
      "grad_norm": 3.7247289697006636e-08,
      "learning_rate": 0.02630993838836987,
      "loss": 0.0,
      "num_input_tokens_seen": 18373152,
      "step": 32345
    },
    {
      "epoch": 567.5486725663717,
      "grad_norm": 2.8088948056392837e-08,
      "learning_rate": 0.026276624525841584,
      "loss": 0.0,
      "num_input_tokens_seen": 18376144,
      "step": 32350
    },
    {
      "epoch": 567.637168141593,
      "grad_norm": 3.939155490684243e-08,
      "learning_rate": 0.026243329743012637,
      "loss": 0.0,
      "num_input_tokens_seen": 18378544,
      "step": 32355
    },
    {
      "epoch": 567.7256637168142,
      "grad_norm": 3.4171399931892665e-08,
      "learning_rate": 0.026210054045017438,
      "loss": 0.0,
      "num_input_tokens_seen": 18381360,
      "step": 32360
    },
    {
      "epoch": 567.8141592920354,
      "grad_norm": 1.4622363764260626e-08,
      "learning_rate": 0.02617679743698755,
      "loss": 0.0,
      "num_input_tokens_seen": 18384032,
      "step": 32365
    },
    {
      "epoch": 567.9026548672566,
      "grad_norm": 2.8346343938778773e-08,
      "learning_rate": 0.02614355992405158,
      "loss": 0.0,
      "num_input_tokens_seen": 18387872,
      "step": 32370
    },
    {
      "epoch": 567.9911504424779,
      "grad_norm": 2.8952751307542712e-08,
      "learning_rate": 0.026110341511335115,
      "loss": 0.0,
      "num_input_tokens_seen": 18390832,
      "step": 32375
    },
    {
      "epoch": 568.070796460177,
      "grad_norm": 3.524535330257095e-08,
      "learning_rate": 0.02607714220396093,
      "loss": 0.0,
      "num_input_tokens_seen": 18393512,
      "step": 32380
    },
    {
      "epoch": 568.1592920353983,
      "grad_norm": 2.264764553672194e-08,
      "learning_rate": 0.02604396200704869,
      "loss": 0.0,
      "num_input_tokens_seen": 18396328,
      "step": 32385
    },
    {
      "epoch": 568.2477876106195,
      "grad_norm": 2.538227406034821e-08,
      "learning_rate": 0.02601080092571523,
      "loss": 0.0,
      "num_input_tokens_seen": 18399416,
      "step": 32390
    },
    {
      "epoch": 568.3362831858407,
      "grad_norm": 1.7929195905708184e-08,
      "learning_rate": 0.025977658965074455,
      "loss": 0.0,
      "num_input_tokens_seen": 18402056,
      "step": 32395
    },
    {
      "epoch": 568.4247787610619,
      "grad_norm": 1.2288979611696504e-08,
      "learning_rate": 0.02594453613023719,
      "loss": 0.0,
      "num_input_tokens_seen": 18404648,
      "step": 32400
    },
    {
      "epoch": 568.4247787610619,
      "eval_loss": 0.8064342141151428,
      "eval_runtime": 0.9491,
      "eval_samples_per_second": 26.341,
      "eval_steps_per_second": 13.698,
      "num_input_tokens_seen": 18404648,
      "step": 32400
    },
    {
      "epoch": 568.5132743362832,
      "grad_norm": 2.4374736895538263e-08,
      "learning_rate": 0.025911432426311443,
      "loss": 0.0,
      "num_input_tokens_seen": 18407176,
      "step": 32405
    },
    {
      "epoch": 568.6017699115044,
      "grad_norm": 4.249823959412424e-08,
      "learning_rate": 0.025878347858402234,
      "loss": 0.0,
      "num_input_tokens_seen": 18410408,
      "step": 32410
    },
    {
      "epoch": 568.6902654867257,
      "grad_norm": 6.509661432119174e-08,
      "learning_rate": 0.025845282431611598,
      "loss": 0.0,
      "num_input_tokens_seen": 18413048,
      "step": 32415
    },
    {
      "epoch": 568.7787610619469,
      "grad_norm": 3.148695526533629e-08,
      "learning_rate": 0.025812236151038608,
      "loss": 0.0,
      "num_input_tokens_seen": 18416088,
      "step": 32420
    },
    {
      "epoch": 568.8672566371681,
      "grad_norm": 4.357064398163857e-08,
      "learning_rate": 0.025779209021779468,
      "loss": 0.0,
      "num_input_tokens_seen": 18418936,
      "step": 32425
    },
    {
      "epoch": 568.9557522123894,
      "grad_norm": 3.8739312202551446e-08,
      "learning_rate": 0.025746201048927324,
      "loss": 0.0,
      "num_input_tokens_seen": 18421816,
      "step": 32430
    },
    {
      "epoch": 569.0353982300885,
      "grad_norm": 3.8993999140757296e-08,
      "learning_rate": 0.025713212237572485,
      "loss": 0.0,
      "num_input_tokens_seen": 18424008,
      "step": 32435
    },
    {
      "epoch": 569.1238938053098,
      "grad_norm": 4.583429813465045e-08,
      "learning_rate": 0.025680242592802164,
      "loss": 0.0,
      "num_input_tokens_seen": 18427272,
      "step": 32440
    },
    {
      "epoch": 569.212389380531,
      "grad_norm": 2.3276712113329268e-08,
      "learning_rate": 0.02564729211970073,
      "loss": 0.0,
      "num_input_tokens_seen": 18429656,
      "step": 32445
    },
    {
      "epoch": 569.3008849557522,
      "grad_norm": 2.673321830570785e-08,
      "learning_rate": 0.025614360823349617,
      "loss": 0.0,
      "num_input_tokens_seen": 18432760,
      "step": 32450
    },
    {
      "epoch": 569.3893805309734,
      "grad_norm": 4.569829314959861e-08,
      "learning_rate": 0.025581448708827146,
      "loss": 0.0,
      "num_input_tokens_seen": 18435352,
      "step": 32455
    },
    {
      "epoch": 569.4778761061947,
      "grad_norm": 5.877436137780023e-08,
      "learning_rate": 0.025548555781208876,
      "loss": 0.0,
      "num_input_tokens_seen": 18438264,
      "step": 32460
    },
    {
      "epoch": 569.566371681416,
      "grad_norm": 3.424300132337521e-08,
      "learning_rate": 0.02551568204556721,
      "loss": 0.0,
      "num_input_tokens_seen": 18441176,
      "step": 32465
    },
    {
      "epoch": 569.6548672566372,
      "grad_norm": 2.2227427010079737e-08,
      "learning_rate": 0.02548282750697173,
      "loss": 0.0,
      "num_input_tokens_seen": 18444120,
      "step": 32470
    },
    {
      "epoch": 569.7433628318585,
      "grad_norm": 3.912763446578538e-08,
      "learning_rate": 0.02544999217048909,
      "loss": 0.0,
      "num_input_tokens_seen": 18447480,
      "step": 32475
    },
    {
      "epoch": 569.8318584070796,
      "grad_norm": 7.86404630304105e-09,
      "learning_rate": 0.025417176041182793,
      "loss": 0.0,
      "num_input_tokens_seen": 18450216,
      "step": 32480
    },
    {
      "epoch": 569.9203539823009,
      "grad_norm": 1.8243085264657566e-08,
      "learning_rate": 0.025384379124113596,
      "loss": 0.0,
      "num_input_tokens_seen": 18452776,
      "step": 32485
    },
    {
      "epoch": 570.0,
      "grad_norm": 4.561353250664979e-08,
      "learning_rate": 0.025351601424339124,
      "loss": 0.0,
      "num_input_tokens_seen": 18455456,
      "step": 32490
    },
    {
      "epoch": 570.0884955752213,
      "grad_norm": 2.160119372263125e-08,
      "learning_rate": 0.025318842946914184,
      "loss": 0.0,
      "num_input_tokens_seen": 18458320,
      "step": 32495
    },
    {
      "epoch": 570.1769911504425,
      "grad_norm": 4.4841126367600737e-08,
      "learning_rate": 0.025286103696890494,
      "loss": 0.0,
      "num_input_tokens_seen": 18461104,
      "step": 32500
    },
    {
      "epoch": 570.2654867256637,
      "grad_norm": 4.565321276572831e-08,
      "learning_rate": 0.025253383679316836,
      "loss": 0.0,
      "num_input_tokens_seen": 18464144,
      "step": 32505
    },
    {
      "epoch": 570.3539823008849,
      "grad_norm": 3.064692677412495e-08,
      "learning_rate": 0.025220682899239077,
      "loss": 0.0,
      "num_input_tokens_seen": 18466960,
      "step": 32510
    },
    {
      "epoch": 570.4424778761062,
      "grad_norm": 2.9505649479233398e-08,
      "learning_rate": 0.02518800136170013,
      "loss": 0.0,
      "num_input_tokens_seen": 18469600,
      "step": 32515
    },
    {
      "epoch": 570.5309734513274,
      "grad_norm": 1.699448404224313e-08,
      "learning_rate": 0.02515533907173981,
      "loss": 0.0,
      "num_input_tokens_seen": 18472368,
      "step": 32520
    },
    {
      "epoch": 570.6194690265487,
      "grad_norm": 3.202852738581896e-08,
      "learning_rate": 0.025122696034395115,
      "loss": 0.0,
      "num_input_tokens_seen": 18474832,
      "step": 32525
    },
    {
      "epoch": 570.70796460177,
      "grad_norm": 1.0117230431205826e-07,
      "learning_rate": 0.025090072254700023,
      "loss": 0.0,
      "num_input_tokens_seen": 18478032,
      "step": 32530
    },
    {
      "epoch": 570.7964601769911,
      "grad_norm": 5.329284746835583e-08,
      "learning_rate": 0.025057467737685468,
      "loss": 0.0,
      "num_input_tokens_seen": 18481072,
      "step": 32535
    },
    {
      "epoch": 570.8849557522124,
      "grad_norm": 3.915934598808235e-08,
      "learning_rate": 0.025024882488379557,
      "loss": 0.0,
      "num_input_tokens_seen": 18484096,
      "step": 32540
    },
    {
      "epoch": 570.9734513274336,
      "grad_norm": 2.3318298403296467e-08,
      "learning_rate": 0.02499231651180727,
      "loss": 0.0,
      "num_input_tokens_seen": 18487328,
      "step": 32545
    },
    {
      "epoch": 571.0530973451328,
      "grad_norm": 2.9327864581318863e-08,
      "learning_rate": 0.024959769812990713,
      "loss": 0.0,
      "num_input_tokens_seen": 18490064,
      "step": 32550
    },
    {
      "epoch": 571.141592920354,
      "grad_norm": 2.7031818561340515e-08,
      "learning_rate": 0.024927242396949045,
      "loss": 0.0,
      "num_input_tokens_seen": 18492528,
      "step": 32555
    },
    {
      "epoch": 571.2300884955753,
      "grad_norm": 4.288986943379314e-08,
      "learning_rate": 0.02489473426869836,
      "loss": 0.0,
      "num_input_tokens_seen": 18495648,
      "step": 32560
    },
    {
      "epoch": 571.3185840707964,
      "grad_norm": 4.415762333564999e-08,
      "learning_rate": 0.024862245433251776,
      "loss": 0.0,
      "num_input_tokens_seen": 18498288,
      "step": 32565
    },
    {
      "epoch": 571.4070796460177,
      "grad_norm": 3.2704079444556555e-08,
      "learning_rate": 0.024829775895619577,
      "loss": 0.0,
      "num_input_tokens_seen": 18500960,
      "step": 32570
    },
    {
      "epoch": 571.4955752212389,
      "grad_norm": 5.661850721594419e-08,
      "learning_rate": 0.024797325660808882,
      "loss": 0.0,
      "num_input_tokens_seen": 18503616,
      "step": 32575
    },
    {
      "epoch": 571.5840707964602,
      "grad_norm": 1.1621274609296961e-08,
      "learning_rate": 0.02476489473382401,
      "loss": 0.0,
      "num_input_tokens_seen": 18506528,
      "step": 32580
    },
    {
      "epoch": 571.6725663716815,
      "grad_norm": 6.148153630647357e-08,
      "learning_rate": 0.024732483119666127,
      "loss": 0.0,
      "num_input_tokens_seen": 18509696,
      "step": 32585
    },
    {
      "epoch": 571.7610619469026,
      "grad_norm": 3.9085314540443505e-08,
      "learning_rate": 0.024700090823333548,
      "loss": 0.0,
      "num_input_tokens_seen": 18512336,
      "step": 32590
    },
    {
      "epoch": 571.8495575221239,
      "grad_norm": 1.3702083379030228e-07,
      "learning_rate": 0.02466771784982163,
      "loss": 0.0,
      "num_input_tokens_seen": 18514880,
      "step": 32595
    },
    {
      "epoch": 571.9380530973451,
      "grad_norm": 2.2061934501493852e-08,
      "learning_rate": 0.024635364204122594,
      "loss": 0.0,
      "num_input_tokens_seen": 18517216,
      "step": 32600
    },
    {
      "epoch": 571.9380530973451,
      "eval_loss": 0.8300523161888123,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 26.705,
      "eval_steps_per_second": 13.886,
      "num_input_tokens_seen": 18517216,
      "step": 32600
    },
    {
      "epoch": 572.0176991150443,
      "grad_norm": 3.810940540915908e-08,
      "learning_rate": 0.024603029891225852,
      "loss": 0.0,
      "num_input_tokens_seen": 18520328,
      "step": 32605
    },
    {
      "epoch": 572.1061946902655,
      "grad_norm": 3.867392450729312e-08,
      "learning_rate": 0.024570714916117748,
      "loss": 0.0,
      "num_input_tokens_seen": 18523496,
      "step": 32610
    },
    {
      "epoch": 572.1946902654868,
      "grad_norm": 3.486623256776511e-08,
      "learning_rate": 0.024538419283781625,
      "loss": 0.0,
      "num_input_tokens_seen": 18526248,
      "step": 32615
    },
    {
      "epoch": 572.2831858407079,
      "grad_norm": 3.9771745008465587e-08,
      "learning_rate": 0.024506142999197938,
      "loss": 0.0,
      "num_input_tokens_seen": 18528968,
      "step": 32620
    },
    {
      "epoch": 572.3716814159292,
      "grad_norm": 2.3197344489744864e-08,
      "learning_rate": 0.024473886067344002,
      "loss": 0.0,
      "num_input_tokens_seen": 18532024,
      "step": 32625
    },
    {
      "epoch": 572.4601769911504,
      "grad_norm": 1.4141008364276786e-08,
      "learning_rate": 0.02444164849319434,
      "loss": 0.0,
      "num_input_tokens_seen": 18534840,
      "step": 32630
    },
    {
      "epoch": 572.5486725663717,
      "grad_norm": 4.011571874684705e-08,
      "learning_rate": 0.024409430281720306,
      "loss": 0.0,
      "num_input_tokens_seen": 18537848,
      "step": 32635
    },
    {
      "epoch": 572.637168141593,
      "grad_norm": 3.42027490773944e-08,
      "learning_rate": 0.024377231437890428,
      "loss": 0.0,
      "num_input_tokens_seen": 18540392,
      "step": 32640
    },
    {
      "epoch": 572.7256637168142,
      "grad_norm": 7.630064402519565e-08,
      "learning_rate": 0.024345051966670115,
      "loss": 0.0,
      "num_input_tokens_seen": 18543512,
      "step": 32645
    },
    {
      "epoch": 572.8141592920354,
      "grad_norm": 2.401521115302785e-08,
      "learning_rate": 0.024312891873021884,
      "loss": 0.0,
      "num_input_tokens_seen": 18546328,
      "step": 32650
    },
    {
      "epoch": 572.9026548672566,
      "grad_norm": 2.8659934869779136e-08,
      "learning_rate": 0.024280751161905183,
      "loss": 0.0,
      "num_input_tokens_seen": 18549272,
      "step": 32655
    },
    {
      "epoch": 572.9911504424779,
      "grad_norm": 2.2277582445440203e-08,
      "learning_rate": 0.02424862983827658,
      "loss": 0.0,
      "num_input_tokens_seen": 18552280,
      "step": 32660
    },
    {
      "epoch": 573.070796460177,
      "grad_norm": 6.892199166941282e-08,
      "learning_rate": 0.024216527907089495,
      "loss": 0.0,
      "num_input_tokens_seen": 18554696,
      "step": 32665
    },
    {
      "epoch": 573.1592920353983,
      "grad_norm": 3.356663214049149e-08,
      "learning_rate": 0.024184445373294505,
      "loss": 0.0,
      "num_input_tokens_seen": 18557464,
      "step": 32670
    },
    {
      "epoch": 573.2477876106195,
      "grad_norm": 3.436805684486899e-08,
      "learning_rate": 0.02415238224183918,
      "loss": 0.0,
      "num_input_tokens_seen": 18560728,
      "step": 32675
    },
    {
      "epoch": 573.3362831858407,
      "grad_norm": 2.6632571703544272e-08,
      "learning_rate": 0.024120338517667973,
      "loss": 0.0,
      "num_input_tokens_seen": 18563304,
      "step": 32680
    },
    {
      "epoch": 573.4247787610619,
      "grad_norm": 6.964615550941744e-08,
      "learning_rate": 0.02408831420572247,
      "loss": 0.0,
      "num_input_tokens_seen": 18566248,
      "step": 32685
    },
    {
      "epoch": 573.5132743362832,
      "grad_norm": 3.350388055878284e-08,
      "learning_rate": 0.024056309310941264,
      "loss": 0.0,
      "num_input_tokens_seen": 18569528,
      "step": 32690
    },
    {
      "epoch": 573.6017699115044,
      "grad_norm": 1.6335230057507033e-08,
      "learning_rate": 0.02402432383825982,
      "loss": 0.0,
      "num_input_tokens_seen": 18571992,
      "step": 32695
    },
    {
      "epoch": 573.6902654867257,
      "grad_norm": 2.749369620858033e-08,
      "learning_rate": 0.023992357792610792,
      "loss": 0.0,
      "num_input_tokens_seen": 18574296,
      "step": 32700
    },
    {
      "epoch": 573.7787610619469,
      "grad_norm": 9.928473332365684e-08,
      "learning_rate": 0.0239604111789237,
      "loss": 0.0,
      "num_input_tokens_seen": 18577624,
      "step": 32705
    },
    {
      "epoch": 573.8672566371681,
      "grad_norm": 1.4345087784306543e-08,
      "learning_rate": 0.023928484002125095,
      "loss": 0.0,
      "num_input_tokens_seen": 18581016,
      "step": 32710
    },
    {
      "epoch": 573.9557522123894,
      "grad_norm": 3.1131055067135094e-08,
      "learning_rate": 0.023896576267138595,
      "loss": 0.0,
      "num_input_tokens_seen": 18583576,
      "step": 32715
    },
    {
      "epoch": 574.0353982300885,
      "grad_norm": 3.827015149227009e-08,
      "learning_rate": 0.02386468797888471,
      "loss": 0.0,
      "num_input_tokens_seen": 18585736,
      "step": 32720
    },
    {
      "epoch": 574.1238938053098,
      "grad_norm": 2.43714293191033e-08,
      "learning_rate": 0.023832819142281057,
      "loss": 0.0,
      "num_input_tokens_seen": 18588888,
      "step": 32725
    },
    {
      "epoch": 574.212389380531,
      "grad_norm": 2.878126714733753e-08,
      "learning_rate": 0.02380096976224225,
      "loss": 0.0,
      "num_input_tokens_seen": 18592088,
      "step": 32730
    },
    {
      "epoch": 574.3008849557522,
      "grad_norm": 1.6510384170942416e-08,
      "learning_rate": 0.023769139843679777,
      "loss": 0.0,
      "num_input_tokens_seen": 18594728,
      "step": 32735
    },
    {
      "epoch": 574.3893805309734,
      "grad_norm": 1.8461784989654006e-08,
      "learning_rate": 0.023737329391502287,
      "loss": 0.0,
      "num_input_tokens_seen": 18598120,
      "step": 32740
    },
    {
      "epoch": 574.4778761061947,
      "grad_norm": 4.290354738145652e-08,
      "learning_rate": 0.023705538410615293,
      "loss": 0.0,
      "num_input_tokens_seen": 18600968,
      "step": 32745
    },
    {
      "epoch": 574.566371681416,
      "grad_norm": 1.0395989846756493e-07,
      "learning_rate": 0.023673766905921396,
      "loss": 0.0,
      "num_input_tokens_seen": 18603944,
      "step": 32750
    },
    {
      "epoch": 574.6548672566372,
      "grad_norm": 2.8595421142085797e-08,
      "learning_rate": 0.0236420148823202,
      "loss": 0.0,
      "num_input_tokens_seen": 18606520,
      "step": 32755
    },
    {
      "epoch": 574.7433628318585,
      "grad_norm": 3.7216370429860035e-08,
      "learning_rate": 0.02361028234470816,
      "loss": 0.0,
      "num_input_tokens_seen": 18609512,
      "step": 32760
    },
    {
      "epoch": 574.8318584070796,
      "grad_norm": 7.77057849177254e-08,
      "learning_rate": 0.023578569297978913,
      "loss": 0.0,
      "num_input_tokens_seen": 18612184,
      "step": 32765
    },
    {
      "epoch": 574.9203539823009,
      "grad_norm": 1.9879667689792768e-08,
      "learning_rate": 0.023546875747023025,
      "loss": 0.0,
      "num_input_tokens_seen": 18614952,
      "step": 32770
    },
    {
      "epoch": 575.0,
      "grad_norm": 2.0458056582128847e-08,
      "learning_rate": 0.02351520169672801,
      "loss": 0.0,
      "num_input_tokens_seen": 18617264,
      "step": 32775
    },
    {
      "epoch": 575.0884955752213,
      "grad_norm": 7.770520227268207e-08,
      "learning_rate": 0.023483547151978357,
      "loss": 0.0,
      "num_input_tokens_seen": 18619920,
      "step": 32780
    },
    {
      "epoch": 575.1769911504425,
      "grad_norm": 8.760043357369796e-08,
      "learning_rate": 0.023451912117655675,
      "loss": 0.0,
      "num_input_tokens_seen": 18622800,
      "step": 32785
    },
    {
      "epoch": 575.2654867256637,
      "grad_norm": 3.2108722791690525e-08,
      "learning_rate": 0.023420296598638417,
      "loss": 0.0,
      "num_input_tokens_seen": 18625872,
      "step": 32790
    },
    {
      "epoch": 575.3539823008849,
      "grad_norm": 3.0624310198845706e-08,
      "learning_rate": 0.023388700599802165,
      "loss": 0.0,
      "num_input_tokens_seen": 18628464,
      "step": 32795
    },
    {
      "epoch": 575.4424778761062,
      "grad_norm": 1.9383962879260253e-08,
      "learning_rate": 0.023357124126019334,
      "loss": 0.0,
      "num_input_tokens_seen": 18631296,
      "step": 32800
    },
    {
      "epoch": 575.4424778761062,
      "eval_loss": 0.8266446590423584,
      "eval_runtime": 0.9399,
      "eval_samples_per_second": 26.597,
      "eval_steps_per_second": 13.831,
      "num_input_tokens_seen": 18631296,
      "step": 32800
    },
    {
      "epoch": 575.5309734513274,
      "grad_norm": 8.214475144541211e-08,
      "learning_rate": 0.02332556718215945,
      "loss": 0.0,
      "num_input_tokens_seen": 18634368,
      "step": 32805
    },
    {
      "epoch": 575.6194690265487,
      "grad_norm": 4.868726222184705e-08,
      "learning_rate": 0.023294029773089035,
      "loss": 0.0,
      "num_input_tokens_seen": 18637280,
      "step": 32810
    },
    {
      "epoch": 575.70796460177,
      "grad_norm": 4.4034806023773854e-08,
      "learning_rate": 0.023262511903671484,
      "loss": 0.0,
      "num_input_tokens_seen": 18640048,
      "step": 32815
    },
    {
      "epoch": 575.7964601769911,
      "grad_norm": 1.9528227923615304e-08,
      "learning_rate": 0.023231013578767324,
      "loss": 0.0,
      "num_input_tokens_seen": 18642432,
      "step": 32820
    },
    {
      "epoch": 575.8849557522124,
      "grad_norm": 5.9432824883742796e-08,
      "learning_rate": 0.0231995348032339,
      "loss": 0.0,
      "num_input_tokens_seen": 18645696,
      "step": 32825
    },
    {
      "epoch": 575.9734513274336,
      "grad_norm": 2.0487650687073256e-08,
      "learning_rate": 0.023168075581925685,
      "loss": 0.0,
      "num_input_tokens_seen": 18649200,
      "step": 32830
    },
    {
      "epoch": 576.0530973451328,
      "grad_norm": 6.523509910039138e-08,
      "learning_rate": 0.023136635919694126,
      "loss": 0.0,
      "num_input_tokens_seen": 18651216,
      "step": 32835
    },
    {
      "epoch": 576.141592920354,
      "grad_norm": 2.8305162658170957e-08,
      "learning_rate": 0.02310521582138753,
      "loss": 0.0,
      "num_input_tokens_seen": 18653952,
      "step": 32840
    },
    {
      "epoch": 576.2300884955753,
      "grad_norm": 3.687975080879369e-08,
      "learning_rate": 0.023073815291851357,
      "loss": 0.0,
      "num_input_tokens_seen": 18656352,
      "step": 32845
    },
    {
      "epoch": 576.3185840707964,
      "grad_norm": 3.313226670798031e-08,
      "learning_rate": 0.02304243433592788,
      "loss": 0.0,
      "num_input_tokens_seen": 18660192,
      "step": 32850
    },
    {
      "epoch": 576.4070796460177,
      "grad_norm": 3.709507012672475e-08,
      "learning_rate": 0.023011072958456513,
      "loss": 0.0,
      "num_input_tokens_seen": 18663600,
      "step": 32855
    },
    {
      "epoch": 576.4955752212389,
      "grad_norm": 3.892803590588301e-08,
      "learning_rate": 0.022979731164273536,
      "loss": 0.0,
      "num_input_tokens_seen": 18666096,
      "step": 32860
    },
    {
      "epoch": 576.5840707964602,
      "grad_norm": 5.393455282387549e-08,
      "learning_rate": 0.022948408958212218,
      "loss": 0.0,
      "num_input_tokens_seen": 18668576,
      "step": 32865
    },
    {
      "epoch": 576.6725663716815,
      "grad_norm": 3.763681633017768e-08,
      "learning_rate": 0.022917106345102876,
      "loss": 0.0,
      "num_input_tokens_seen": 18671504,
      "step": 32870
    },
    {
      "epoch": 576.7610619469026,
      "grad_norm": 3.3128120691117147e-08,
      "learning_rate": 0.022885823329772785,
      "loss": 0.0,
      "num_input_tokens_seen": 18674448,
      "step": 32875
    },
    {
      "epoch": 576.8495575221239,
      "grad_norm": 2.713363578266126e-08,
      "learning_rate": 0.02285455991704612,
      "loss": 0.0,
      "num_input_tokens_seen": 18677680,
      "step": 32880
    },
    {
      "epoch": 576.9380530973451,
      "grad_norm": 1.3707372126248174e-08,
      "learning_rate": 0.022823316111744117,
      "loss": 0.0,
      "num_input_tokens_seen": 18680368,
      "step": 32885
    },
    {
      "epoch": 577.0176991150443,
      "grad_norm": 2.506114782363511e-08,
      "learning_rate": 0.022792091918685014,
      "loss": 0.0,
      "num_input_tokens_seen": 18682512,
      "step": 32890
    },
    {
      "epoch": 577.1061946902655,
      "grad_norm": 1.3443615998198766e-08,
      "learning_rate": 0.022760887342683906,
      "loss": 0.0,
      "num_input_tokens_seen": 18685056,
      "step": 32895
    },
    {
      "epoch": 577.1946902654868,
      "grad_norm": 1.3644072538454566e-08,
      "learning_rate": 0.022729702388552975,
      "loss": 0.0,
      "num_input_tokens_seen": 18688352,
      "step": 32900
    },
    {
      "epoch": 577.2831858407079,
      "grad_norm": 2.430778067719075e-08,
      "learning_rate": 0.022698537061101292,
      "loss": 0.0,
      "num_input_tokens_seen": 18691200,
      "step": 32905
    },
    {
      "epoch": 577.3716814159292,
      "grad_norm": 2.278237509756309e-08,
      "learning_rate": 0.022667391365134962,
      "loss": 0.0,
      "num_input_tokens_seen": 18693472,
      "step": 32910
    },
    {
      "epoch": 577.4601769911504,
      "grad_norm": 3.43506094679924e-08,
      "learning_rate": 0.022636265305457065,
      "loss": 0.0,
      "num_input_tokens_seen": 18696320,
      "step": 32915
    },
    {
      "epoch": 577.5486725663717,
      "grad_norm": 3.028706174745821e-08,
      "learning_rate": 0.02260515888686764,
      "loss": 0.0,
      "num_input_tokens_seen": 18699424,
      "step": 32920
    },
    {
      "epoch": 577.637168141593,
      "grad_norm": 5.34940376439863e-08,
      "learning_rate": 0.022574072114163596,
      "loss": 0.0,
      "num_input_tokens_seen": 18702352,
      "step": 32925
    },
    {
      "epoch": 577.7256637168142,
      "grad_norm": 1.7653098538517042e-08,
      "learning_rate": 0.022543004992139005,
      "loss": 0.0,
      "num_input_tokens_seen": 18705824,
      "step": 32930
    },
    {
      "epoch": 577.8141592920354,
      "grad_norm": 2.9083150110409406e-08,
      "learning_rate": 0.022511957525584745,
      "loss": 0.0,
      "num_input_tokens_seen": 18708848,
      "step": 32935
    },
    {
      "epoch": 577.9026548672566,
      "grad_norm": 4.0491411112952846e-08,
      "learning_rate": 0.022480929719288778,
      "loss": 0.0,
      "num_input_tokens_seen": 18711600,
      "step": 32940
    },
    {
      "epoch": 577.9911504424779,
      "grad_norm": 1.3237036355917553e-08,
      "learning_rate": 0.02244992157803592,
      "loss": 0.0,
      "num_input_tokens_seen": 18714512,
      "step": 32945
    },
    {
      "epoch": 578.070796460177,
      "grad_norm": 3.4463468523426855e-08,
      "learning_rate": 0.022418933106608047,
      "loss": 0.0,
      "num_input_tokens_seen": 18716952,
      "step": 32950
    },
    {
      "epoch": 578.1592920353983,
      "grad_norm": 2.81897563070288e-08,
      "learning_rate": 0.022387964309784018,
      "loss": 0.0,
      "num_input_tokens_seen": 18719576,
      "step": 32955
    },
    {
      "epoch": 578.2477876106195,
      "grad_norm": 3.93933525799639e-08,
      "learning_rate": 0.022357015192339517,
      "loss": 0.0,
      "num_input_tokens_seen": 18722424,
      "step": 32960
    },
    {
      "epoch": 578.3362831858407,
      "grad_norm": 2.1646968662025756e-08,
      "learning_rate": 0.02232608575904734,
      "loss": 0.0,
      "num_input_tokens_seen": 18725784,
      "step": 32965
    },
    {
      "epoch": 578.4247787610619,
      "grad_norm": 3.3252820941243044e-08,
      "learning_rate": 0.022295176014677225,
      "loss": 0.0,
      "num_input_tokens_seen": 18728648,
      "step": 32970
    },
    {
      "epoch": 578.5132743362832,
      "grad_norm": 4.0535486078852045e-08,
      "learning_rate": 0.02226428596399577,
      "loss": 0.0,
      "num_input_tokens_seen": 18731080,
      "step": 32975
    },
    {
      "epoch": 578.6017699115044,
      "grad_norm": 1.2499777213292873e-08,
      "learning_rate": 0.02223341561176669,
      "loss": 0.0,
      "num_input_tokens_seen": 18733704,
      "step": 32980
    },
    {
      "epoch": 578.6902654867257,
      "grad_norm": 1.4350925781059232e-08,
      "learning_rate": 0.0222025649627505,
      "loss": 0.0,
      "num_input_tokens_seen": 18736664,
      "step": 32985
    },
    {
      "epoch": 578.7787610619469,
      "grad_norm": 2.3933978354762075e-08,
      "learning_rate": 0.022171734021704814,
      "loss": 0.0,
      "num_input_tokens_seen": 18739928,
      "step": 32990
    },
    {
      "epoch": 578.8672566371681,
      "grad_norm": 3.0690504360109117e-08,
      "learning_rate": 0.022140922793384116,
      "loss": 0.0,
      "num_input_tokens_seen": 18742552,
      "step": 32995
    },
    {
      "epoch": 578.9557522123894,
      "grad_norm": 9.433294678728998e-08,
      "learning_rate": 0.022110131282539934,
      "loss": 0.0,
      "num_input_tokens_seen": 18745416,
      "step": 33000
    },
    {
      "epoch": 578.9557522123894,
      "eval_loss": 0.8064060807228088,
      "eval_runtime": 0.9353,
      "eval_samples_per_second": 26.728,
      "eval_steps_per_second": 13.899,
      "num_input_tokens_seen": 18745416,
      "step": 33000
    },
    {
      "epoch": 579.0353982300885,
      "grad_norm": 4.561278643677724e-08,
      "learning_rate": 0.022079359493920675,
      "loss": 0.0,
      "num_input_tokens_seen": 18747704,
      "step": 33005
    },
    {
      "epoch": 579.1238938053098,
      "grad_norm": 7.839597060410597e-08,
      "learning_rate": 0.02204860743227169,
      "loss": 0.0,
      "num_input_tokens_seen": 18750712,
      "step": 33010
    },
    {
      "epoch": 579.212389380531,
      "grad_norm": 4.3777475866590976e-08,
      "learning_rate": 0.022017875102335365,
      "loss": 0.0,
      "num_input_tokens_seen": 18753192,
      "step": 33015
    },
    {
      "epoch": 579.3008849557522,
      "grad_norm": 3.1071223816070415e-08,
      "learning_rate": 0.02198716250885108,
      "loss": 0.0,
      "num_input_tokens_seen": 18756168,
      "step": 33020
    },
    {
      "epoch": 579.3893805309734,
      "grad_norm": 2.4335571779943166e-08,
      "learning_rate": 0.021956469656555,
      "loss": 0.0,
      "num_input_tokens_seen": 18759176,
      "step": 33025
    },
    {
      "epoch": 579.4778761061947,
      "grad_norm": 2.1685004014670994e-08,
      "learning_rate": 0.0219257965501804,
      "loss": 0.0,
      "num_input_tokens_seen": 18761576,
      "step": 33030
    },
    {
      "epoch": 579.566371681416,
      "grad_norm": 3.811739901493638e-08,
      "learning_rate": 0.021895143194457494,
      "loss": 0.0,
      "num_input_tokens_seen": 18764488,
      "step": 33035
    },
    {
      "epoch": 579.6548672566372,
      "grad_norm": 1.8267517276626677e-08,
      "learning_rate": 0.021864509594113322,
      "loss": 0.0,
      "num_input_tokens_seen": 18767128,
      "step": 33040
    },
    {
      "epoch": 579.7433628318585,
      "grad_norm": 4.0056733041637926e-08,
      "learning_rate": 0.02183389575387207,
      "loss": 0.0,
      "num_input_tokens_seen": 18770376,
      "step": 33045
    },
    {
      "epoch": 579.8318584070796,
      "grad_norm": 9.194931038791765e-08,
      "learning_rate": 0.021803301678454682,
      "loss": 0.0,
      "num_input_tokens_seen": 18773064,
      "step": 33050
    },
    {
      "epoch": 579.9203539823009,
      "grad_norm": 2.902517692859874e-08,
      "learning_rate": 0.021772727372579213,
      "loss": 0.0,
      "num_input_tokens_seen": 18776232,
      "step": 33055
    },
    {
      "epoch": 580.0,
      "grad_norm": 4.246261653406691e-08,
      "learning_rate": 0.02174217284096061,
      "loss": 0.0,
      "num_input_tokens_seen": 18778768,
      "step": 33060
    },
    {
      "epoch": 580.0884955752213,
      "grad_norm": 3.27967804025775e-08,
      "learning_rate": 0.0217116380883107,
      "loss": 0.0,
      "num_input_tokens_seen": 18781232,
      "step": 33065
    },
    {
      "epoch": 580.1769911504425,
      "grad_norm": 3.4815631266837954e-08,
      "learning_rate": 0.021681123119338425,
      "loss": 0.0,
      "num_input_tokens_seen": 18784032,
      "step": 33070
    },
    {
      "epoch": 580.2654867256637,
      "grad_norm": 1.936228777310589e-08,
      "learning_rate": 0.02165062793874951,
      "loss": 0.0,
      "num_input_tokens_seen": 18786496,
      "step": 33075
    },
    {
      "epoch": 580.3539823008849,
      "grad_norm": 2.428509127128109e-08,
      "learning_rate": 0.021620152551246666,
      "loss": 0.0,
      "num_input_tokens_seen": 18789728,
      "step": 33080
    },
    {
      "epoch": 580.4424778761062,
      "grad_norm": 1.3759491324094597e-08,
      "learning_rate": 0.02158969696152967,
      "loss": 0.0,
      "num_input_tokens_seen": 18792496,
      "step": 33085
    },
    {
      "epoch": 580.5309734513274,
      "grad_norm": 2.2603783733643468e-08,
      "learning_rate": 0.021559261174295057,
      "loss": 0.0,
      "num_input_tokens_seen": 18795040,
      "step": 33090
    },
    {
      "epoch": 580.6194690265487,
      "grad_norm": 6.647105266210929e-08,
      "learning_rate": 0.02152884519423646,
      "loss": 0.0,
      "num_input_tokens_seen": 18798128,
      "step": 33095
    },
    {
      "epoch": 580.70796460177,
      "grad_norm": 5.505593847487944e-08,
      "learning_rate": 0.021498449026044447,
      "loss": 0.0,
      "num_input_tokens_seen": 18801024,
      "step": 33100
    },
    {
      "epoch": 580.7964601769911,
      "grad_norm": 3.197598630322318e-08,
      "learning_rate": 0.021468072674406414,
      "loss": 0.0,
      "num_input_tokens_seen": 18803984,
      "step": 33105
    },
    {
      "epoch": 580.8849557522124,
      "grad_norm": 3.206887910778278e-08,
      "learning_rate": 0.021437716144006795,
      "loss": 0.0,
      "num_input_tokens_seen": 18806832,
      "step": 33110
    },
    {
      "epoch": 580.9734513274336,
      "grad_norm": 5.1960629576797146e-08,
      "learning_rate": 0.021407379439527002,
      "loss": 0.0,
      "num_input_tokens_seen": 18809840,
      "step": 33115
    },
    {
      "epoch": 581.0530973451328,
      "grad_norm": 2.0184886651009037e-08,
      "learning_rate": 0.021377062565645255,
      "loss": 0.0,
      "num_input_tokens_seen": 18812080,
      "step": 33120
    },
    {
      "epoch": 581.141592920354,
      "grad_norm": 2.8578758914932223e-08,
      "learning_rate": 0.02134676552703688,
      "loss": 0.0,
      "num_input_tokens_seen": 18814976,
      "step": 33125
    },
    {
      "epoch": 581.2300884955753,
      "grad_norm": 3.902390588450544e-08,
      "learning_rate": 0.02131648832837398,
      "loss": 0.0,
      "num_input_tokens_seen": 18817808,
      "step": 33130
    },
    {
      "epoch": 581.3185840707964,
      "grad_norm": 1.9326996891777526e-08,
      "learning_rate": 0.02128623097432574,
      "loss": 0.0,
      "num_input_tokens_seen": 18820512,
      "step": 33135
    },
    {
      "epoch": 581.4070796460177,
      "grad_norm": 3.8728888540617845e-08,
      "learning_rate": 0.021255993469558192,
      "loss": 0.0,
      "num_input_tokens_seen": 18823232,
      "step": 33140
    },
    {
      "epoch": 581.4955752212389,
      "grad_norm": 2.793134434853073e-08,
      "learning_rate": 0.021225775818734364,
      "loss": 0.0,
      "num_input_tokens_seen": 18825856,
      "step": 33145
    },
    {
      "epoch": 581.5840707964602,
      "grad_norm": 4.8859803314371675e-08,
      "learning_rate": 0.021195578026514166,
      "loss": 0.0,
      "num_input_tokens_seen": 18828608,
      "step": 33150
    },
    {
      "epoch": 581.6725663716815,
      "grad_norm": 3.295809491987711e-08,
      "learning_rate": 0.02116540009755452,
      "loss": 0.0,
      "num_input_tokens_seen": 18831824,
      "step": 33155
    },
    {
      "epoch": 581.7610619469026,
      "grad_norm": 4.7761314903027596e-08,
      "learning_rate": 0.021135242036509173,
      "loss": 0.0,
      "num_input_tokens_seen": 18834416,
      "step": 33160
    },
    {
      "epoch": 581.8495575221239,
      "grad_norm": 6.672532038010104e-08,
      "learning_rate": 0.021105103848028967,
      "loss": 0.0,
      "num_input_tokens_seen": 18837776,
      "step": 33165
    },
    {
      "epoch": 581.9380530973451,
      "grad_norm": 7.822969649851075e-08,
      "learning_rate": 0.021074985536761504,
      "loss": 0.0,
      "num_input_tokens_seen": 18840480,
      "step": 33170
    },
    {
      "epoch": 582.0176991150443,
      "grad_norm": 4.0053386385352496e-08,
      "learning_rate": 0.021044887107351435,
      "loss": 0.0,
      "num_input_tokens_seen": 18843368,
      "step": 33175
    },
    {
      "epoch": 582.1061946902655,
      "grad_norm": 4.239971573838375e-08,
      "learning_rate": 0.021014808564440362,
      "loss": 0.0,
      "num_input_tokens_seen": 18846072,
      "step": 33180
    },
    {
      "epoch": 582.1946902654868,
      "grad_norm": 1.145107404454393e-07,
      "learning_rate": 0.02098474991266671,
      "loss": 0.0,
      "num_input_tokens_seen": 18849240,
      "step": 33185
    },
    {
      "epoch": 582.2831858407079,
      "grad_norm": 6.087695680889738e-08,
      "learning_rate": 0.02095471115666592,
      "loss": 0.0,
      "num_input_tokens_seen": 18851896,
      "step": 33190
    },
    {
      "epoch": 582.3716814159292,
      "grad_norm": 2.6855319745777706e-08,
      "learning_rate": 0.020924692301070406,
      "loss": 0.0,
      "num_input_tokens_seen": 18855000,
      "step": 33195
    },
    {
      "epoch": 582.4601769911504,
      "grad_norm": 3.614686505670761e-08,
      "learning_rate": 0.020894693350509346,
      "loss": 0.0,
      "num_input_tokens_seen": 18857896,
      "step": 33200
    },
    {
      "epoch": 582.4601769911504,
      "eval_loss": 0.8053354024887085,
      "eval_runtime": 0.9378,
      "eval_samples_per_second": 26.657,
      "eval_steps_per_second": 13.862,
      "num_input_tokens_seen": 18857896,
      "step": 33200
    },
    {
      "epoch": 582.5486725663717,
      "grad_norm": 3.525465430698205e-08,
      "learning_rate": 0.020864714309609057,
      "loss": 0.0,
      "num_input_tokens_seen": 18860696,
      "step": 33205
    },
    {
      "epoch": 582.637168141593,
      "grad_norm": 6.400674834594611e-08,
      "learning_rate": 0.020834755182992604,
      "loss": 0.0,
      "num_input_tokens_seen": 18863448,
      "step": 33210
    },
    {
      "epoch": 582.7256637168142,
      "grad_norm": 1.732410481736224e-08,
      "learning_rate": 0.02080481597528011,
      "loss": 0.0,
      "num_input_tokens_seen": 18866344,
      "step": 33215
    },
    {
      "epoch": 582.8141592920354,
      "grad_norm": 1.8322291239769584e-08,
      "learning_rate": 0.020774896691088583,
      "loss": 0.0,
      "num_input_tokens_seen": 18869256,
      "step": 33220
    },
    {
      "epoch": 582.9026548672566,
      "grad_norm": 4.440773437863754e-08,
      "learning_rate": 0.020744997335031882,
      "loss": 0.0,
      "num_input_tokens_seen": 18871912,
      "step": 33225
    },
    {
      "epoch": 582.9911504424779,
      "grad_norm": 3.361461864415105e-08,
      "learning_rate": 0.02071511791172092,
      "loss": 0.0,
      "num_input_tokens_seen": 18874824,
      "step": 33230
    },
    {
      "epoch": 583.070796460177,
      "grad_norm": 3.6656341961816e-08,
      "learning_rate": 0.02068525842576351,
      "loss": 0.0,
      "num_input_tokens_seen": 18877168,
      "step": 33235
    },
    {
      "epoch": 583.1592920353983,
      "grad_norm": 8.482483337957092e-08,
      "learning_rate": 0.020655418881764264,
      "loss": 0.0,
      "num_input_tokens_seen": 18879680,
      "step": 33240
    },
    {
      "epoch": 583.2477876106195,
      "grad_norm": 4.6300129952214775e-08,
      "learning_rate": 0.020625599284324923,
      "loss": 0.0,
      "num_input_tokens_seen": 18882672,
      "step": 33245
    },
    {
      "epoch": 583.3362831858407,
      "grad_norm": 4.437966794057502e-08,
      "learning_rate": 0.02059579963804396,
      "loss": 0.0,
      "num_input_tokens_seen": 18885328,
      "step": 33250
    },
    {
      "epoch": 583.4247787610619,
      "grad_norm": 3.786076163692087e-08,
      "learning_rate": 0.02056601994751688,
      "loss": 0.0,
      "num_input_tokens_seen": 18888608,
      "step": 33255
    },
    {
      "epoch": 583.5132743362832,
      "grad_norm": 2.1147542383914697e-08,
      "learning_rate": 0.02053626021733614,
      "loss": 0.0,
      "num_input_tokens_seen": 18891328,
      "step": 33260
    },
    {
      "epoch": 583.6017699115044,
      "grad_norm": 4.345436366293143e-08,
      "learning_rate": 0.02050652045209097,
      "loss": 0.0,
      "num_input_tokens_seen": 18893664,
      "step": 33265
    },
    {
      "epoch": 583.6902654867257,
      "grad_norm": 6.313324973916679e-08,
      "learning_rate": 0.020476800656367672,
      "loss": 0.0,
      "num_input_tokens_seen": 18896768,
      "step": 33270
    },
    {
      "epoch": 583.7787610619469,
      "grad_norm": 4.85277631412373e-08,
      "learning_rate": 0.020447100834749425,
      "loss": 0.0,
      "num_input_tokens_seen": 18899664,
      "step": 33275
    },
    {
      "epoch": 583.8672566371681,
      "grad_norm": 3.6376604839460924e-08,
      "learning_rate": 0.02041742099181627,
      "loss": 0.0,
      "num_input_tokens_seen": 18902896,
      "step": 33280
    },
    {
      "epoch": 583.9557522123894,
      "grad_norm": 1.9471771750545486e-08,
      "learning_rate": 0.02038776113214526,
      "loss": 0.0,
      "num_input_tokens_seen": 18906096,
      "step": 33285
    },
    {
      "epoch": 584.0353982300885,
      "grad_norm": 2.512257601949841e-08,
      "learning_rate": 0.0203581212603103,
      "loss": 0.0,
      "num_input_tokens_seen": 18908552,
      "step": 33290
    },
    {
      "epoch": 584.1238938053098,
      "grad_norm": 2.8062096646408463e-08,
      "learning_rate": 0.02032850138088219,
      "loss": 0.0,
      "num_input_tokens_seen": 18911352,
      "step": 33295
    },
    {
      "epoch": 584.212389380531,
      "grad_norm": 3.236671375361766e-08,
      "learning_rate": 0.020298901498428754,
      "loss": 0.0,
      "num_input_tokens_seen": 18914264,
      "step": 33300
    },
    {
      "epoch": 584.3008849557522,
      "grad_norm": 2.803760779102049e-08,
      "learning_rate": 0.020269321617514595,
      "loss": 0.0,
      "num_input_tokens_seen": 18916792,
      "step": 33305
    },
    {
      "epoch": 584.3893805309734,
      "grad_norm": 3.6190570540384215e-08,
      "learning_rate": 0.020239761742701343,
      "loss": 0.0,
      "num_input_tokens_seen": 18919640,
      "step": 33310
    },
    {
      "epoch": 584.4778761061947,
      "grad_norm": 3.511195245664567e-08,
      "learning_rate": 0.02021022187854754,
      "loss": 0.0,
      "num_input_tokens_seen": 18922456,
      "step": 33315
    },
    {
      "epoch": 584.566371681416,
      "grad_norm": 3.435371098703399e-08,
      "learning_rate": 0.020180702029608522,
      "loss": 0.0,
      "num_input_tokens_seen": 18925608,
      "step": 33320
    },
    {
      "epoch": 584.6548672566372,
      "grad_norm": 3.3487154382783046e-08,
      "learning_rate": 0.020151202200436695,
      "loss": 0.0,
      "num_input_tokens_seen": 18928568,
      "step": 33325
    },
    {
      "epoch": 584.7433628318585,
      "grad_norm": 2.550681266200172e-08,
      "learning_rate": 0.020121722395581226,
      "loss": 0.0,
      "num_input_tokens_seen": 18931592,
      "step": 33330
    },
    {
      "epoch": 584.8318584070796,
      "grad_norm": 6.279096709249643e-08,
      "learning_rate": 0.020092262619588342,
      "loss": 0.0,
      "num_input_tokens_seen": 18933800,
      "step": 33335
    },
    {
      "epoch": 584.9203539823009,
      "grad_norm": 1.1999500770798477e-07,
      "learning_rate": 0.02006282287700109,
      "loss": 0.0,
      "num_input_tokens_seen": 18936952,
      "step": 33340
    },
    {
      "epoch": 585.0,
      "grad_norm": 5.337918551617804e-09,
      "learning_rate": 0.020033403172359427,
      "loss": 0.0,
      "num_input_tokens_seen": 18939504,
      "step": 33345
    },
    {
      "epoch": 585.0884955752213,
      "grad_norm": 1.3576371138412924e-08,
      "learning_rate": 0.020004003510200284,
      "loss": 0.0,
      "num_input_tokens_seen": 18942224,
      "step": 33350
    },
    {
      "epoch": 585.1769911504425,
      "grad_norm": 2.6340652325984593e-08,
      "learning_rate": 0.019974623895057407,
      "loss": 0.0,
      "num_input_tokens_seen": 18944976,
      "step": 33355
    },
    {
      "epoch": 585.2654867256637,
      "grad_norm": 3.017911254232786e-08,
      "learning_rate": 0.019945264331461553,
      "loss": 0.0,
      "num_input_tokens_seen": 18947680,
      "step": 33360
    },
    {
      "epoch": 585.3539823008849,
      "grad_norm": 2.0215717100313668e-08,
      "learning_rate": 0.019915924823940317,
      "loss": 0.0,
      "num_input_tokens_seen": 18950320,
      "step": 33365
    },
    {
      "epoch": 585.4424778761062,
      "grad_norm": 3.243862778390394e-08,
      "learning_rate": 0.01988660537701816,
      "loss": 0.0,
      "num_input_tokens_seen": 18953200,
      "step": 33370
    },
    {
      "epoch": 585.5309734513274,
      "grad_norm": 3.250254820841292e-08,
      "learning_rate": 0.01985730599521659,
      "loss": 0.0,
      "num_input_tokens_seen": 18955760,
      "step": 33375
    },
    {
      "epoch": 585.6194690265487,
      "grad_norm": 8.767333525838694e-08,
      "learning_rate": 0.019828026683053918,
      "loss": 0.0,
      "num_input_tokens_seen": 18959216,
      "step": 33380
    },
    {
      "epoch": 585.70796460177,
      "grad_norm": 4.211726789549175e-08,
      "learning_rate": 0.01979876744504535,
      "loss": 0.0,
      "num_input_tokens_seen": 18962608,
      "step": 33385
    },
    {
      "epoch": 585.7964601769911,
      "grad_norm": 3.674260184993727e-08,
      "learning_rate": 0.019769528285703046,
      "loss": 0.0,
      "num_input_tokens_seen": 18966064,
      "step": 33390
    },
    {
      "epoch": 585.8849557522124,
      "grad_norm": 1.803787341714269e-08,
      "learning_rate": 0.019740309209536098,
      "loss": 0.0,
      "num_input_tokens_seen": 18968752,
      "step": 33395
    },
    {
      "epoch": 585.9734513274336,
      "grad_norm": 6.683244180294423e-08,
      "learning_rate": 0.019711110221050387,
      "loss": 0.0,
      "num_input_tokens_seen": 18971344,
      "step": 33400
    },
    {
      "epoch": 585.9734513274336,
      "eval_loss": 0.825111985206604,
      "eval_runtime": 0.9451,
      "eval_samples_per_second": 26.451,
      "eval_steps_per_second": 13.754,
      "num_input_tokens_seen": 18971344,
      "step": 33400
    },
    {
      "epoch": 586.0530973451328,
      "grad_norm": 4.59213502779221e-08,
      "learning_rate": 0.019681931324748825,
      "loss": 0.0,
      "num_input_tokens_seen": 18973736,
      "step": 33405
    },
    {
      "epoch": 586.141592920354,
      "grad_norm": 3.744888132928281e-08,
      "learning_rate": 0.019652772525131094,
      "loss": 0.0,
      "num_input_tokens_seen": 18976392,
      "step": 33410
    },
    {
      "epoch": 586.2300884955753,
      "grad_norm": 4.132190056793661e-08,
      "learning_rate": 0.019623633826693885,
      "loss": 0.0,
      "num_input_tokens_seen": 18979576,
      "step": 33415
    },
    {
      "epoch": 586.3185840707964,
      "grad_norm": 3.739424769833022e-08,
      "learning_rate": 0.019594515233930788,
      "loss": 0.0,
      "num_input_tokens_seen": 18982136,
      "step": 33420
    },
    {
      "epoch": 586.4070796460177,
      "grad_norm": 3.733569187147623e-08,
      "learning_rate": 0.019565416751332186,
      "loss": 0.0,
      "num_input_tokens_seen": 18985288,
      "step": 33425
    },
    {
      "epoch": 586.4955752212389,
      "grad_norm": 5.660006863195122e-08,
      "learning_rate": 0.019536338383385497,
      "loss": 0.0,
      "num_input_tokens_seen": 18987864,
      "step": 33430
    },
    {
      "epoch": 586.5840707964602,
      "grad_norm": 2.4174513058028424e-08,
      "learning_rate": 0.019507280134574933,
      "loss": 0.0,
      "num_input_tokens_seen": 18991144,
      "step": 33435
    },
    {
      "epoch": 586.6725663716815,
      "grad_norm": 4.439606371420268e-08,
      "learning_rate": 0.019478242009381624,
      "loss": 0.0,
      "num_input_tokens_seen": 18993976,
      "step": 33440
    },
    {
      "epoch": 586.7610619469026,
      "grad_norm": 7.264313950372525e-08,
      "learning_rate": 0.01944922401228367,
      "loss": 0.0,
      "num_input_tokens_seen": 18997032,
      "step": 33445
    },
    {
      "epoch": 586.8495575221239,
      "grad_norm": 2.5145034499018948e-08,
      "learning_rate": 0.01942022614775593,
      "loss": 0.0,
      "num_input_tokens_seen": 18999736,
      "step": 33450
    },
    {
      "epoch": 586.9380530973451,
      "grad_norm": 1.4579027762806618e-08,
      "learning_rate": 0.01939124842027029,
      "loss": 0.0,
      "num_input_tokens_seen": 19002936,
      "step": 33455
    },
    {
      "epoch": 587.0176991150443,
      "grad_norm": 2.8067978163903717e-08,
      "learning_rate": 0.01936229083429551,
      "loss": 0.0,
      "num_input_tokens_seen": 19005400,
      "step": 33460
    },
    {
      "epoch": 587.1061946902655,
      "grad_norm": 4.707524681180075e-08,
      "learning_rate": 0.019333353394297148,
      "loss": 0.0,
      "num_input_tokens_seen": 19007928,
      "step": 33465
    },
    {
      "epoch": 587.1946902654868,
      "grad_norm": 3.6698079242114545e-08,
      "learning_rate": 0.019304436104737754,
      "loss": 0.0,
      "num_input_tokens_seen": 19010600,
      "step": 33470
    },
    {
      "epoch": 587.2831858407079,
      "grad_norm": 3.985255503380358e-08,
      "learning_rate": 0.019275538970076778,
      "loss": 0.0,
      "num_input_tokens_seen": 19014456,
      "step": 33475
    },
    {
      "epoch": 587.3716814159292,
      "grad_norm": 5.961628346540238e-08,
      "learning_rate": 0.019246661994770434,
      "loss": 0.0,
      "num_input_tokens_seen": 19016920,
      "step": 33480
    },
    {
      "epoch": 587.4601769911504,
      "grad_norm": 3.4593309550245976e-08,
      "learning_rate": 0.019217805183271985,
      "loss": 0.0,
      "num_input_tokens_seen": 19019432,
      "step": 33485
    },
    {
      "epoch": 587.5486725663717,
      "grad_norm": 1.953532446918871e-08,
      "learning_rate": 0.019188968540031465,
      "loss": 0.0,
      "num_input_tokens_seen": 19022376,
      "step": 33490
    },
    {
      "epoch": 587.637168141593,
      "grad_norm": 2.6066350855558085e-08,
      "learning_rate": 0.019160152069495867,
      "loss": 0.0,
      "num_input_tokens_seen": 19025416,
      "step": 33495
    },
    {
      "epoch": 587.7256637168142,
      "grad_norm": 3.321051167404221e-08,
      "learning_rate": 0.019131355776109103,
      "loss": 0.0,
      "num_input_tokens_seen": 19028360,
      "step": 33500
    },
    {
      "epoch": 587.8141592920354,
      "grad_norm": 4.229014294310218e-08,
      "learning_rate": 0.019102579664311857,
      "loss": 0.0,
      "num_input_tokens_seen": 19031432,
      "step": 33505
    },
    {
      "epoch": 587.9026548672566,
      "grad_norm": 2.823850842048614e-08,
      "learning_rate": 0.019073823738541763,
      "loss": 0.0,
      "num_input_tokens_seen": 19034456,
      "step": 33510
    },
    {
      "epoch": 587.9911504424779,
      "grad_norm": 5.61822659506106e-08,
      "learning_rate": 0.0190450880032334,
      "loss": 0.0,
      "num_input_tokens_seen": 19037032,
      "step": 33515
    },
    {
      "epoch": 588.070796460177,
      "grad_norm": 1.7747391112266087e-08,
      "learning_rate": 0.019016372462818114,
      "loss": 0.0,
      "num_input_tokens_seen": 19039232,
      "step": 33520
    },
    {
      "epoch": 588.1592920353983,
      "grad_norm": 2.1038021102981475e-08,
      "learning_rate": 0.018987677121724278,
      "loss": 0.0,
      "num_input_tokens_seen": 19042032,
      "step": 33525
    },
    {
      "epoch": 588.2477876106195,
      "grad_norm": 2.1408277817158705e-08,
      "learning_rate": 0.018959001984377,
      "loss": 0.0,
      "num_input_tokens_seen": 19045184,
      "step": 33530
    },
    {
      "epoch": 588.3362831858407,
      "grad_norm": 3.2540821592874636e-08,
      "learning_rate": 0.018930347055198377,
      "loss": 0.0,
      "num_input_tokens_seen": 19048288,
      "step": 33535
    },
    {
      "epoch": 588.4247787610619,
      "grad_norm": 4.6715939561181585e-08,
      "learning_rate": 0.01890171233860739,
      "loss": 0.0,
      "num_input_tokens_seen": 19051520,
      "step": 33540
    },
    {
      "epoch": 588.5132743362832,
      "grad_norm": 4.914562623525853e-08,
      "learning_rate": 0.018873097839019807,
      "loss": 0.0,
      "num_input_tokens_seen": 19054576,
      "step": 33545
    },
    {
      "epoch": 588.6017699115044,
      "grad_norm": 3.270942627864315e-08,
      "learning_rate": 0.0188445035608484,
      "loss": 0.0,
      "num_input_tokens_seen": 19057552,
      "step": 33550
    },
    {
      "epoch": 588.6902654867257,
      "grad_norm": 4.055561575455613e-08,
      "learning_rate": 0.018815929508502777,
      "loss": 0.0,
      "num_input_tokens_seen": 19060400,
      "step": 33555
    },
    {
      "epoch": 588.7787610619469,
      "grad_norm": 6.182366973916942e-08,
      "learning_rate": 0.01878737568638934,
      "loss": 0.0,
      "num_input_tokens_seen": 19062864,
      "step": 33560
    },
    {
      "epoch": 588.8672566371681,
      "grad_norm": 4.15416394616841e-08,
      "learning_rate": 0.01875884209891152,
      "loss": 0.0,
      "num_input_tokens_seen": 19065840,
      "step": 33565
    },
    {
      "epoch": 588.9557522123894,
      "grad_norm": 2.043438307453016e-08,
      "learning_rate": 0.018730328750469514,
      "loss": 0.0,
      "num_input_tokens_seen": 19068432,
      "step": 33570
    },
    {
      "epoch": 589.0353982300885,
      "grad_norm": 7.736015561476961e-08,
      "learning_rate": 0.018701835645460473,
      "loss": 0.0,
      "num_input_tokens_seen": 19070672,
      "step": 33575
    },
    {
      "epoch": 589.1238938053098,
      "grad_norm": 1.1331991345286951e-07,
      "learning_rate": 0.01867336278827838,
      "loss": 0.0,
      "num_input_tokens_seen": 19073664,
      "step": 33580
    },
    {
      "epoch": 589.212389380531,
      "grad_norm": 3.2448326692247065e-08,
      "learning_rate": 0.018644910183314056,
      "loss": 0.0,
      "num_input_tokens_seen": 19076400,
      "step": 33585
    },
    {
      "epoch": 589.3008849557522,
      "grad_norm": 1.7112533612362313e-08,
      "learning_rate": 0.01861647783495531,
      "loss": 0.0,
      "num_input_tokens_seen": 19079264,
      "step": 33590
    },
    {
      "epoch": 589.3893805309734,
      "grad_norm": 4.081685034407201e-08,
      "learning_rate": 0.01858806574758676,
      "loss": 0.0,
      "num_input_tokens_seen": 19081904,
      "step": 33595
    },
    {
      "epoch": 589.4778761061947,
      "grad_norm": 2.3020753303626407e-08,
      "learning_rate": 0.01855967392558988,
      "loss": 0.0,
      "num_input_tokens_seen": 19085248,
      "step": 33600
    },
    {
      "epoch": 589.4778761061947,
      "eval_loss": 0.8042502999305725,
      "eval_runtime": 0.9229,
      "eval_samples_per_second": 27.087,
      "eval_steps_per_second": 14.085,
      "num_input_tokens_seen": 19085248,
      "step": 33600
    },
    {
      "epoch": 589.566371681416,
      "grad_norm": 6.197748803060676e-08,
      "learning_rate": 0.018531302373343096,
      "loss": 0.0,
      "num_input_tokens_seen": 19087968,
      "step": 33605
    },
    {
      "epoch": 589.6548672566372,
      "grad_norm": 1.650734837710388e-08,
      "learning_rate": 0.018502951095221588,
      "loss": 0.0,
      "num_input_tokens_seen": 19090672,
      "step": 33610
    },
    {
      "epoch": 589.7433628318585,
      "grad_norm": 4.1191679400753856e-08,
      "learning_rate": 0.01847462009559751,
      "loss": 0.0,
      "num_input_tokens_seen": 19093360,
      "step": 33615
    },
    {
      "epoch": 589.8318584070796,
      "grad_norm": 5.1916622112457844e-08,
      "learning_rate": 0.01844630937883992,
      "loss": 0.0,
      "num_input_tokens_seen": 19096592,
      "step": 33620
    },
    {
      "epoch": 589.9203539823009,
      "grad_norm": 8.469802992294717e-08,
      "learning_rate": 0.018418018949314573,
      "loss": 0.0,
      "num_input_tokens_seen": 19099872,
      "step": 33625
    },
    {
      "epoch": 590.0,
      "grad_norm": 3.449208918482327e-08,
      "learning_rate": 0.018389748811384315,
      "loss": 0.0,
      "num_input_tokens_seen": 19101944,
      "step": 33630
    },
    {
      "epoch": 590.0884955752213,
      "grad_norm": 3.6248295032237365e-08,
      "learning_rate": 0.018361498969408658,
      "loss": 0.0,
      "num_input_tokens_seen": 19105160,
      "step": 33635
    },
    {
      "epoch": 590.1769911504425,
      "grad_norm": 3.4686344463352725e-08,
      "learning_rate": 0.01833326942774415,
      "loss": 0.0,
      "num_input_tokens_seen": 19108088,
      "step": 33640
    },
    {
      "epoch": 590.2654867256637,
      "grad_norm": 2.3227768153333272e-08,
      "learning_rate": 0.018305060190744155,
      "loss": 0.0,
      "num_input_tokens_seen": 19111032,
      "step": 33645
    },
    {
      "epoch": 590.3539823008849,
      "grad_norm": 4.0627924136060756e-08,
      "learning_rate": 0.018276871262758846,
      "loss": 0.0,
      "num_input_tokens_seen": 19113704,
      "step": 33650
    },
    {
      "epoch": 590.4424778761062,
      "grad_norm": 4.1117335314311276e-08,
      "learning_rate": 0.0182487026481353,
      "loss": 0.0,
      "num_input_tokens_seen": 19116456,
      "step": 33655
    },
    {
      "epoch": 590.5309734513274,
      "grad_norm": 4.2765158525526203e-08,
      "learning_rate": 0.018220554351217538,
      "loss": 0.0,
      "num_input_tokens_seen": 19119416,
      "step": 33660
    },
    {
      "epoch": 590.6194690265487,
      "grad_norm": 6.612514624748655e-08,
      "learning_rate": 0.01819242637634629,
      "loss": 0.0,
      "num_input_tokens_seen": 19122520,
      "step": 33665
    },
    {
      "epoch": 590.70796460177,
      "grad_norm": 3.485343214038039e-08,
      "learning_rate": 0.01816431872785933,
      "loss": 0.0,
      "num_input_tokens_seen": 19125240,
      "step": 33670
    },
    {
      "epoch": 590.7964601769911,
      "grad_norm": 2.7006162639509057e-08,
      "learning_rate": 0.018136231410091148,
      "loss": 0.0,
      "num_input_tokens_seen": 19127912,
      "step": 33675
    },
    {
      "epoch": 590.8849557522124,
      "grad_norm": 3.821957861305236e-08,
      "learning_rate": 0.018108164427373175,
      "loss": 0.0,
      "num_input_tokens_seen": 19130696,
      "step": 33680
    },
    {
      "epoch": 590.9734513274336,
      "grad_norm": 3.4309096008655615e-08,
      "learning_rate": 0.01808011778403375,
      "loss": 0.0,
      "num_input_tokens_seen": 19133816,
      "step": 33685
    },
    {
      "epoch": 591.0530973451328,
      "grad_norm": 2.23403269217215e-08,
      "learning_rate": 0.01805209148439793,
      "loss": 0.0,
      "num_input_tokens_seen": 19136248,
      "step": 33690
    },
    {
      "epoch": 591.141592920354,
      "grad_norm": 5.6722488039895325e-08,
      "learning_rate": 0.018024085532787757,
      "loss": 0.0,
      "num_input_tokens_seen": 19138872,
      "step": 33695
    },
    {
      "epoch": 591.2300884955753,
      "grad_norm": 1.87525959205459e-08,
      "learning_rate": 0.017996099933522164,
      "loss": 0.0,
      "num_input_tokens_seen": 19141784,
      "step": 33700
    },
    {
      "epoch": 591.3185840707964,
      "grad_norm": 5.397211921831513e-08,
      "learning_rate": 0.017968134690916775,
      "loss": 0.0,
      "num_input_tokens_seen": 19144808,
      "step": 33705
    },
    {
      "epoch": 591.4070796460177,
      "grad_norm": 3.013817817532072e-08,
      "learning_rate": 0.017940189809284263,
      "loss": 0.0,
      "num_input_tokens_seen": 19147336,
      "step": 33710
    },
    {
      "epoch": 591.4955752212389,
      "grad_norm": 2.5857371355186842e-08,
      "learning_rate": 0.017912265292934024,
      "loss": 0.0,
      "num_input_tokens_seen": 19149992,
      "step": 33715
    },
    {
      "epoch": 591.5840707964602,
      "grad_norm": 2.49486244996433e-08,
      "learning_rate": 0.017884361146172423,
      "loss": 0.0,
      "num_input_tokens_seen": 19153048,
      "step": 33720
    },
    {
      "epoch": 591.6725663716815,
      "grad_norm": 1.9978255494379482e-08,
      "learning_rate": 0.01785647737330261,
      "loss": 0.0,
      "num_input_tokens_seen": 19156120,
      "step": 33725
    },
    {
      "epoch": 591.7610619469026,
      "grad_norm": 2.540214616431058e-08,
      "learning_rate": 0.017828613978624563,
      "loss": 0.0,
      "num_input_tokens_seen": 19159192,
      "step": 33730
    },
    {
      "epoch": 591.8495575221239,
      "grad_norm": 5.505745193090661e-08,
      "learning_rate": 0.01780077096643523,
      "loss": 0.0,
      "num_input_tokens_seen": 19162104,
      "step": 33735
    },
    {
      "epoch": 591.9380530973451,
      "grad_norm": 4.223941019176891e-08,
      "learning_rate": 0.017772948341028345,
      "loss": 0.0,
      "num_input_tokens_seen": 19165496,
      "step": 33740
    },
    {
      "epoch": 592.0176991150443,
      "grad_norm": 7.353830966394526e-08,
      "learning_rate": 0.01774514610669447,
      "loss": 0.0,
      "num_input_tokens_seen": 19168032,
      "step": 33745
    },
    {
      "epoch": 592.1061946902655,
      "grad_norm": 5.3071136818516607e-08,
      "learning_rate": 0.017717364267721112,
      "loss": 0.0,
      "num_input_tokens_seen": 19170752,
      "step": 33750
    },
    {
      "epoch": 592.1946902654868,
      "grad_norm": 8.303129561681999e-08,
      "learning_rate": 0.017689602828392513,
      "loss": 0.0,
      "num_input_tokens_seen": 19173712,
      "step": 33755
    },
    {
      "epoch": 592.2831858407079,
      "grad_norm": 3.943177873111381e-08,
      "learning_rate": 0.017661861792989897,
      "loss": 0.0,
      "num_input_tokens_seen": 19176624,
      "step": 33760
    },
    {
      "epoch": 592.3716814159292,
      "grad_norm": 2.8203135826743164e-08,
      "learning_rate": 0.017634141165791272,
      "loss": 0.0,
      "num_input_tokens_seen": 19179296,
      "step": 33765
    },
    {
      "epoch": 592.4601769911504,
      "grad_norm": 3.4449129771019216e-08,
      "learning_rate": 0.017606440951071455,
      "loss": 0.0,
      "num_input_tokens_seen": 19182064,
      "step": 33770
    },
    {
      "epoch": 592.5486725663717,
      "grad_norm": 3.5270122822339545e-08,
      "learning_rate": 0.017578761153102213,
      "loss": 0.0,
      "num_input_tokens_seen": 19184848,
      "step": 33775
    },
    {
      "epoch": 592.637168141593,
      "grad_norm": 3.145044047414558e-08,
      "learning_rate": 0.017551101776152146,
      "loss": 0.0,
      "num_input_tokens_seen": 19187488,
      "step": 33780
    },
    {
      "epoch": 592.7256637168142,
      "grad_norm": 5.000208247452065e-08,
      "learning_rate": 0.017523462824486608,
      "loss": 0.0,
      "num_input_tokens_seen": 19190832,
      "step": 33785
    },
    {
      "epoch": 592.8141592920354,
      "grad_norm": 5.5709094226585876e-08,
      "learning_rate": 0.01749584430236794,
      "loss": 0.0,
      "num_input_tokens_seen": 19193536,
      "step": 33790
    },
    {
      "epoch": 592.9026548672566,
      "grad_norm": 6.691288945148699e-08,
      "learning_rate": 0.01746824621405524,
      "loss": 0.0,
      "num_input_tokens_seen": 19196176,
      "step": 33795
    },
    {
      "epoch": 592.9911504424779,
      "grad_norm": 2.3795967862838552e-08,
      "learning_rate": 0.017440668563804412,
      "loss": 0.0,
      "num_input_tokens_seen": 19199136,
      "step": 33800
    },
    {
      "epoch": 592.9911504424779,
      "eval_loss": 0.801821768283844,
      "eval_runtime": 0.931,
      "eval_samples_per_second": 26.852,
      "eval_steps_per_second": 13.963,
      "num_input_tokens_seen": 19199136,
      "step": 33800
    },
    {
      "epoch": 593.070796460177,
      "grad_norm": 1.3524123687602696e-07,
      "learning_rate": 0.017413111355868392,
      "loss": 0.0,
      "num_input_tokens_seen": 19201480,
      "step": 33805
    },
    {
      "epoch": 593.1592920353983,
      "grad_norm": 3.6091588384579154e-08,
      "learning_rate": 0.017385574594496748,
      "loss": 0.0,
      "num_input_tokens_seen": 19204632,
      "step": 33810
    },
    {
      "epoch": 593.2477876106195,
      "grad_norm": 5.1029079628506224e-08,
      "learning_rate": 0.01735805828393605,
      "loss": 0.0,
      "num_input_tokens_seen": 19207288,
      "step": 33815
    },
    {
      "epoch": 593.3362831858407,
      "grad_norm": 4.808816811419092e-08,
      "learning_rate": 0.017330562428429667,
      "loss": 0.0,
      "num_input_tokens_seen": 19210072,
      "step": 33820
    },
    {
      "epoch": 593.4247787610619,
      "grad_norm": 7.892942477383258e-08,
      "learning_rate": 0.01730308703221776,
      "loss": 0.0,
      "num_input_tokens_seen": 19212712,
      "step": 33825
    },
    {
      "epoch": 593.5132743362832,
      "grad_norm": 4.0078155905121093e-08,
      "learning_rate": 0.01727563209953744,
      "loss": 0.0,
      "num_input_tokens_seen": 19215864,
      "step": 33830
    },
    {
      "epoch": 593.6017699115044,
      "grad_norm": 2.3389475245494395e-08,
      "learning_rate": 0.017248197634622535,
      "loss": 0.0,
      "num_input_tokens_seen": 19218776,
      "step": 33835
    },
    {
      "epoch": 593.6902654867257,
      "grad_norm": 4.139816667247942e-08,
      "learning_rate": 0.01722078364170383,
      "loss": 0.0,
      "num_input_tokens_seen": 19221544,
      "step": 33840
    },
    {
      "epoch": 593.7787610619469,
      "grad_norm": 5.1973039205677196e-08,
      "learning_rate": 0.017193390125008905,
      "loss": 0.0,
      "num_input_tokens_seen": 19224920,
      "step": 33845
    },
    {
      "epoch": 593.8672566371681,
      "grad_norm": 4.039899081931253e-08,
      "learning_rate": 0.017166017088762153,
      "loss": 0.0,
      "num_input_tokens_seen": 19227400,
      "step": 33850
    },
    {
      "epoch": 593.9557522123894,
      "grad_norm": 2.9167159354415162e-08,
      "learning_rate": 0.017138664537184878,
      "loss": 0.0,
      "num_input_tokens_seen": 19230360,
      "step": 33855
    },
    {
      "epoch": 594.0353982300885,
      "grad_norm": 3.147283322846306e-08,
      "learning_rate": 0.017111332474495172,
      "loss": 0.0,
      "num_input_tokens_seen": 19232576,
      "step": 33860
    },
    {
      "epoch": 594.1238938053098,
      "grad_norm": 2.9758442821048448e-08,
      "learning_rate": 0.017084020904907998,
      "loss": 0.0,
      "num_input_tokens_seen": 19235184,
      "step": 33865
    },
    {
      "epoch": 594.212389380531,
      "grad_norm": 5.626391441637679e-08,
      "learning_rate": 0.017056729832635103,
      "loss": 0.0,
      "num_input_tokens_seen": 19237568,
      "step": 33870
    },
    {
      "epoch": 594.3008849557522,
      "grad_norm": 3.172694817976662e-08,
      "learning_rate": 0.017029459261885153,
      "loss": 0.0,
      "num_input_tokens_seen": 19240160,
      "step": 33875
    },
    {
      "epoch": 594.3893805309734,
      "grad_norm": 2.9742821539002762e-08,
      "learning_rate": 0.01700220919686359,
      "loss": 0.0,
      "num_input_tokens_seen": 19242960,
      "step": 33880
    },
    {
      "epoch": 594.4778761061947,
      "grad_norm": 4.1837211028905585e-08,
      "learning_rate": 0.016974979641772723,
      "loss": 0.0,
      "num_input_tokens_seen": 19245824,
      "step": 33885
    },
    {
      "epoch": 594.566371681416,
      "grad_norm": 2.6871381564319563e-08,
      "learning_rate": 0.01694777060081169,
      "loss": 0.0,
      "num_input_tokens_seen": 19248544,
      "step": 33890
    },
    {
      "epoch": 594.6548672566372,
      "grad_norm": 3.3365754603664755e-08,
      "learning_rate": 0.016920582078176444,
      "loss": 0.0,
      "num_input_tokens_seen": 19252112,
      "step": 33895
    },
    {
      "epoch": 594.7433628318585,
      "grad_norm": 6.217400283503594e-08,
      "learning_rate": 0.016893414078059863,
      "loss": 0.0,
      "num_input_tokens_seen": 19255504,
      "step": 33900
    },
    {
      "epoch": 594.8318584070796,
      "grad_norm": 5.302649697114248e-08,
      "learning_rate": 0.016866266604651535,
      "loss": 0.0,
      "num_input_tokens_seen": 19258432,
      "step": 33905
    },
    {
      "epoch": 594.9203539823009,
      "grad_norm": 3.104529966435621e-08,
      "learning_rate": 0.016839139662137976,
      "loss": 0.0,
      "num_input_tokens_seen": 19262000,
      "step": 33910
    },
    {
      "epoch": 595.0,
      "grad_norm": 2.0906432141032383e-08,
      "learning_rate": 0.01681203325470245,
      "loss": 0.0,
      "num_input_tokens_seen": 19264208,
      "step": 33915
    },
    {
      "epoch": 595.0884955752213,
      "grad_norm": 4.591467117620596e-08,
      "learning_rate": 0.016784947386525157,
      "loss": 0.0,
      "num_input_tokens_seen": 19266768,
      "step": 33920
    },
    {
      "epoch": 595.1769911504425,
      "grad_norm": 3.779815571647305e-08,
      "learning_rate": 0.01675788206178308,
      "loss": 0.0,
      "num_input_tokens_seen": 19269536,
      "step": 33925
    },
    {
      "epoch": 595.2654867256637,
      "grad_norm": 4.8437183153282604e-08,
      "learning_rate": 0.016730837284649986,
      "loss": 0.0,
      "num_input_tokens_seen": 19272336,
      "step": 33930
    },
    {
      "epoch": 595.3539823008849,
      "grad_norm": 2.2709031100021093e-08,
      "learning_rate": 0.016703813059296583,
      "loss": 0.0,
      "num_input_tokens_seen": 19275168,
      "step": 33935
    },
    {
      "epoch": 595.4424778761062,
      "grad_norm": 8.979455401458836e-08,
      "learning_rate": 0.016676809389890294,
      "loss": 0.0,
      "num_input_tokens_seen": 19278048,
      "step": 33940
    },
    {
      "epoch": 595.5309734513274,
      "grad_norm": 2.9489573449836826e-08,
      "learning_rate": 0.016649826280595435,
      "loss": 0.0,
      "num_input_tokens_seen": 19280864,
      "step": 33945
    },
    {
      "epoch": 595.6194690265487,
      "grad_norm": 5.9400342422577523e-08,
      "learning_rate": 0.016622863735573163,
      "loss": 0.0,
      "num_input_tokens_seen": 19283920,
      "step": 33950
    },
    {
      "epoch": 595.70796460177,
      "grad_norm": 4.8348411496590415e-08,
      "learning_rate": 0.016595921758981395,
      "loss": 0.0,
      "num_input_tokens_seen": 19287024,
      "step": 33955
    },
    {
      "epoch": 595.7964601769911,
      "grad_norm": 1.4757672417431422e-08,
      "learning_rate": 0.01656900035497495,
      "loss": 0.0,
      "num_input_tokens_seen": 19289616,
      "step": 33960
    },
    {
      "epoch": 595.8849557522124,
      "grad_norm": 4.382055962537379e-08,
      "learning_rate": 0.016542099527705485,
      "loss": 0.0,
      "num_input_tokens_seen": 19292528,
      "step": 33965
    },
    {
      "epoch": 595.9734513274336,
      "grad_norm": 2.105832308529898e-08,
      "learning_rate": 0.01651521928132138,
      "loss": 0.0,
      "num_input_tokens_seen": 19295648,
      "step": 33970
    },
    {
      "epoch": 596.0530973451328,
      "grad_norm": 2.88289836447575e-08,
      "learning_rate": 0.01648835961996794,
      "loss": 0.0,
      "num_input_tokens_seen": 19297952,
      "step": 33975
    },
    {
      "epoch": 596.141592920354,
      "grad_norm": 2.4784018393120277e-08,
      "learning_rate": 0.016461520547787285,
      "loss": 0.0,
      "num_input_tokens_seen": 19300784,
      "step": 33980
    },
    {
      "epoch": 596.2300884955753,
      "grad_norm": 2.8511818683796264e-08,
      "learning_rate": 0.016434702068918266,
      "loss": 0.0,
      "num_input_tokens_seen": 19303312,
      "step": 33985
    },
    {
      "epoch": 596.3185840707964,
      "grad_norm": 1.148695716324255e-07,
      "learning_rate": 0.01640790418749673,
      "loss": 0.0,
      "num_input_tokens_seen": 19306544,
      "step": 33990
    },
    {
      "epoch": 596.4070796460177,
      "grad_norm": 4.711025525239165e-08,
      "learning_rate": 0.016381126907655134,
      "loss": 0.0,
      "num_input_tokens_seen": 19308912,
      "step": 33995
    },
    {
      "epoch": 596.4955752212389,
      "grad_norm": 2.217750605382207e-08,
      "learning_rate": 0.016354370233522948,
      "loss": 0.0,
      "num_input_tokens_seen": 19311344,
      "step": 34000
    },
    {
      "epoch": 596.4955752212389,
      "eval_loss": 0.8218154311180115,
      "eval_runtime": 0.9422,
      "eval_samples_per_second": 26.533,
      "eval_steps_per_second": 13.797,
      "num_input_tokens_seen": 19311344,
      "step": 34000
    },
    {
      "epoch": 596.5840707964602,
      "grad_norm": 2.166191848118615e-08,
      "learning_rate": 0.016327634169226394,
      "loss": 0.0,
      "num_input_tokens_seen": 19313920,
      "step": 34005
    },
    {
      "epoch": 596.6725663716815,
      "grad_norm": 4.9452729911081406e-08,
      "learning_rate": 0.016300918718888485,
      "loss": 0.0,
      "num_input_tokens_seen": 19317408,
      "step": 34010
    },
    {
      "epoch": 596.7610619469026,
      "grad_norm": 3.5323111546858854e-08,
      "learning_rate": 0.016274223886629052,
      "loss": 0.0,
      "num_input_tokens_seen": 19320352,
      "step": 34015
    },
    {
      "epoch": 596.8495575221239,
      "grad_norm": 3.6161985406124586e-08,
      "learning_rate": 0.01624754967656482,
      "loss": 0.0,
      "num_input_tokens_seen": 19323360,
      "step": 34020
    },
    {
      "epoch": 596.9380530973451,
      "grad_norm": 6.497041482589339e-08,
      "learning_rate": 0.016220896092809235,
      "loss": 0.0,
      "num_input_tokens_seen": 19326640,
      "step": 34025
    },
    {
      "epoch": 597.0176991150443,
      "grad_norm": 7.145755631654538e-08,
      "learning_rate": 0.01619426313947267,
      "loss": 0.0,
      "num_input_tokens_seen": 19328992,
      "step": 34030
    },
    {
      "epoch": 597.1061946902655,
      "grad_norm": 7.355523479191106e-08,
      "learning_rate": 0.016167650820662228,
      "loss": 0.0,
      "num_input_tokens_seen": 19331680,
      "step": 34035
    },
    {
      "epoch": 597.1946902654868,
      "grad_norm": 4.4806657939489014e-08,
      "learning_rate": 0.016141059140481855,
      "loss": 0.0,
      "num_input_tokens_seen": 19334352,
      "step": 34040
    },
    {
      "epoch": 597.2831858407079,
      "grad_norm": 5.163122551721244e-08,
      "learning_rate": 0.016114488103032374,
      "loss": 0.0,
      "num_input_tokens_seen": 19337600,
      "step": 34045
    },
    {
      "epoch": 597.3716814159292,
      "grad_norm": 3.1968454550224124e-08,
      "learning_rate": 0.016087937712411293,
      "loss": 0.0,
      "num_input_tokens_seen": 19340288,
      "step": 34050
    },
    {
      "epoch": 597.4601769911504,
      "grad_norm": 4.618501492359428e-08,
      "learning_rate": 0.01606140797271308,
      "loss": 0.0,
      "num_input_tokens_seen": 19343552,
      "step": 34055
    },
    {
      "epoch": 597.5486725663717,
      "grad_norm": 5.4706518426428374e-08,
      "learning_rate": 0.01603489888802897,
      "loss": 0.0,
      "num_input_tokens_seen": 19346272,
      "step": 34060
    },
    {
      "epoch": 597.637168141593,
      "grad_norm": 3.3993991621628084e-08,
      "learning_rate": 0.016008410462446918,
      "loss": 0.0,
      "num_input_tokens_seen": 19349104,
      "step": 34065
    },
    {
      "epoch": 597.7256637168142,
      "grad_norm": 4.23997406073795e-08,
      "learning_rate": 0.01598194270005185,
      "loss": 0.0,
      "num_input_tokens_seen": 19352032,
      "step": 34070
    },
    {
      "epoch": 597.8141592920354,
      "grad_norm": 3.4767968060123167e-08,
      "learning_rate": 0.015955495604925356,
      "loss": 0.0,
      "num_input_tokens_seen": 19354688,
      "step": 34075
    },
    {
      "epoch": 597.9026548672566,
      "grad_norm": 3.18040740410197e-08,
      "learning_rate": 0.01592906918114598,
      "loss": 0.0,
      "num_input_tokens_seen": 19357392,
      "step": 34080
    },
    {
      "epoch": 597.9911504424779,
      "grad_norm": 4.059582892068647e-08,
      "learning_rate": 0.015902663432788965,
      "loss": 0.0,
      "num_input_tokens_seen": 19360480,
      "step": 34085
    },
    {
      "epoch": 598.070796460177,
      "grad_norm": 3.594236019921482e-08,
      "learning_rate": 0.01587627836392643,
      "loss": 0.0,
      "num_input_tokens_seen": 19362624,
      "step": 34090
    },
    {
      "epoch": 598.1592920353983,
      "grad_norm": 2.1414141571085565e-08,
      "learning_rate": 0.01584991397862726,
      "loss": 0.0,
      "num_input_tokens_seen": 19365200,
      "step": 34095
    },
    {
      "epoch": 598.2477876106195,
      "grad_norm": 2.5895806388120945e-08,
      "learning_rate": 0.015823570280957214,
      "loss": 0.0,
      "num_input_tokens_seen": 19368704,
      "step": 34100
    },
    {
      "epoch": 598.3362831858407,
      "grad_norm": 3.2858682885716917e-08,
      "learning_rate": 0.015797247274978766,
      "loss": 0.0,
      "num_input_tokens_seen": 19371456,
      "step": 34105
    },
    {
      "epoch": 598.4247787610619,
      "grad_norm": 5.6344998000668056e-08,
      "learning_rate": 0.015770944964751326,
      "loss": 0.0,
      "num_input_tokens_seen": 19374720,
      "step": 34110
    },
    {
      "epoch": 598.5132743362832,
      "grad_norm": 6.042516531579167e-08,
      "learning_rate": 0.015744663354330956,
      "loss": 0.0,
      "num_input_tokens_seen": 19377728,
      "step": 34115
    },
    {
      "epoch": 598.6017699115044,
      "grad_norm": 4.385353236102674e-08,
      "learning_rate": 0.015718402447770664,
      "loss": 0.0,
      "num_input_tokens_seen": 19380768,
      "step": 34120
    },
    {
      "epoch": 598.6902654867257,
      "grad_norm": 2.2873789973232306e-08,
      "learning_rate": 0.015692162249120224,
      "loss": 0.0,
      "num_input_tokens_seen": 19383360,
      "step": 34125
    },
    {
      "epoch": 598.7787610619469,
      "grad_norm": 4.501018580072014e-08,
      "learning_rate": 0.01566594276242615,
      "loss": 0.0,
      "num_input_tokens_seen": 19386320,
      "step": 34130
    },
    {
      "epoch": 598.8672566371681,
      "grad_norm": 3.3305653346360486e-08,
      "learning_rate": 0.015639743991731857,
      "loss": 0.0,
      "num_input_tokens_seen": 19389280,
      "step": 34135
    },
    {
      "epoch": 598.9557522123894,
      "grad_norm": 2.895045092543569e-08,
      "learning_rate": 0.01561356594107755,
      "loss": 0.0,
      "num_input_tokens_seen": 19392048,
      "step": 34140
    },
    {
      "epoch": 599.0353982300885,
      "grad_norm": 3.2208532729782746e-08,
      "learning_rate": 0.015587408614500147,
      "loss": 0.0,
      "num_input_tokens_seen": 19394200,
      "step": 34145
    },
    {
      "epoch": 599.1238938053098,
      "grad_norm": 2.4105583307232337e-08,
      "learning_rate": 0.015561272016033505,
      "loss": 0.0,
      "num_input_tokens_seen": 19396904,
      "step": 34150
    },
    {
      "epoch": 599.212389380531,
      "grad_norm": 3.321252606269809e-08,
      "learning_rate": 0.015535156149708167,
      "loss": 0.0,
      "num_input_tokens_seen": 19399688,
      "step": 34155
    },
    {
      "epoch": 599.3008849557522,
      "grad_norm": 5.33317994211302e-08,
      "learning_rate": 0.015509061019551528,
      "loss": 0.0,
      "num_input_tokens_seen": 19402568,
      "step": 34160
    },
    {
      "epoch": 599.3893805309734,
      "grad_norm": 5.3897903029564986e-08,
      "learning_rate": 0.015482986629587818,
      "loss": 0.0,
      "num_input_tokens_seen": 19405336,
      "step": 34165
    },
    {
      "epoch": 599.4778761061947,
      "grad_norm": 9.562670300056197e-08,
      "learning_rate": 0.01545693298383799,
      "loss": 0.0,
      "num_input_tokens_seen": 19408136,
      "step": 34170
    },
    {
      "epoch": 599.566371681416,
      "grad_norm": 3.628835187896584e-08,
      "learning_rate": 0.015430900086319858,
      "loss": 0.0,
      "num_input_tokens_seen": 19410840,
      "step": 34175
    },
    {
      "epoch": 599.6548672566372,
      "grad_norm": 3.3337084204276834e-08,
      "learning_rate": 0.015404887941048084,
      "loss": 0.0,
      "num_input_tokens_seen": 19413720,
      "step": 34180
    },
    {
      "epoch": 599.7433628318585,
      "grad_norm": 3.048496566293579e-08,
      "learning_rate": 0.01537889655203397,
      "loss": 0.0,
      "num_input_tokens_seen": 19416952,
      "step": 34185
    },
    {
      "epoch": 599.8318584070796,
      "grad_norm": 5.79546757251137e-08,
      "learning_rate": 0.015352925923285798,
      "loss": 0.0,
      "num_input_tokens_seen": 19419912,
      "step": 34190
    },
    {
      "epoch": 599.9203539823009,
      "grad_norm": 8.062872325353965e-08,
      "learning_rate": 0.015326976058808511,
      "loss": 0.0,
      "num_input_tokens_seen": 19423032,
      "step": 34195
    },
    {
      "epoch": 600.0,
      "grad_norm": 2.4700595346871523e-08,
      "learning_rate": 0.015301046962603908,
      "loss": 0.0,
      "num_input_tokens_seen": 19425472,
      "step": 34200
    },
    {
      "epoch": 600.0,
      "eval_loss": 0.8327299356460571,
      "eval_runtime": 0.9146,
      "eval_samples_per_second": 27.336,
      "eval_steps_per_second": 14.215,
      "num_input_tokens_seen": 19425472,
      "step": 34200
    },
    {
      "epoch": 600.0884955752213,
      "grad_norm": 4.320324364925909e-08,
      "learning_rate": 0.015275138638670626,
      "loss": 0.0,
      "num_input_tokens_seen": 19428848,
      "step": 34205
    },
    {
      "epoch": 600.1769911504425,
      "grad_norm": 3.2690866902385096e-08,
      "learning_rate": 0.015249251091004001,
      "loss": 0.0,
      "num_input_tokens_seen": 19432096,
      "step": 34210
    },
    {
      "epoch": 600.2654867256637,
      "grad_norm": 2.863466086466815e-08,
      "learning_rate": 0.01522338432359624,
      "loss": 0.0,
      "num_input_tokens_seen": 19434656,
      "step": 34215
    },
    {
      "epoch": 600.3539823008849,
      "grad_norm": 1.4271901704887568e-07,
      "learning_rate": 0.01519753834043635,
      "loss": 0.0,
      "num_input_tokens_seen": 19437280,
      "step": 34220
    },
    {
      "epoch": 600.4424778761062,
      "grad_norm": 2.7366452215460413e-08,
      "learning_rate": 0.015171713145510095,
      "loss": 0.0,
      "num_input_tokens_seen": 19440832,
      "step": 34225
    },
    {
      "epoch": 600.5309734513274,
      "grad_norm": 2.7776865252349126e-08,
      "learning_rate": 0.01514590874279999,
      "loss": 0.0,
      "num_input_tokens_seen": 19443712,
      "step": 34230
    },
    {
      "epoch": 600.6194690265487,
      "grad_norm": 1.9850736165949456e-08,
      "learning_rate": 0.015120125136285467,
      "loss": 0.0,
      "num_input_tokens_seen": 19446624,
      "step": 34235
    },
    {
      "epoch": 600.70796460177,
      "grad_norm": 1.6413535419701475e-08,
      "learning_rate": 0.015094362329942629,
      "loss": 0.0,
      "num_input_tokens_seen": 19449424,
      "step": 34240
    },
    {
      "epoch": 600.7964601769911,
      "grad_norm": 2.5233420686276986e-08,
      "learning_rate": 0.01506862032774448,
      "loss": 0.0,
      "num_input_tokens_seen": 19452176,
      "step": 34245
    },
    {
      "epoch": 600.8849557522124,
      "grad_norm": 3.56529774592218e-08,
      "learning_rate": 0.015042899133660697,
      "loss": 0.0,
      "num_input_tokens_seen": 19454512,
      "step": 34250
    },
    {
      "epoch": 600.9734513274336,
      "grad_norm": 2.789635900057874e-08,
      "learning_rate": 0.01501719875165789,
      "loss": 0.0,
      "num_input_tokens_seen": 19457248,
      "step": 34255
    },
    {
      "epoch": 601.0530973451328,
      "grad_norm": 1.4806207815354355e-08,
      "learning_rate": 0.014991519185699286,
      "loss": 0.0,
      "num_input_tokens_seen": 19459592,
      "step": 34260
    },
    {
      "epoch": 601.141592920354,
      "grad_norm": 3.874421494742819e-08,
      "learning_rate": 0.014965860439745054,
      "loss": 0.0,
      "num_input_tokens_seen": 19462248,
      "step": 34265
    },
    {
      "epoch": 601.2300884955753,
      "grad_norm": 2.2196088522719037e-08,
      "learning_rate": 0.01494022251775211,
      "loss": 0.0,
      "num_input_tokens_seen": 19465224,
      "step": 34270
    },
    {
      "epoch": 601.3185840707964,
      "grad_norm": 4.8973777921901274e-08,
      "learning_rate": 0.014914605423674109,
      "loss": 0.0,
      "num_input_tokens_seen": 19468040,
      "step": 34275
    },
    {
      "epoch": 601.4070796460177,
      "grad_norm": 2.13692565864676e-08,
      "learning_rate": 0.014889009161461525,
      "loss": 0.0,
      "num_input_tokens_seen": 19470920,
      "step": 34280
    },
    {
      "epoch": 601.4955752212389,
      "grad_norm": 1.0232218983219354e-07,
      "learning_rate": 0.014863433735061665,
      "loss": 0.0,
      "num_input_tokens_seen": 19473928,
      "step": 34285
    },
    {
      "epoch": 601.5840707964602,
      "grad_norm": 1.0989203325095787e-08,
      "learning_rate": 0.014837879148418541,
      "loss": 0.0,
      "num_input_tokens_seen": 19477480,
      "step": 34290
    },
    {
      "epoch": 601.6725663716815,
      "grad_norm": 1.3838260315424122e-07,
      "learning_rate": 0.01481234540547302,
      "loss": 0.0,
      "num_input_tokens_seen": 19479848,
      "step": 34295
    },
    {
      "epoch": 601.7610619469026,
      "grad_norm": 2.6027333177580658e-08,
      "learning_rate": 0.014786832510162717,
      "loss": 0.0,
      "num_input_tokens_seen": 19482648,
      "step": 34300
    },
    {
      "epoch": 601.8495575221239,
      "grad_norm": 3.642574952777977e-08,
      "learning_rate": 0.014761340466422017,
      "loss": 0.0,
      "num_input_tokens_seen": 19485432,
      "step": 34305
    },
    {
      "epoch": 601.9380530973451,
      "grad_norm": 3.509548207603075e-08,
      "learning_rate": 0.014735869278182144,
      "loss": 0.0,
      "num_input_tokens_seen": 19488392,
      "step": 34310
    },
    {
      "epoch": 602.0176991150443,
      "grad_norm": 2.823318112632478e-08,
      "learning_rate": 0.014710418949371057,
      "loss": 0.0,
      "num_input_tokens_seen": 19490824,
      "step": 34315
    },
    {
      "epoch": 602.1061946902655,
      "grad_norm": 2.3232180623722343e-08,
      "learning_rate": 0.014684989483913495,
      "loss": 0.0,
      "num_input_tokens_seen": 19493352,
      "step": 34320
    },
    {
      "epoch": 602.1946902654868,
      "grad_norm": 2.2475546757050324e-08,
      "learning_rate": 0.014659580885731077,
      "loss": 0.0,
      "num_input_tokens_seen": 19496392,
      "step": 34325
    },
    {
      "epoch": 602.2831858407079,
      "grad_norm": 2.0846218973247232e-08,
      "learning_rate": 0.014634193158742047,
      "loss": 0.0,
      "num_input_tokens_seen": 19499208,
      "step": 34330
    },
    {
      "epoch": 602.3716814159292,
      "grad_norm": 7.018467584885002e-08,
      "learning_rate": 0.014608826306861576,
      "loss": 0.0,
      "num_input_tokens_seen": 19502056,
      "step": 34335
    },
    {
      "epoch": 602.4601769911504,
      "grad_norm": 3.861937258875514e-08,
      "learning_rate": 0.014583480334001486,
      "loss": 0.0,
      "num_input_tokens_seen": 19504536,
      "step": 34340
    },
    {
      "epoch": 602.5486725663717,
      "grad_norm": 7.8321022556338e-08,
      "learning_rate": 0.014558155244070496,
      "loss": 0.0,
      "num_input_tokens_seen": 19507144,
      "step": 34345
    },
    {
      "epoch": 602.637168141593,
      "grad_norm": 4.482329174493316e-08,
      "learning_rate": 0.014532851040974036,
      "loss": 0.0,
      "num_input_tokens_seen": 19510280,
      "step": 34350
    },
    {
      "epoch": 602.7256637168142,
      "grad_norm": 3.0667024475405924e-08,
      "learning_rate": 0.014507567728614335,
      "loss": 0.0,
      "num_input_tokens_seen": 19513240,
      "step": 34355
    },
    {
      "epoch": 602.8141592920354,
      "grad_norm": 2.4785869356946932e-08,
      "learning_rate": 0.01448230531089037,
      "loss": 0.0,
      "num_input_tokens_seen": 19516184,
      "step": 34360
    },
    {
      "epoch": 602.9026548672566,
      "grad_norm": 2.0210691786815005e-08,
      "learning_rate": 0.014457063791697993,
      "loss": 0.0,
      "num_input_tokens_seen": 19519640,
      "step": 34365
    },
    {
      "epoch": 602.9911504424779,
      "grad_norm": 3.797019232365528e-08,
      "learning_rate": 0.01443184317492971,
      "loss": 0.0,
      "num_input_tokens_seen": 19522104,
      "step": 34370
    },
    {
      "epoch": 603.070796460177,
      "grad_norm": 7.321847306229756e-08,
      "learning_rate": 0.014406643464474822,
      "loss": 0.0,
      "num_input_tokens_seen": 19524120,
      "step": 34375
    },
    {
      "epoch": 603.1592920353983,
      "grad_norm": 2.855558101089173e-08,
      "learning_rate": 0.014381464664219539,
      "loss": 0.0,
      "num_input_tokens_seen": 19526824,
      "step": 34380
    },
    {
      "epoch": 603.2477876106195,
      "grad_norm": 9.317967908373248e-08,
      "learning_rate": 0.014356306778046656,
      "loss": 0.0,
      "num_input_tokens_seen": 19529816,
      "step": 34385
    },
    {
      "epoch": 603.3362831858407,
      "grad_norm": 1.1371088959322151e-08,
      "learning_rate": 0.014331169809835885,
      "loss": 0.0,
      "num_input_tokens_seen": 19532840,
      "step": 34390
    },
    {
      "epoch": 603.4247787610619,
      "grad_norm": 2.363742801492208e-08,
      "learning_rate": 0.014306053763463644,
      "loss": 0.0,
      "num_input_tokens_seen": 19536040,
      "step": 34395
    },
    {
      "epoch": 603.5132743362832,
      "grad_norm": 2.125154807686158e-08,
      "learning_rate": 0.014280958642803147,
      "loss": 0.0,
      "num_input_tokens_seen": 19539112,
      "step": 34400
    },
    {
      "epoch": 603.5132743362832,
      "eval_loss": 0.7973053455352783,
      "eval_runtime": 0.9341,
      "eval_samples_per_second": 26.763,
      "eval_steps_per_second": 13.917,
      "num_input_tokens_seen": 19539112,
      "step": 34400
    },
    {
      "epoch": 603.6017699115044,
      "grad_norm": 3.959589989221968e-08,
      "learning_rate": 0.014255884451724404,
      "loss": 0.0,
      "num_input_tokens_seen": 19541960,
      "step": 34405
    },
    {
      "epoch": 603.6902654867257,
      "grad_norm": 4.577048073883816e-08,
      "learning_rate": 0.014230831194094101,
      "loss": 0.0,
      "num_input_tokens_seen": 19544568,
      "step": 34410
    },
    {
      "epoch": 603.7787610619469,
      "grad_norm": 7.573844129638019e-08,
      "learning_rate": 0.014205798873775865,
      "loss": 0.0,
      "num_input_tokens_seen": 19547272,
      "step": 34415
    },
    {
      "epoch": 603.8672566371681,
      "grad_norm": 1.1166066826717724e-07,
      "learning_rate": 0.014180787494629893,
      "loss": 0.0,
      "num_input_tokens_seen": 19550296,
      "step": 34420
    },
    {
      "epoch": 603.9557522123894,
      "grad_norm": 3.446922036687283e-08,
      "learning_rate": 0.014155797060513314,
      "loss": 0.0,
      "num_input_tokens_seen": 19553336,
      "step": 34425
    },
    {
      "epoch": 604.0353982300885,
      "grad_norm": 5.899133270759194e-08,
      "learning_rate": 0.014130827575279963,
      "loss": 0.0,
      "num_input_tokens_seen": 19555984,
      "step": 34430
    },
    {
      "epoch": 604.1238938053098,
      "grad_norm": 4.690216570679695e-08,
      "learning_rate": 0.014105879042780427,
      "loss": 0.0,
      "num_input_tokens_seen": 19559088,
      "step": 34435
    },
    {
      "epoch": 604.212389380531,
      "grad_norm": 2.4524625885646856e-08,
      "learning_rate": 0.014080951466862113,
      "loss": 0.0,
      "num_input_tokens_seen": 19562400,
      "step": 34440
    },
    {
      "epoch": 604.3008849557522,
      "grad_norm": 2.854947034336419e-08,
      "learning_rate": 0.014056044851369126,
      "loss": 0.0,
      "num_input_tokens_seen": 19565104,
      "step": 34445
    },
    {
      "epoch": 604.3893805309734,
      "grad_norm": 4.9980648952896445e-08,
      "learning_rate": 0.014031159200142428,
      "loss": 0.0,
      "num_input_tokens_seen": 19567440,
      "step": 34450
    },
    {
      "epoch": 604.4778761061947,
      "grad_norm": 8.072211699072795e-08,
      "learning_rate": 0.014006294517019667,
      "loss": 0.0,
      "num_input_tokens_seen": 19570256,
      "step": 34455
    },
    {
      "epoch": 604.566371681416,
      "grad_norm": 3.6168714245832234e-08,
      "learning_rate": 0.013981450805835276,
      "loss": 0.0,
      "num_input_tokens_seen": 19572928,
      "step": 34460
    },
    {
      "epoch": 604.6548672566372,
      "grad_norm": 1.2096158741314866e-08,
      "learning_rate": 0.01395662807042049,
      "loss": 0.0,
      "num_input_tokens_seen": 19575920,
      "step": 34465
    },
    {
      "epoch": 604.7433628318585,
      "grad_norm": 3.621848065904487e-08,
      "learning_rate": 0.013931826314603296,
      "loss": 0.0,
      "num_input_tokens_seen": 19579440,
      "step": 34470
    },
    {
      "epoch": 604.8318584070796,
      "grad_norm": 4.804483921816427e-08,
      "learning_rate": 0.013907045542208401,
      "loss": 0.0,
      "num_input_tokens_seen": 19581968,
      "step": 34475
    },
    {
      "epoch": 604.9203539823009,
      "grad_norm": 3.2710030239968546e-08,
      "learning_rate": 0.013882285757057333,
      "loss": 0.0,
      "num_input_tokens_seen": 19584656,
      "step": 34480
    },
    {
      "epoch": 605.0,
      "grad_norm": 5.822453630344171e-08,
      "learning_rate": 0.013857546962968403,
      "loss": 0.0,
      "num_input_tokens_seen": 19587152,
      "step": 34485
    },
    {
      "epoch": 605.0884955752213,
      "grad_norm": 2.5866460973134053e-08,
      "learning_rate": 0.013832829163756577,
      "loss": 0.0,
      "num_input_tokens_seen": 19589984,
      "step": 34490
    },
    {
      "epoch": 605.1769911504425,
      "grad_norm": 5.654170820434956e-08,
      "learning_rate": 0.013808132363233689,
      "loss": 0.0,
      "num_input_tokens_seen": 19592784,
      "step": 34495
    },
    {
      "epoch": 605.2654867256637,
      "grad_norm": 2.5182297136439047e-08,
      "learning_rate": 0.013783456565208256,
      "loss": 0.0,
      "num_input_tokens_seen": 19595488,
      "step": 34500
    },
    {
      "epoch": 605.3539823008849,
      "grad_norm": 6.181851830433516e-08,
      "learning_rate": 0.01375880177348564,
      "loss": 0.0,
      "num_input_tokens_seen": 19597968,
      "step": 34505
    },
    {
      "epoch": 605.4424778761062,
      "grad_norm": 1.9925732175352096e-08,
      "learning_rate": 0.013734167991867928,
      "loss": 0.0,
      "num_input_tokens_seen": 19600800,
      "step": 34510
    },
    {
      "epoch": 605.5309734513274,
      "grad_norm": 4.933725605837935e-08,
      "learning_rate": 0.013709555224153935,
      "loss": 0.0,
      "num_input_tokens_seen": 19604080,
      "step": 34515
    },
    {
      "epoch": 605.6194690265487,
      "grad_norm": 2.0059204075550952e-07,
      "learning_rate": 0.013684963474139222,
      "loss": 0.0,
      "num_input_tokens_seen": 19607456,
      "step": 34520
    },
    {
      "epoch": 605.70796460177,
      "grad_norm": 4.2033938996155484e-08,
      "learning_rate": 0.013660392745616224,
      "loss": 0.0,
      "num_input_tokens_seen": 19610304,
      "step": 34525
    },
    {
      "epoch": 605.7964601769911,
      "grad_norm": 5.8566079985666875e-08,
      "learning_rate": 0.013635843042373974,
      "loss": 0.0,
      "num_input_tokens_seen": 19612976,
      "step": 34530
    },
    {
      "epoch": 605.8849557522124,
      "grad_norm": 2.7801563717844147e-08,
      "learning_rate": 0.01361131436819843,
      "loss": 0.0,
      "num_input_tokens_seen": 19616224,
      "step": 34535
    },
    {
      "epoch": 605.9734513274336,
      "grad_norm": 5.1369802633871586e-08,
      "learning_rate": 0.013586806726872147,
      "loss": 0.0,
      "num_input_tokens_seen": 19618912,
      "step": 34540
    },
    {
      "epoch": 606.0530973451328,
      "grad_norm": 4.192483515907952e-08,
      "learning_rate": 0.013562320122174537,
      "loss": 0.0,
      "num_input_tokens_seen": 19621232,
      "step": 34545
    },
    {
      "epoch": 606.141592920354,
      "grad_norm": 4.741044890010926e-08,
      "learning_rate": 0.013537854557881762,
      "loss": 0.0,
      "num_input_tokens_seen": 19623680,
      "step": 34550
    },
    {
      "epoch": 606.2300884955753,
      "grad_norm": 6.466916602221318e-08,
      "learning_rate": 0.013513410037766687,
      "loss": 0.0,
      "num_input_tokens_seen": 19626768,
      "step": 34555
    },
    {
      "epoch": 606.3185840707964,
      "grad_norm": 3.2793511905993e-08,
      "learning_rate": 0.013488986565598998,
      "loss": 0.0,
      "num_input_tokens_seen": 19630224,
      "step": 34560
    },
    {
      "epoch": 606.4070796460177,
      "grad_norm": 6.081109660271977e-08,
      "learning_rate": 0.013464584145145097,
      "loss": 0.0,
      "num_input_tokens_seen": 19633392,
      "step": 34565
    },
    {
      "epoch": 606.4955752212389,
      "grad_norm": 1.9985655796972424e-08,
      "learning_rate": 0.013440202780168109,
      "loss": 0.0,
      "num_input_tokens_seen": 19635872,
      "step": 34570
    },
    {
      "epoch": 606.5840707964602,
      "grad_norm": 1.7025310938834082e-08,
      "learning_rate": 0.01341584247442799,
      "loss": 0.0,
      "num_input_tokens_seen": 19638784,
      "step": 34575
    },
    {
      "epoch": 606.6725663716815,
      "grad_norm": 1.4414576199328621e-08,
      "learning_rate": 0.013391503231681355,
      "loss": 0.0,
      "num_input_tokens_seen": 19641376,
      "step": 34580
    },
    {
      "epoch": 606.7610619469026,
      "grad_norm": 5.820317383609108e-08,
      "learning_rate": 0.013367185055681685,
      "loss": 0.0,
      "num_input_tokens_seen": 19644176,
      "step": 34585
    },
    {
      "epoch": 606.8495575221239,
      "grad_norm": 1.5729256119811907e-08,
      "learning_rate": 0.013342887950179095,
      "loss": 0.0,
      "num_input_tokens_seen": 19647680,
      "step": 34590
    },
    {
      "epoch": 606.9380530973451,
      "grad_norm": 6.455587708842359e-08,
      "learning_rate": 0.013318611918920554,
      "loss": 0.0,
      "num_input_tokens_seen": 19650336,
      "step": 34595
    },
    {
      "epoch": 607.0176991150443,
      "grad_norm": 4.8529997798141267e-08,
      "learning_rate": 0.01329435696564965,
      "loss": 0.0,
      "num_input_tokens_seen": 19652392,
      "step": 34600
    },
    {
      "epoch": 607.0176991150443,
      "eval_loss": 0.7865257263183594,
      "eval_runtime": 0.932,
      "eval_samples_per_second": 26.823,
      "eval_steps_per_second": 13.948,
      "num_input_tokens_seen": 19652392,
      "step": 34600
    },
    {
      "epoch": 607.1061946902655,
      "grad_norm": 2.528213016717018e-08,
      "learning_rate": 0.013270123094106894,
      "loss": 0.0,
      "num_input_tokens_seen": 19654904,
      "step": 34605
    },
    {
      "epoch": 607.1946902654868,
      "grad_norm": 5.326219110202146e-08,
      "learning_rate": 0.013245910308029395,
      "loss": 0.0,
      "num_input_tokens_seen": 19657784,
      "step": 34610
    },
    {
      "epoch": 607.2831858407079,
      "grad_norm": 4.627883853913772e-08,
      "learning_rate": 0.0132217186111511,
      "loss": 0.0,
      "num_input_tokens_seen": 19660856,
      "step": 34615
    },
    {
      "epoch": 607.3716814159292,
      "grad_norm": 6.936889462849649e-08,
      "learning_rate": 0.013197548007202626,
      "loss": 0.0,
      "num_input_tokens_seen": 19663528,
      "step": 34620
    },
    {
      "epoch": 607.4601769911504,
      "grad_norm": 3.5578388235535385e-08,
      "learning_rate": 0.01317339849991142,
      "loss": 0.0,
      "num_input_tokens_seen": 19666168,
      "step": 34625
    },
    {
      "epoch": 607.5486725663717,
      "grad_norm": 4.3130729210361096e-08,
      "learning_rate": 0.013149270093001675,
      "loss": 0.0,
      "num_input_tokens_seen": 19669096,
      "step": 34630
    },
    {
      "epoch": 607.637168141593,
      "grad_norm": 3.978034612828196e-08,
      "learning_rate": 0.013125162790194227,
      "loss": 0.0,
      "num_input_tokens_seen": 19672424,
      "step": 34635
    },
    {
      "epoch": 607.7256637168142,
      "grad_norm": 8.650069993620946e-09,
      "learning_rate": 0.01310107659520674,
      "loss": 0.0,
      "num_input_tokens_seen": 19675128,
      "step": 34640
    },
    {
      "epoch": 607.8141592920354,
      "grad_norm": 4.8252982054464155e-08,
      "learning_rate": 0.013077011511753655,
      "loss": 0.0,
      "num_input_tokens_seen": 19677944,
      "step": 34645
    },
    {
      "epoch": 607.9026548672566,
      "grad_norm": 5.1047468474507696e-08,
      "learning_rate": 0.013052967543546056,
      "loss": 0.0,
      "num_input_tokens_seen": 19681160,
      "step": 34650
    },
    {
      "epoch": 607.9911504424779,
      "grad_norm": 3.349270727426301e-08,
      "learning_rate": 0.01302894469429186,
      "loss": 0.0,
      "num_input_tokens_seen": 19684296,
      "step": 34655
    },
    {
      "epoch": 608.070796460177,
      "grad_norm": 1.819583062001584e-08,
      "learning_rate": 0.013004942967695653,
      "loss": 0.0,
      "num_input_tokens_seen": 19686560,
      "step": 34660
    },
    {
      "epoch": 608.1592920353983,
      "grad_norm": 2.187780623330582e-08,
      "learning_rate": 0.012980962367458859,
      "loss": 0.0,
      "num_input_tokens_seen": 19689328,
      "step": 34665
    },
    {
      "epoch": 608.2477876106195,
      "grad_norm": 5.2718384324634826e-08,
      "learning_rate": 0.012957002897279567,
      "loss": 0.0,
      "num_input_tokens_seen": 19692768,
      "step": 34670
    },
    {
      "epoch": 608.3362831858407,
      "grad_norm": 2.6670706532172517e-08,
      "learning_rate": 0.012933064560852576,
      "loss": 0.0,
      "num_input_tokens_seen": 19695840,
      "step": 34675
    },
    {
      "epoch": 608.4247787610619,
      "grad_norm": 1.8209431118521024e-07,
      "learning_rate": 0.012909147361869527,
      "loss": 0.0,
      "num_input_tokens_seen": 19699120,
      "step": 34680
    },
    {
      "epoch": 608.5132743362832,
      "grad_norm": 4.640748230144709e-08,
      "learning_rate": 0.012885251304018774,
      "loss": 0.0,
      "num_input_tokens_seen": 19701616,
      "step": 34685
    },
    {
      "epoch": 608.6017699115044,
      "grad_norm": 1.3244311425353317e-08,
      "learning_rate": 0.012861376390985335,
      "loss": 0.0,
      "num_input_tokens_seen": 19704512,
      "step": 34690
    },
    {
      "epoch": 608.6902654867257,
      "grad_norm": 5.488977095069458e-08,
      "learning_rate": 0.012837522626451063,
      "loss": 0.0,
      "num_input_tokens_seen": 19706896,
      "step": 34695
    },
    {
      "epoch": 608.7787610619469,
      "grad_norm": 6.77704363738485e-08,
      "learning_rate": 0.01281369001409447,
      "loss": 0.0,
      "num_input_tokens_seen": 19709456,
      "step": 34700
    },
    {
      "epoch": 608.8672566371681,
      "grad_norm": 2.7425150150861555e-08,
      "learning_rate": 0.012789878557590877,
      "loss": 0.0,
      "num_input_tokens_seen": 19712480,
      "step": 34705
    },
    {
      "epoch": 608.9557522123894,
      "grad_norm": 9.182645754890473e-08,
      "learning_rate": 0.012766088260612334,
      "loss": 0.0,
      "num_input_tokens_seen": 19715312,
      "step": 34710
    },
    {
      "epoch": 609.0353982300885,
      "grad_norm": 5.572996641944883e-08,
      "learning_rate": 0.012742319126827523,
      "loss": 0.0,
      "num_input_tokens_seen": 19717576,
      "step": 34715
    },
    {
      "epoch": 609.1238938053098,
      "grad_norm": 3.9513473382157827e-08,
      "learning_rate": 0.012718571159902008,
      "loss": 0.0,
      "num_input_tokens_seen": 19720744,
      "step": 34720
    },
    {
      "epoch": 609.212389380531,
      "grad_norm": 6.126664686689764e-08,
      "learning_rate": 0.01269484436349803,
      "loss": 0.0,
      "num_input_tokens_seen": 19723784,
      "step": 34725
    },
    {
      "epoch": 609.3008849557522,
      "grad_norm": 2.438356716538692e-08,
      "learning_rate": 0.012671138741274528,
      "loss": 0.0,
      "num_input_tokens_seen": 19726920,
      "step": 34730
    },
    {
      "epoch": 609.3893805309734,
      "grad_norm": 6.616791381475196e-08,
      "learning_rate": 0.012647454296887194,
      "loss": 0.0,
      "num_input_tokens_seen": 19730088,
      "step": 34735
    },
    {
      "epoch": 609.4778761061947,
      "grad_norm": 3.945961424278721e-08,
      "learning_rate": 0.012623791033988507,
      "loss": 0.0,
      "num_input_tokens_seen": 19733432,
      "step": 34740
    },
    {
      "epoch": 609.566371681416,
      "grad_norm": 3.909693191417318e-08,
      "learning_rate": 0.012600148956227597,
      "loss": 0.0,
      "num_input_tokens_seen": 19736040,
      "step": 34745
    },
    {
      "epoch": 609.6548672566372,
      "grad_norm": 4.3181085374044414e-08,
      "learning_rate": 0.012576528067250414,
      "loss": 0.0,
      "num_input_tokens_seen": 19738776,
      "step": 34750
    },
    {
      "epoch": 609.7433628318585,
      "grad_norm": 3.492865729981531e-08,
      "learning_rate": 0.012552928370699561,
      "loss": 0.0,
      "num_input_tokens_seen": 19741384,
      "step": 34755
    },
    {
      "epoch": 609.8318584070796,
      "grad_norm": 4.0844671644890695e-08,
      "learning_rate": 0.012529349870214411,
      "loss": 0.0,
      "num_input_tokens_seen": 19744232,
      "step": 34760
    },
    {
      "epoch": 609.9203539823009,
      "grad_norm": 2.2538223731771723e-08,
      "learning_rate": 0.012505792569431106,
      "loss": 0.0,
      "num_input_tokens_seen": 19746936,
      "step": 34765
    },
    {
      "epoch": 610.0,
      "grad_norm": 4.902017280983273e-08,
      "learning_rate": 0.012482256471982422,
      "loss": 0.0,
      "num_input_tokens_seen": 19749320,
      "step": 34770
    },
    {
      "epoch": 610.0884955752213,
      "grad_norm": 3.280094418300905e-08,
      "learning_rate": 0.012458741581497956,
      "loss": 0.0,
      "num_input_tokens_seen": 19752776,
      "step": 34775
    },
    {
      "epoch": 610.1769911504425,
      "grad_norm": 4.829253086313656e-08,
      "learning_rate": 0.012435247901603974,
      "loss": 0.0,
      "num_input_tokens_seen": 19755656,
      "step": 34780
    },
    {
      "epoch": 610.2654867256637,
      "grad_norm": 1.8944950497257196e-08,
      "learning_rate": 0.012411775435923528,
      "loss": 0.0,
      "num_input_tokens_seen": 19758344,
      "step": 34785
    },
    {
      "epoch": 610.3539823008849,
      "grad_norm": 3.239629720042103e-08,
      "learning_rate": 0.012388324188076354,
      "loss": 0.0,
      "num_input_tokens_seen": 19761240,
      "step": 34790
    },
    {
      "epoch": 610.4424778761062,
      "grad_norm": 1.4895189082153593e-08,
      "learning_rate": 0.012364894161678913,
      "loss": 0.0,
      "num_input_tokens_seen": 19764248,
      "step": 34795
    },
    {
      "epoch": 610.5309734513274,
      "grad_norm": 3.365513734365777e-08,
      "learning_rate": 0.012341485360344445,
      "loss": 0.0,
      "num_input_tokens_seen": 19766904,
      "step": 34800
    },
    {
      "epoch": 610.5309734513274,
      "eval_loss": 0.8026770353317261,
      "eval_runtime": 0.9397,
      "eval_samples_per_second": 26.603,
      "eval_steps_per_second": 13.834,
      "num_input_tokens_seen": 19766904,
      "step": 34800
    },
    {
      "epoch": 610.6194690265487,
      "grad_norm": 4.859318991634609e-08,
      "learning_rate": 0.01231809778768283,
      "loss": 0.0,
      "num_input_tokens_seen": 19769528,
      "step": 34805
    },
    {
      "epoch": 610.70796460177,
      "grad_norm": 2.663139397895975e-08,
      "learning_rate": 0.012294731447300799,
      "loss": 0.0,
      "num_input_tokens_seen": 19772296,
      "step": 34810
    },
    {
      "epoch": 610.7964601769911,
      "grad_norm": 4.611985815472508e-08,
      "learning_rate": 0.012271386342801671,
      "loss": 0.0,
      "num_input_tokens_seen": 19775096,
      "step": 34815
    },
    {
      "epoch": 610.8849557522124,
      "grad_norm": 6.418071762936961e-08,
      "learning_rate": 0.012248062477785565,
      "loss": 0.0,
      "num_input_tokens_seen": 19777944,
      "step": 34820
    },
    {
      "epoch": 610.9734513274336,
      "grad_norm": 6.66990658260147e-08,
      "learning_rate": 0.012224759855849305,
      "loss": 0.0,
      "num_input_tokens_seen": 19780600,
      "step": 34825
    },
    {
      "epoch": 611.0530973451328,
      "grad_norm": 3.9752379166202445e-08,
      "learning_rate": 0.012201478480586513,
      "loss": 0.0,
      "num_input_tokens_seen": 19783248,
      "step": 34830
    },
    {
      "epoch": 611.141592920354,
      "grad_norm": 1.5444573620015944e-08,
      "learning_rate": 0.012178218355587389,
      "loss": 0.0,
      "num_input_tokens_seen": 19786496,
      "step": 34835
    },
    {
      "epoch": 611.2300884955753,
      "grad_norm": 2.3326267140078016e-08,
      "learning_rate": 0.01215497948443896,
      "loss": 0.0,
      "num_input_tokens_seen": 19789584,
      "step": 34840
    },
    {
      "epoch": 611.3185840707964,
      "grad_norm": 1.889429057655434e-08,
      "learning_rate": 0.012131761870724993,
      "loss": 0.0,
      "num_input_tokens_seen": 19792288,
      "step": 34845
    },
    {
      "epoch": 611.4070796460177,
      "grad_norm": 3.698351846992409e-08,
      "learning_rate": 0.012108565518025893,
      "loss": 0.0,
      "num_input_tokens_seen": 19794864,
      "step": 34850
    },
    {
      "epoch": 611.4955752212389,
      "grad_norm": 2.988483771559913e-08,
      "learning_rate": 0.012085390429918862,
      "loss": 0.0,
      "num_input_tokens_seen": 19797264,
      "step": 34855
    },
    {
      "epoch": 611.5840707964602,
      "grad_norm": 1.948395755846377e-08,
      "learning_rate": 0.012062236609977744,
      "loss": 0.0,
      "num_input_tokens_seen": 19800448,
      "step": 34860
    },
    {
      "epoch": 611.6725663716815,
      "grad_norm": 4.1631899705407704e-08,
      "learning_rate": 0.01203910406177318,
      "loss": 0.0,
      "num_input_tokens_seen": 19803584,
      "step": 34865
    },
    {
      "epoch": 611.7610619469026,
      "grad_norm": 5.5365347861879854e-08,
      "learning_rate": 0.01201599278887252,
      "loss": 0.0,
      "num_input_tokens_seen": 19806336,
      "step": 34870
    },
    {
      "epoch": 611.8495575221239,
      "grad_norm": 1.2117820524792933e-08,
      "learning_rate": 0.011992902794839744,
      "loss": 0.0,
      "num_input_tokens_seen": 19809456,
      "step": 34875
    },
    {
      "epoch": 611.9380530973451,
      "grad_norm": 6.346784431343622e-08,
      "learning_rate": 0.011969834083235703,
      "loss": 0.0,
      "num_input_tokens_seen": 19812656,
      "step": 34880
    },
    {
      "epoch": 612.0176991150443,
      "grad_norm": 3.2140679451231335e-08,
      "learning_rate": 0.011946786657617836,
      "loss": 0.0,
      "num_input_tokens_seen": 19814672,
      "step": 34885
    },
    {
      "epoch": 612.1061946902655,
      "grad_norm": 6.996198465003545e-08,
      "learning_rate": 0.011923760521540332,
      "loss": 0.0,
      "num_input_tokens_seen": 19817376,
      "step": 34890
    },
    {
      "epoch": 612.1946902654868,
      "grad_norm": 2.9415412328148705e-08,
      "learning_rate": 0.011900755678554153,
      "loss": 0.0,
      "num_input_tokens_seen": 19820384,
      "step": 34895
    },
    {
      "epoch": 612.2831858407079,
      "grad_norm": 1.082010214759066e-08,
      "learning_rate": 0.011877772132206893,
      "loss": 0.0,
      "num_input_tokens_seen": 19822992,
      "step": 34900
    },
    {
      "epoch": 612.3716814159292,
      "grad_norm": 1.7334333080043507e-08,
      "learning_rate": 0.011854809886042915,
      "loss": 0.0,
      "num_input_tokens_seen": 19825552,
      "step": 34905
    },
    {
      "epoch": 612.4601769911504,
      "grad_norm": 3.836788664557389e-08,
      "learning_rate": 0.011831868943603325,
      "loss": 0.0,
      "num_input_tokens_seen": 19828320,
      "step": 34910
    },
    {
      "epoch": 612.5486725663717,
      "grad_norm": 5.7816169629631986e-08,
      "learning_rate": 0.011808949308425836,
      "loss": 0.0,
      "num_input_tokens_seen": 19832016,
      "step": 34915
    },
    {
      "epoch": 612.637168141593,
      "grad_norm": 4.6166842793127216e-08,
      "learning_rate": 0.01178605098404501,
      "loss": 0.0,
      "num_input_tokens_seen": 19835632,
      "step": 34920
    },
    {
      "epoch": 612.7256637168142,
      "grad_norm": 3.1490738905404214e-08,
      "learning_rate": 0.011763173973992002,
      "loss": 0.0,
      "num_input_tokens_seen": 19838384,
      "step": 34925
    },
    {
      "epoch": 612.8141592920354,
      "grad_norm": 5.288305970907459e-08,
      "learning_rate": 0.011740318281794776,
      "loss": 0.0,
      "num_input_tokens_seen": 19840784,
      "step": 34930
    },
    {
      "epoch": 612.9026548672566,
      "grad_norm": 4.346510351638244e-08,
      "learning_rate": 0.01171748391097796,
      "loss": 0.0,
      "num_input_tokens_seen": 19843472,
      "step": 34935
    },
    {
      "epoch": 612.9911504424779,
      "grad_norm": 4.9419025316410625e-08,
      "learning_rate": 0.011694670865062873,
      "loss": 0.0,
      "num_input_tokens_seen": 19846304,
      "step": 34940
    },
    {
      "epoch": 613.070796460177,
      "grad_norm": 3.235821210978429e-08,
      "learning_rate": 0.011671879147567616,
      "loss": 0.0,
      "num_input_tokens_seen": 19848624,
      "step": 34945
    },
    {
      "epoch": 613.1592920353983,
      "grad_norm": 4.659067442958076e-08,
      "learning_rate": 0.011649108762006893,
      "loss": 0.0,
      "num_input_tokens_seen": 19851312,
      "step": 34950
    },
    {
      "epoch": 613.2477876106195,
      "grad_norm": 2.0164609537687284e-08,
      "learning_rate": 0.011626359711892265,
      "loss": 0.0,
      "num_input_tokens_seen": 19854032,
      "step": 34955
    },
    {
      "epoch": 613.3362831858407,
      "grad_norm": 3.346449517493966e-08,
      "learning_rate": 0.01160363200073189,
      "loss": 0.0,
      "num_input_tokens_seen": 19856832,
      "step": 34960
    },
    {
      "epoch": 613.4247787610619,
      "grad_norm": 4.131363340320604e-08,
      "learning_rate": 0.011580925632030614,
      "loss": 0.0,
      "num_input_tokens_seen": 19859984,
      "step": 34965
    },
    {
      "epoch": 613.5132743362832,
      "grad_norm": 2.9904583698225906e-08,
      "learning_rate": 0.011558240609290104,
      "loss": 0.0,
      "num_input_tokens_seen": 19863152,
      "step": 34970
    },
    {
      "epoch": 613.6017699115044,
      "grad_norm": 1.0214664314389665e-07,
      "learning_rate": 0.011535576936008679,
      "loss": 0.0,
      "num_input_tokens_seen": 19865824,
      "step": 34975
    },
    {
      "epoch": 613.6902654867257,
      "grad_norm": 4.3640252300747306e-08,
      "learning_rate": 0.011512934615681309,
      "loss": 0.0,
      "num_input_tokens_seen": 19868768,
      "step": 34980
    },
    {
      "epoch": 613.7787610619469,
      "grad_norm": 3.753305577447463e-08,
      "learning_rate": 0.011490313651799765,
      "loss": 0.0,
      "num_input_tokens_seen": 19871184,
      "step": 34985
    },
    {
      "epoch": 613.8672566371681,
      "grad_norm": 3.473439846857218e-08,
      "learning_rate": 0.011467714047852512,
      "loss": 0.0,
      "num_input_tokens_seen": 19874480,
      "step": 34990
    },
    {
      "epoch": 613.9557522123894,
      "grad_norm": 9.138671686059752e-08,
      "learning_rate": 0.011445135807324624,
      "loss": 0.0,
      "num_input_tokens_seen": 19877808,
      "step": 34995
    },
    {
      "epoch": 614.0353982300885,
      "grad_norm": 1.5541480991032586e-08,
      "learning_rate": 0.011422578933698002,
      "loss": 0.0,
      "num_input_tokens_seen": 19879808,
      "step": 35000
    },
    {
      "epoch": 614.0353982300885,
      "eval_loss": 0.8159120082855225,
      "eval_runtime": 0.9473,
      "eval_samples_per_second": 26.392,
      "eval_steps_per_second": 13.724,
      "num_input_tokens_seen": 19879808,
      "step": 35000
    },
    {
      "epoch": 614.1238938053098,
      "grad_norm": 5.97066076579722e-08,
      "learning_rate": 0.011400043430451161,
      "loss": 0.0,
      "num_input_tokens_seen": 19882944,
      "step": 35005
    },
    {
      "epoch": 614.212389380531,
      "grad_norm": 2.3124076875546962e-08,
      "learning_rate": 0.011377529301059392,
      "loss": 0.0,
      "num_input_tokens_seen": 19885712,
      "step": 35010
    },
    {
      "epoch": 614.3008849557522,
      "grad_norm": 6.204471247883703e-08,
      "learning_rate": 0.011355036548994646,
      "loss": 0.0,
      "num_input_tokens_seen": 19889200,
      "step": 35015
    },
    {
      "epoch": 614.3893805309734,
      "grad_norm": 1.2840845720063498e-08,
      "learning_rate": 0.011332565177725584,
      "loss": 0.0,
      "num_input_tokens_seen": 19892128,
      "step": 35020
    },
    {
      "epoch": 614.4778761061947,
      "grad_norm": 2.21002949274407e-08,
      "learning_rate": 0.011310115190717585,
      "loss": 0.0,
      "num_input_tokens_seen": 19894768,
      "step": 35025
    },
    {
      "epoch": 614.566371681416,
      "grad_norm": 7.042755356678754e-08,
      "learning_rate": 0.01128768659143271,
      "loss": 0.0,
      "num_input_tokens_seen": 19897520,
      "step": 35030
    },
    {
      "epoch": 614.6548672566372,
      "grad_norm": 4.884065418764294e-08,
      "learning_rate": 0.011265279383329713,
      "loss": 0.0,
      "num_input_tokens_seen": 19899888,
      "step": 35035
    },
    {
      "epoch": 614.7433628318585,
      "grad_norm": 6.549825570800749e-08,
      "learning_rate": 0.01124289356986411,
      "loss": 0.0,
      "num_input_tokens_seen": 19903024,
      "step": 35040
    },
    {
      "epoch": 614.8318584070796,
      "grad_norm": 7.186221750998811e-08,
      "learning_rate": 0.011220529154488023,
      "loss": 0.0,
      "num_input_tokens_seen": 19905856,
      "step": 35045
    },
    {
      "epoch": 614.9203539823009,
      "grad_norm": 1.5788732099508707e-08,
      "learning_rate": 0.011198186140650346,
      "loss": 0.0,
      "num_input_tokens_seen": 19908624,
      "step": 35050
    },
    {
      "epoch": 615.0,
      "grad_norm": 3.5770401751733516e-08,
      "learning_rate": 0.011175864531796685,
      "loss": 0.0,
      "num_input_tokens_seen": 19911632,
      "step": 35055
    },
    {
      "epoch": 615.0884955752213,
      "grad_norm": 3.6758141419568346e-08,
      "learning_rate": 0.011153564331369258,
      "loss": 0.0,
      "num_input_tokens_seen": 19914528,
      "step": 35060
    },
    {
      "epoch": 615.1769911504425,
      "grad_norm": 1.9911778892378607e-08,
      "learning_rate": 0.011131285542807078,
      "loss": 0.0,
      "num_input_tokens_seen": 19917424,
      "step": 35065
    },
    {
      "epoch": 615.2654867256637,
      "grad_norm": 2.267776011422029e-08,
      "learning_rate": 0.011109028169545815,
      "loss": 0.0,
      "num_input_tokens_seen": 19919952,
      "step": 35070
    },
    {
      "epoch": 615.3539823008849,
      "grad_norm": 3.426308481380147e-08,
      "learning_rate": 0.011086792215017804,
      "loss": 0.0,
      "num_input_tokens_seen": 19922608,
      "step": 35075
    },
    {
      "epoch": 615.4424778761062,
      "grad_norm": 2.1737935895771443e-08,
      "learning_rate": 0.011064577682652137,
      "loss": 0.0,
      "num_input_tokens_seen": 19925584,
      "step": 35080
    },
    {
      "epoch": 615.5309734513274,
      "grad_norm": 4.587133517475195e-08,
      "learning_rate": 0.011042384575874559,
      "loss": 0.0,
      "num_input_tokens_seen": 19928576,
      "step": 35085
    },
    {
      "epoch": 615.6194690265487,
      "grad_norm": 1.0263689631528905e-07,
      "learning_rate": 0.011020212898107512,
      "loss": 0.0,
      "num_input_tokens_seen": 19931808,
      "step": 35090
    },
    {
      "epoch": 615.70796460177,
      "grad_norm": 3.592248631889561e-08,
      "learning_rate": 0.010998062652770197,
      "loss": 0.0,
      "num_input_tokens_seen": 19934528,
      "step": 35095
    },
    {
      "epoch": 615.7964601769911,
      "grad_norm": 4.342821213754178e-08,
      "learning_rate": 0.010975933843278428,
      "loss": 0.0,
      "num_input_tokens_seen": 19937040,
      "step": 35100
    },
    {
      "epoch": 615.8849557522124,
      "grad_norm": 3.890325928068705e-08,
      "learning_rate": 0.010953826473044714,
      "loss": 0.0,
      "num_input_tokens_seen": 19939840,
      "step": 35105
    },
    {
      "epoch": 615.9734513274336,
      "grad_norm": 2.5001044789974003e-08,
      "learning_rate": 0.010931740545478357,
      "loss": 0.0,
      "num_input_tokens_seen": 19943088,
      "step": 35110
    },
    {
      "epoch": 616.0530973451328,
      "grad_norm": 2.819897382266845e-08,
      "learning_rate": 0.010909676063985218,
      "loss": 0.0,
      "num_input_tokens_seen": 19945608,
      "step": 35115
    },
    {
      "epoch": 616.141592920354,
      "grad_norm": 5.501622268866413e-08,
      "learning_rate": 0.010887633031967974,
      "loss": 0.0,
      "num_input_tokens_seen": 19949160,
      "step": 35120
    },
    {
      "epoch": 616.2300884955753,
      "grad_norm": 3.439635776203431e-08,
      "learning_rate": 0.01086561145282589,
      "loss": 0.0,
      "num_input_tokens_seen": 19951752,
      "step": 35125
    },
    {
      "epoch": 616.3185840707964,
      "grad_norm": 2.006376753627137e-08,
      "learning_rate": 0.010843611329954983,
      "loss": 0.0,
      "num_input_tokens_seen": 19954360,
      "step": 35130
    },
    {
      "epoch": 616.4070796460177,
      "grad_norm": 1.9099939407851707e-08,
      "learning_rate": 0.010821632666747988,
      "loss": 0.0,
      "num_input_tokens_seen": 19957512,
      "step": 35135
    },
    {
      "epoch": 616.4955752212389,
      "grad_norm": 3.260218761624856e-08,
      "learning_rate": 0.010799675466594244,
      "loss": 0.0,
      "num_input_tokens_seen": 19959912,
      "step": 35140
    },
    {
      "epoch": 616.5840707964602,
      "grad_norm": 9.55368903987619e-08,
      "learning_rate": 0.010777739732879826,
      "loss": 0.0,
      "num_input_tokens_seen": 19962712,
      "step": 35145
    },
    {
      "epoch": 616.6725663716815,
      "grad_norm": 6.233225491314442e-08,
      "learning_rate": 0.010755825468987562,
      "loss": 0.0,
      "num_input_tokens_seen": 19965896,
      "step": 35150
    },
    {
      "epoch": 616.7610619469026,
      "grad_norm": 9.251436949853087e-08,
      "learning_rate": 0.010733932678296814,
      "loss": 0.0,
      "num_input_tokens_seen": 19968424,
      "step": 35155
    },
    {
      "epoch": 616.8495575221239,
      "grad_norm": 2.561074730067503e-08,
      "learning_rate": 0.010712061364183817,
      "loss": 0.0,
      "num_input_tokens_seen": 19971368,
      "step": 35160
    },
    {
      "epoch": 616.9380530973451,
      "grad_norm": 3.712296603453069e-08,
      "learning_rate": 0.010690211530021337,
      "loss": 0.0,
      "num_input_tokens_seen": 19974552,
      "step": 35165
    },
    {
      "epoch": 617.0176991150443,
      "grad_norm": 5.181991014069354e-08,
      "learning_rate": 0.01066838317917893,
      "loss": 0.0,
      "num_input_tokens_seen": 19976784,
      "step": 35170
    },
    {
      "epoch": 617.1061946902655,
      "grad_norm": 1.2893822898263352e-08,
      "learning_rate": 0.010646576315022787,
      "loss": 0.0,
      "num_input_tokens_seen": 19979376,
      "step": 35175
    },
    {
      "epoch": 617.1946902654868,
      "grad_norm": 6.653246487076103e-08,
      "learning_rate": 0.010624790940915785,
      "loss": 0.0,
      "num_input_tokens_seen": 19982160,
      "step": 35180
    },
    {
      "epoch": 617.2831858407079,
      "grad_norm": 3.2854412523875e-08,
      "learning_rate": 0.0106030270602175,
      "loss": 0.0,
      "num_input_tokens_seen": 19985040,
      "step": 35185
    },
    {
      "epoch": 617.3716814159292,
      "grad_norm": 1.8023262882138624e-08,
      "learning_rate": 0.010581284676284252,
      "loss": 0.0,
      "num_input_tokens_seen": 19987680,
      "step": 35190
    },
    {
      "epoch": 617.4601769911504,
      "grad_norm": 3.05828677937825e-08,
      "learning_rate": 0.010559563792468923,
      "loss": 0.0,
      "num_input_tokens_seen": 19990416,
      "step": 35195
    },
    {
      "epoch": 617.5486725663717,
      "grad_norm": 2.6095971605855084e-08,
      "learning_rate": 0.010537864412121217,
      "loss": 0.0,
      "num_input_tokens_seen": 19993952,
      "step": 35200
    },
    {
      "epoch": 617.5486725663717,
      "eval_loss": 0.8087599873542786,
      "eval_runtime": 0.9461,
      "eval_samples_per_second": 26.425,
      "eval_steps_per_second": 13.741,
      "num_input_tokens_seen": 19993952,
      "step": 35200
    },
    {
      "epoch": 617.637168141593,
      "grad_norm": 4.017178767412588e-08,
      "learning_rate": 0.010516186538587357,
      "loss": 0.0,
      "num_input_tokens_seen": 19997328,
      "step": 35205
    },
    {
      "epoch": 617.7256637168142,
      "grad_norm": 1.970399843287396e-08,
      "learning_rate": 0.01049453017521042,
      "loss": 0.0,
      "num_input_tokens_seen": 20000320,
      "step": 35210
    },
    {
      "epoch": 617.8141592920354,
      "grad_norm": 3.592591468759565e-08,
      "learning_rate": 0.010472895325330083,
      "loss": 0.0,
      "num_input_tokens_seen": 20003296,
      "step": 35215
    },
    {
      "epoch": 617.9026548672566,
      "grad_norm": 3.9218154057607535e-08,
      "learning_rate": 0.010451281992282662,
      "loss": 0.0,
      "num_input_tokens_seen": 20006144,
      "step": 35220
    },
    {
      "epoch": 617.9911504424779,
      "grad_norm": 4.758028282481064e-08,
      "learning_rate": 0.01042969017940124,
      "loss": 0.0,
      "num_input_tokens_seen": 20008832,
      "step": 35225
    },
    {
      "epoch": 618.070796460177,
      "grad_norm": 1.4923386970622232e-08,
      "learning_rate": 0.01040811989001557,
      "loss": 0.0,
      "num_input_tokens_seen": 20011784,
      "step": 35230
    },
    {
      "epoch": 618.1592920353983,
      "grad_norm": 5.965829785736787e-08,
      "learning_rate": 0.010386571127451992,
      "loss": 0.0,
      "num_input_tokens_seen": 20014936,
      "step": 35235
    },
    {
      "epoch": 618.2477876106195,
      "grad_norm": 4.773392703327772e-08,
      "learning_rate": 0.010365043895033682,
      "loss": 0.0,
      "num_input_tokens_seen": 20017608,
      "step": 35240
    },
    {
      "epoch": 618.3362831858407,
      "grad_norm": 4.6699060618493604e-08,
      "learning_rate": 0.010343538196080365,
      "loss": 0.0,
      "num_input_tokens_seen": 20020616,
      "step": 35245
    },
    {
      "epoch": 618.4247787610619,
      "grad_norm": 2.3207705979189086e-08,
      "learning_rate": 0.010322054033908457,
      "loss": 0.0,
      "num_input_tokens_seen": 20023352,
      "step": 35250
    },
    {
      "epoch": 618.5132743362832,
      "grad_norm": 3.117172298061632e-08,
      "learning_rate": 0.010300591411831156,
      "loss": 0.0,
      "num_input_tokens_seen": 20025912,
      "step": 35255
    },
    {
      "epoch": 618.6017699115044,
      "grad_norm": 3.8519306855278046e-08,
      "learning_rate": 0.010279150333158198,
      "loss": 0.0,
      "num_input_tokens_seen": 20028280,
      "step": 35260
    },
    {
      "epoch": 618.6902654867257,
      "grad_norm": 2.2896029960861597e-08,
      "learning_rate": 0.010257730801196107,
      "loss": 0.0,
      "num_input_tokens_seen": 20031176,
      "step": 35265
    },
    {
      "epoch": 618.7787610619469,
      "grad_norm": 2.1631787916476242e-08,
      "learning_rate": 0.010236332819248056,
      "loss": 0.0,
      "num_input_tokens_seen": 20033912,
      "step": 35270
    },
    {
      "epoch": 618.8672566371681,
      "grad_norm": 2.7715607586742408e-08,
      "learning_rate": 0.010214956390613854,
      "loss": 0.0,
      "num_input_tokens_seen": 20036936,
      "step": 35275
    },
    {
      "epoch": 618.9557522123894,
      "grad_norm": 5.082806353584601e-08,
      "learning_rate": 0.010193601518590034,
      "loss": 0.0,
      "num_input_tokens_seen": 20039928,
      "step": 35280
    },
    {
      "epoch": 619.0353982300885,
      "grad_norm": 4.730117808549039e-08,
      "learning_rate": 0.010172268206469758,
      "loss": 0.0,
      "num_input_tokens_seen": 20041984,
      "step": 35285
    },
    {
      "epoch": 619.1238938053098,
      "grad_norm": 1.6364094079790448e-08,
      "learning_rate": 0.010150956457542897,
      "loss": 0.0,
      "num_input_tokens_seen": 20044912,
      "step": 35290
    },
    {
      "epoch": 619.212389380531,
      "grad_norm": 3.1816863810263385e-08,
      "learning_rate": 0.010129666275096054,
      "loss": 0.0,
      "num_input_tokens_seen": 20047888,
      "step": 35295
    },
    {
      "epoch": 619.3008849557522,
      "grad_norm": 5.427827431958576e-08,
      "learning_rate": 0.010108397662412338,
      "loss": 0.0,
      "num_input_tokens_seen": 20050784,
      "step": 35300
    },
    {
      "epoch": 619.3893805309734,
      "grad_norm": 2.447749203327021e-08,
      "learning_rate": 0.010087150622771707,
      "loss": 0.0,
      "num_input_tokens_seen": 20053376,
      "step": 35305
    },
    {
      "epoch": 619.4778761061947,
      "grad_norm": 1.2337507371285028e-07,
      "learning_rate": 0.010065925159450739,
      "loss": 0.0,
      "num_input_tokens_seen": 20056288,
      "step": 35310
    },
    {
      "epoch": 619.566371681416,
      "grad_norm": 1.0604500033650766e-07,
      "learning_rate": 0.010044721275722618,
      "loss": 0.0,
      "num_input_tokens_seen": 20059344,
      "step": 35315
    },
    {
      "epoch": 619.6548672566372,
      "grad_norm": 7.608757357502327e-08,
      "learning_rate": 0.01002353897485726,
      "loss": 0.0,
      "num_input_tokens_seen": 20062768,
      "step": 35320
    },
    {
      "epoch": 619.7433628318585,
      "grad_norm": 2.5282563598238994e-08,
      "learning_rate": 0.010002378260121236,
      "loss": 0.0,
      "num_input_tokens_seen": 20065344,
      "step": 35325
    },
    {
      "epoch": 619.8318584070796,
      "grad_norm": 3.467662423872753e-08,
      "learning_rate": 0.009981239134777786,
      "loss": 0.0,
      "num_input_tokens_seen": 20068240,
      "step": 35330
    },
    {
      "epoch": 619.9203539823009,
      "grad_norm": 3.0295165487359554e-08,
      "learning_rate": 0.009960121602086884,
      "loss": 0.0,
      "num_input_tokens_seen": 20070784,
      "step": 35335
    },
    {
      "epoch": 620.0,
      "grad_norm": 2.7444103878337955e-08,
      "learning_rate": 0.009939025665305062,
      "loss": 0.0,
      "num_input_tokens_seen": 20073496,
      "step": 35340
    },
    {
      "epoch": 620.0884955752213,
      "grad_norm": 1.7573780652924142e-08,
      "learning_rate": 0.009917951327685597,
      "loss": 0.0,
      "num_input_tokens_seen": 20077080,
      "step": 35345
    },
    {
      "epoch": 620.1769911504425,
      "grad_norm": 1.6410249159548584e-08,
      "learning_rate": 0.009896898592478425,
      "loss": 0.0,
      "num_input_tokens_seen": 20079960,
      "step": 35350
    },
    {
      "epoch": 620.2654867256637,
      "grad_norm": 5.122269897128717e-08,
      "learning_rate": 0.009875867462930132,
      "loss": 0.0,
      "num_input_tokens_seen": 20082744,
      "step": 35355
    },
    {
      "epoch": 620.3539823008849,
      "grad_norm": 1.903304891470725e-08,
      "learning_rate": 0.009854857942284006,
      "loss": 0.0,
      "num_input_tokens_seen": 20085416,
      "step": 35360
    },
    {
      "epoch": 620.4424778761062,
      "grad_norm": 4.9211937636073344e-08,
      "learning_rate": 0.009833870033779923,
      "loss": 0.0,
      "num_input_tokens_seen": 20087928,
      "step": 35365
    },
    {
      "epoch": 620.5309734513274,
      "grad_norm": 2.1951443329726317e-08,
      "learning_rate": 0.009812903740654527,
      "loss": 0.0,
      "num_input_tokens_seen": 20090600,
      "step": 35370
    },
    {
      "epoch": 620.6194690265487,
      "grad_norm": 6.082048997768652e-08,
      "learning_rate": 0.009791959066141097,
      "loss": 0.0,
      "num_input_tokens_seen": 20093560,
      "step": 35375
    },
    {
      "epoch": 620.70796460177,
      "grad_norm": 3.256888803093716e-08,
      "learning_rate": 0.009771036013469537,
      "loss": 0.0,
      "num_input_tokens_seen": 20096488,
      "step": 35380
    },
    {
      "epoch": 620.7964601769911,
      "grad_norm": 4.282190957383136e-08,
      "learning_rate": 0.00975013458586646,
      "loss": 0.0,
      "num_input_tokens_seen": 20099512,
      "step": 35385
    },
    {
      "epoch": 620.8849557522124,
      "grad_norm": 2.204463811494861e-08,
      "learning_rate": 0.009729254786555107,
      "loss": 0.0,
      "num_input_tokens_seen": 20102568,
      "step": 35390
    },
    {
      "epoch": 620.9734513274336,
      "grad_norm": 1.0572760800187098e-07,
      "learning_rate": 0.009708396618755421,
      "loss": 0.0,
      "num_input_tokens_seen": 20105112,
      "step": 35395
    },
    {
      "epoch": 621.0530973451328,
      "grad_norm": 7.004344126926298e-08,
      "learning_rate": 0.009687560085683994,
      "loss": 0.0,
      "num_input_tokens_seen": 20107560,
      "step": 35400
    },
    {
      "epoch": 621.0530973451328,
      "eval_loss": 0.8044732809066772,
      "eval_runtime": 0.9434,
      "eval_samples_per_second": 26.499,
      "eval_steps_per_second": 13.78,
      "num_input_tokens_seen": 20107560,
      "step": 35400
    },
    {
      "epoch": 621.141592920354,
      "grad_norm": 2.5927237246037294e-08,
      "learning_rate": 0.009666745190554054,
      "loss": 0.0,
      "num_input_tokens_seen": 20110056,
      "step": 35405
    },
    {
      "epoch": 621.2300884955753,
      "grad_norm": 5.724987772737222e-08,
      "learning_rate": 0.009645951936575553,
      "loss": 0.0,
      "num_input_tokens_seen": 20113336,
      "step": 35410
    },
    {
      "epoch": 621.3185840707964,
      "grad_norm": 7.078506314428523e-08,
      "learning_rate": 0.00962518032695509,
      "loss": 0.0,
      "num_input_tokens_seen": 20115784,
      "step": 35415
    },
    {
      "epoch": 621.4070796460177,
      "grad_norm": 3.4440930107848544e-08,
      "learning_rate": 0.009604430364895855,
      "loss": 0.0,
      "num_input_tokens_seen": 20118760,
      "step": 35420
    },
    {
      "epoch": 621.4955752212389,
      "grad_norm": 5.671471470236611e-08,
      "learning_rate": 0.00958370205359777,
      "loss": 0.0,
      "num_input_tokens_seen": 20121896,
      "step": 35425
    },
    {
      "epoch": 621.5840707964602,
      "grad_norm": 5.932748337045268e-08,
      "learning_rate": 0.009562995396257445,
      "loss": 0.0,
      "num_input_tokens_seen": 20124536,
      "step": 35430
    },
    {
      "epoch": 621.6725663716815,
      "grad_norm": 2.9823905123294026e-08,
      "learning_rate": 0.009542310396068026,
      "loss": 0.0,
      "num_input_tokens_seen": 20127448,
      "step": 35435
    },
    {
      "epoch": 621.7610619469026,
      "grad_norm": 4.3993924947471896e-08,
      "learning_rate": 0.009521647056219495,
      "loss": 0.0,
      "num_input_tokens_seen": 20130504,
      "step": 35440
    },
    {
      "epoch": 621.8495575221239,
      "grad_norm": 3.279973626035826e-08,
      "learning_rate": 0.00950100537989832,
      "loss": 0.0,
      "num_input_tokens_seen": 20133064,
      "step": 35445
    },
    {
      "epoch": 621.9380530973451,
      "grad_norm": 6.478993697101032e-08,
      "learning_rate": 0.00948038537028772,
      "loss": 0.0,
      "num_input_tokens_seen": 20136632,
      "step": 35450
    },
    {
      "epoch": 622.0176991150443,
      "grad_norm": 1.964955487210318e-08,
      "learning_rate": 0.009459787030567617,
      "loss": 0.0,
      "num_input_tokens_seen": 20139096,
      "step": 35455
    },
    {
      "epoch": 622.1061946902655,
      "grad_norm": 6.54334826322156e-08,
      "learning_rate": 0.00943921036391449,
      "loss": 0.0,
      "num_input_tokens_seen": 20141880,
      "step": 35460
    },
    {
      "epoch": 622.1946902654868,
      "grad_norm": 5.1866233974351417e-08,
      "learning_rate": 0.009418655373501483,
      "loss": 0.0,
      "num_input_tokens_seen": 20144792,
      "step": 35465
    },
    {
      "epoch": 622.2831858407079,
      "grad_norm": 5.844966821655362e-08,
      "learning_rate": 0.00939812206249851,
      "loss": 0.0,
      "num_input_tokens_seen": 20147560,
      "step": 35470
    },
    {
      "epoch": 622.3716814159292,
      "grad_norm": 3.7143713882414886e-08,
      "learning_rate": 0.009377610434072004,
      "loss": 0.0,
      "num_input_tokens_seen": 20150424,
      "step": 35475
    },
    {
      "epoch": 622.4601769911504,
      "grad_norm": 3.593940078872038e-08,
      "learning_rate": 0.009357120491385167,
      "loss": 0.0,
      "num_input_tokens_seen": 20152920,
      "step": 35480
    },
    {
      "epoch": 622.5486725663717,
      "grad_norm": 7.042199001716654e-08,
      "learning_rate": 0.009336652237597743,
      "loss": 0.0,
      "num_input_tokens_seen": 20156088,
      "step": 35485
    },
    {
      "epoch": 622.637168141593,
      "grad_norm": 1.0109066295171942e-07,
      "learning_rate": 0.009316205675866251,
      "loss": 0.0,
      "num_input_tokens_seen": 20158776,
      "step": 35490
    },
    {
      "epoch": 622.7256637168142,
      "grad_norm": 2.8772509708119287e-08,
      "learning_rate": 0.00929578080934379,
      "loss": 0.0,
      "num_input_tokens_seen": 20161576,
      "step": 35495
    },
    {
      "epoch": 622.8141592920354,
      "grad_norm": 5.3139267208734964e-08,
      "learning_rate": 0.00927537764118012,
      "loss": 0.0,
      "num_input_tokens_seen": 20164376,
      "step": 35500
    },
    {
      "epoch": 622.9026548672566,
      "grad_norm": 5.6552266869402956e-08,
      "learning_rate": 0.009254996174521678,
      "loss": 0.0,
      "num_input_tokens_seen": 20167736,
      "step": 35505
    },
    {
      "epoch": 622.9911504424779,
      "grad_norm": 2.1567437613612128e-08,
      "learning_rate": 0.009234636412511531,
      "loss": 0.0,
      "num_input_tokens_seen": 20170728,
      "step": 35510
    },
    {
      "epoch": 623.070796460177,
      "grad_norm": 3.3073956018370154e-08,
      "learning_rate": 0.009214298358289418,
      "loss": 0.0,
      "num_input_tokens_seen": 20173000,
      "step": 35515
    },
    {
      "epoch": 623.1592920353983,
      "grad_norm": 2.995533421312757e-08,
      "learning_rate": 0.00919398201499173,
      "loss": 0.0,
      "num_input_tokens_seen": 20175752,
      "step": 35520
    },
    {
      "epoch": 623.2477876106195,
      "grad_norm": 6.778059713496987e-08,
      "learning_rate": 0.009173687385751495,
      "loss": 0.0,
      "num_input_tokens_seen": 20178216,
      "step": 35525
    },
    {
      "epoch": 623.3362831858407,
      "grad_norm": 2.55919410108163e-08,
      "learning_rate": 0.009153414473698407,
      "loss": 0.0,
      "num_input_tokens_seen": 20180904,
      "step": 35530
    },
    {
      "epoch": 623.4247787610619,
      "grad_norm": 5.2506500480831164e-08,
      "learning_rate": 0.009133163281958784,
      "loss": 0.0,
      "num_input_tokens_seen": 20183448,
      "step": 35535
    },
    {
      "epoch": 623.5132743362832,
      "grad_norm": 2.921013653178761e-08,
      "learning_rate": 0.009112933813655627,
      "loss": 0.0,
      "num_input_tokens_seen": 20186696,
      "step": 35540
    },
    {
      "epoch": 623.6017699115044,
      "grad_norm": 2.7071591190974686e-08,
      "learning_rate": 0.009092726071908573,
      "loss": 0.0,
      "num_input_tokens_seen": 20189512,
      "step": 35545
    },
    {
      "epoch": 623.6902654867257,
      "grad_norm": 1.6254396939530125e-08,
      "learning_rate": 0.0090725400598339,
      "loss": 0.0,
      "num_input_tokens_seen": 20192360,
      "step": 35550
    },
    {
      "epoch": 623.7787610619469,
      "grad_norm": 3.782583846145826e-08,
      "learning_rate": 0.009052375780544563,
      "loss": 0.0,
      "num_input_tokens_seen": 20195640,
      "step": 35555
    },
    {
      "epoch": 623.8672566371681,
      "grad_norm": 4.152096622078716e-08,
      "learning_rate": 0.009032233237150144,
      "loss": 0.0,
      "num_input_tokens_seen": 20198856,
      "step": 35560
    },
    {
      "epoch": 623.9557522123894,
      "grad_norm": 4.2623188534207657e-08,
      "learning_rate": 0.009012112432756875,
      "loss": 0.0,
      "num_input_tokens_seen": 20201464,
      "step": 35565
    },
    {
      "epoch": 624.0353982300885,
      "grad_norm": 2.980875635216762e-08,
      "learning_rate": 0.008992013370467605,
      "loss": 0.0,
      "num_input_tokens_seen": 20204120,
      "step": 35570
    },
    {
      "epoch": 624.1238938053098,
      "grad_norm": 3.719810237612364e-08,
      "learning_rate": 0.008971936053381924,
      "loss": 0.0,
      "num_input_tokens_seen": 20206728,
      "step": 35575
    },
    {
      "epoch": 624.212389380531,
      "grad_norm": 3.724227681800585e-08,
      "learning_rate": 0.008951880484595953,
      "loss": 0.0,
      "num_input_tokens_seen": 20209912,
      "step": 35580
    },
    {
      "epoch": 624.3008849557522,
      "grad_norm": 4.57954101307223e-08,
      "learning_rate": 0.008931846667202552,
      "loss": 0.0,
      "num_input_tokens_seen": 20212520,
      "step": 35585
    },
    {
      "epoch": 624.3893805309734,
      "grad_norm": 1.4452055552283127e-08,
      "learning_rate": 0.008911834604291152,
      "loss": 0.0,
      "num_input_tokens_seen": 20215464,
      "step": 35590
    },
    {
      "epoch": 624.4778761061947,
      "grad_norm": 4.0956809499448354e-08,
      "learning_rate": 0.008891844298947882,
      "loss": 0.0,
      "num_input_tokens_seen": 20218184,
      "step": 35595
    },
    {
      "epoch": 624.566371681416,
      "grad_norm": 4.347544191318775e-08,
      "learning_rate": 0.008871875754255508,
      "loss": 0.0,
      "num_input_tokens_seen": 20220888,
      "step": 35600
    },
    {
      "epoch": 624.566371681416,
      "eval_loss": 0.8094174861907959,
      "eval_runtime": 0.9174,
      "eval_samples_per_second": 27.25,
      "eval_steps_per_second": 14.17,
      "num_input_tokens_seen": 20220888,
      "step": 35600
    },
    {
      "epoch": 624.6548672566372,
      "grad_norm": 3.153455097049118e-08,
      "learning_rate": 0.008851928973293422,
      "loss": 0.0,
      "num_input_tokens_seen": 20224344,
      "step": 35605
    },
    {
      "epoch": 624.7433628318585,
      "grad_norm": 1.0101405223394977e-07,
      "learning_rate": 0.00883200395913764,
      "loss": 0.0,
      "num_input_tokens_seen": 20226984,
      "step": 35610
    },
    {
      "epoch": 624.8318584070796,
      "grad_norm": 2.511393226711789e-08,
      "learning_rate": 0.00881210071486091,
      "loss": 0.0,
      "num_input_tokens_seen": 20229896,
      "step": 35615
    },
    {
      "epoch": 624.9203539823009,
      "grad_norm": 8.639434412316405e-08,
      "learning_rate": 0.008792219243532505,
      "loss": 0.0,
      "num_input_tokens_seen": 20232712,
      "step": 35620
    },
    {
      "epoch": 625.0,
      "grad_norm": 7.420732117680018e-08,
      "learning_rate": 0.008772359548218428,
      "loss": 0.0,
      "num_input_tokens_seen": 20235192,
      "step": 35625
    },
    {
      "epoch": 625.0884955752213,
      "grad_norm": 5.356084287200247e-08,
      "learning_rate": 0.008752521631981274,
      "loss": 0.0,
      "num_input_tokens_seen": 20237512,
      "step": 35630
    },
    {
      "epoch": 625.1769911504425,
      "grad_norm": 2.1732088129056137e-08,
      "learning_rate": 0.008732705497880315,
      "loss": 0.0,
      "num_input_tokens_seen": 20240280,
      "step": 35635
    },
    {
      "epoch": 625.2654867256637,
      "grad_norm": 4.843345280391986e-08,
      "learning_rate": 0.008712911148971459,
      "loss": 0.0,
      "num_input_tokens_seen": 20243192,
      "step": 35640
    },
    {
      "epoch": 625.3539823008849,
      "grad_norm": 2.8718343259015455e-08,
      "learning_rate": 0.008693138588307208,
      "loss": 0.0,
      "num_input_tokens_seen": 20246152,
      "step": 35645
    },
    {
      "epoch": 625.4424778761062,
      "grad_norm": 2.000305521221435e-08,
      "learning_rate": 0.008673387818936762,
      "loss": 0.0,
      "num_input_tokens_seen": 20248936,
      "step": 35650
    },
    {
      "epoch": 625.5309734513274,
      "grad_norm": 3.401690662485635e-08,
      "learning_rate": 0.008653658843905948,
      "loss": 0.0,
      "num_input_tokens_seen": 20251816,
      "step": 35655
    },
    {
      "epoch": 625.6194690265487,
      "grad_norm": 2.207826099720478e-08,
      "learning_rate": 0.0086339516662572,
      "loss": 0.0,
      "num_input_tokens_seen": 20254776,
      "step": 35660
    },
    {
      "epoch": 625.70796460177,
      "grad_norm": 1.5731281166608824e-08,
      "learning_rate": 0.008614266289029638,
      "loss": 0.0,
      "num_input_tokens_seen": 20257576,
      "step": 35665
    },
    {
      "epoch": 625.7964601769911,
      "grad_norm": 1.4142019111318405e-08,
      "learning_rate": 0.008594602715258965,
      "loss": 0.0,
      "num_input_tokens_seen": 20260120,
      "step": 35670
    },
    {
      "epoch": 625.8849557522124,
      "grad_norm": 2.546413924164881e-08,
      "learning_rate": 0.008574960947977573,
      "loss": 0.0,
      "num_input_tokens_seen": 20263528,
      "step": 35675
    },
    {
      "epoch": 625.9734513274336,
      "grad_norm": 6.332965085675824e-08,
      "learning_rate": 0.008555340990214438,
      "loss": 0.0,
      "num_input_tokens_seen": 20266504,
      "step": 35680
    },
    {
      "epoch": 626.0530973451328,
      "grad_norm": 2.920368125103323e-08,
      "learning_rate": 0.008535742844995258,
      "loss": 0.0,
      "num_input_tokens_seen": 20269536,
      "step": 35685
    },
    {
      "epoch": 626.141592920354,
      "grad_norm": 6.489102588602691e-08,
      "learning_rate": 0.008516166515342266,
      "loss": 0.0,
      "num_input_tokens_seen": 20272560,
      "step": 35690
    },
    {
      "epoch": 626.2300884955753,
      "grad_norm": 4.927399288590095e-08,
      "learning_rate": 0.008496612004274411,
      "loss": 0.0,
      "num_input_tokens_seen": 20275440,
      "step": 35695
    },
    {
      "epoch": 626.3185840707964,
      "grad_norm": 5.51581216257091e-08,
      "learning_rate": 0.008477079314807201,
      "loss": 0.0,
      "num_input_tokens_seen": 20278272,
      "step": 35700
    },
    {
      "epoch": 626.4070796460177,
      "grad_norm": 2.2516957187690423e-08,
      "learning_rate": 0.008457568449952874,
      "loss": 0.0,
      "num_input_tokens_seen": 20280976,
      "step": 35705
    },
    {
      "epoch": 626.4955752212389,
      "grad_norm": 4.316599344633687e-08,
      "learning_rate": 0.008438079412720189,
      "loss": 0.0,
      "num_input_tokens_seen": 20283744,
      "step": 35710
    },
    {
      "epoch": 626.5840707964602,
      "grad_norm": 4.1895265923130864e-08,
      "learning_rate": 0.00841861220611466,
      "loss": 0.0,
      "num_input_tokens_seen": 20286256,
      "step": 35715
    },
    {
      "epoch": 626.6725663716815,
      "grad_norm": 9.099159825609604e-08,
      "learning_rate": 0.008399166833138355,
      "loss": 0.0,
      "num_input_tokens_seen": 20288976,
      "step": 35720
    },
    {
      "epoch": 626.7610619469026,
      "grad_norm": 3.074380572343216e-08,
      "learning_rate": 0.008379743296789987,
      "loss": 0.0,
      "num_input_tokens_seen": 20292032,
      "step": 35725
    },
    {
      "epoch": 626.8495575221239,
      "grad_norm": 1.9826773112185947e-08,
      "learning_rate": 0.008360341600064896,
      "loss": 0.0,
      "num_input_tokens_seen": 20294672,
      "step": 35730
    },
    {
      "epoch": 626.9380530973451,
      "grad_norm": 3.884901289552545e-08,
      "learning_rate": 0.008340961745955121,
      "loss": 0.0,
      "num_input_tokens_seen": 20297520,
      "step": 35735
    },
    {
      "epoch": 627.0176991150443,
      "grad_norm": 3.1170625192089574e-08,
      "learning_rate": 0.008321603737449224,
      "loss": 0.0,
      "num_input_tokens_seen": 20300392,
      "step": 35740
    },
    {
      "epoch": 627.1061946902655,
      "grad_norm": 4.406467368767153e-08,
      "learning_rate": 0.008302267577532479,
      "loss": 0.0,
      "num_input_tokens_seen": 20303432,
      "step": 35745
    },
    {
      "epoch": 627.1946902654868,
      "grad_norm": 3.586815111589203e-08,
      "learning_rate": 0.008282953269186771,
      "loss": 0.0,
      "num_input_tokens_seen": 20306616,
      "step": 35750
    },
    {
      "epoch": 627.2831858407079,
      "grad_norm": 4.4431541112999184e-08,
      "learning_rate": 0.008263660815390567,
      "loss": 0.0,
      "num_input_tokens_seen": 20309384,
      "step": 35755
    },
    {
      "epoch": 627.3716814159292,
      "grad_norm": 2.1609176670267516e-08,
      "learning_rate": 0.008244390219119069,
      "loss": 0.0,
      "num_input_tokens_seen": 20312280,
      "step": 35760
    },
    {
      "epoch": 627.4601769911504,
      "grad_norm": 3.723109642805866e-08,
      "learning_rate": 0.008225141483343967,
      "loss": 0.0,
      "num_input_tokens_seen": 20314664,
      "step": 35765
    },
    {
      "epoch": 627.5486725663717,
      "grad_norm": 5.4024980045142e-08,
      "learning_rate": 0.00820591461103372,
      "loss": 0.0,
      "num_input_tokens_seen": 20317512,
      "step": 35770
    },
    {
      "epoch": 627.637168141593,
      "grad_norm": 4.8130448959682326e-08,
      "learning_rate": 0.008186709605153358,
      "loss": 0.0,
      "num_input_tokens_seen": 20321016,
      "step": 35775
    },
    {
      "epoch": 627.7256637168142,
      "grad_norm": 2.0222481467158104e-08,
      "learning_rate": 0.008167526468664492,
      "loss": 0.0,
      "num_input_tokens_seen": 20323928,
      "step": 35780
    },
    {
      "epoch": 627.8141592920354,
      "grad_norm": 3.620104394030932e-08,
      "learning_rate": 0.008148365204525443,
      "loss": 0.0,
      "num_input_tokens_seen": 20326360,
      "step": 35785
    },
    {
      "epoch": 627.9026548672566,
      "grad_norm": 5.8218947884824956e-08,
      "learning_rate": 0.00812922581569106,
      "loss": 0.0,
      "num_input_tokens_seen": 20329016,
      "step": 35790
    },
    {
      "epoch": 627.9911504424779,
      "grad_norm": 5.059553132014116e-08,
      "learning_rate": 0.008110108305112934,
      "loss": 0.0,
      "num_input_tokens_seen": 20331704,
      "step": 35795
    },
    {
      "epoch": 628.070796460177,
      "grad_norm": 6.420307130383662e-08,
      "learning_rate": 0.008091012675739223,
      "loss": 0.0,
      "num_input_tokens_seen": 20333904,
      "step": 35800
    },
    {
      "epoch": 628.070796460177,
      "eval_loss": 0.8172165751457214,
      "eval_runtime": 0.9258,
      "eval_samples_per_second": 27.002,
      "eval_steps_per_second": 14.041,
      "num_input_tokens_seen": 20333904,
      "step": 35800
    },
    {
      "epoch": 628.1592920353983,
      "grad_norm": 5.507838807261578e-08,
      "learning_rate": 0.008071938930514671,
      "loss": 0.0,
      "num_input_tokens_seen": 20337120,
      "step": 35805
    },
    {
      "epoch": 628.2477876106195,
      "grad_norm": 7.193754214540604e-08,
      "learning_rate": 0.008052887072380726,
      "loss": 0.0,
      "num_input_tokens_seen": 20339840,
      "step": 35810
    },
    {
      "epoch": 628.3362831858407,
      "grad_norm": 5.749597420390273e-08,
      "learning_rate": 0.008033857104275437,
      "loss": 0.0,
      "num_input_tokens_seen": 20342368,
      "step": 35815
    },
    {
      "epoch": 628.4247787610619,
      "grad_norm": 2.355075956472774e-08,
      "learning_rate": 0.008014849029133424,
      "loss": 0.0,
      "num_input_tokens_seen": 20344928,
      "step": 35820
    },
    {
      "epoch": 628.5132743362832,
      "grad_norm": 6.004756158972668e-08,
      "learning_rate": 0.007995862849885975,
      "loss": 0.0,
      "num_input_tokens_seen": 20347616,
      "step": 35825
    },
    {
      "epoch": 628.6017699115044,
      "grad_norm": 5.076590170460804e-08,
      "learning_rate": 0.007976898569461032,
      "loss": 0.0,
      "num_input_tokens_seen": 20350736,
      "step": 35830
    },
    {
      "epoch": 628.6902654867257,
      "grad_norm": 6.563828236494373e-08,
      "learning_rate": 0.007957956190783088,
      "loss": 0.0,
      "num_input_tokens_seen": 20354064,
      "step": 35835
    },
    {
      "epoch": 628.7787610619469,
      "grad_norm": 5.243937550858391e-08,
      "learning_rate": 0.007939035716773324,
      "loss": 0.0,
      "num_input_tokens_seen": 20356624,
      "step": 35840
    },
    {
      "epoch": 628.8672566371681,
      "grad_norm": 5.846408157594851e-08,
      "learning_rate": 0.007920137150349487,
      "loss": 0.0,
      "num_input_tokens_seen": 20359232,
      "step": 35845
    },
    {
      "epoch": 628.9557522123894,
      "grad_norm": 3.7002585884238215e-08,
      "learning_rate": 0.007901260494425981,
      "loss": 0.0,
      "num_input_tokens_seen": 20362432,
      "step": 35850
    },
    {
      "epoch": 629.0353982300885,
      "grad_norm": 1.683111250372349e-08,
      "learning_rate": 0.007882405751913861,
      "loss": 0.0,
      "num_input_tokens_seen": 20364664,
      "step": 35855
    },
    {
      "epoch": 629.1238938053098,
      "grad_norm": 3.6837107586507045e-08,
      "learning_rate": 0.007863572925720702,
      "loss": 0.0,
      "num_input_tokens_seen": 20367576,
      "step": 35860
    },
    {
      "epoch": 629.212389380531,
      "grad_norm": 2.186189718145215e-08,
      "learning_rate": 0.007844762018750827,
      "loss": 0.0,
      "num_input_tokens_seen": 20370120,
      "step": 35865
    },
    {
      "epoch": 629.3008849557522,
      "grad_norm": 5.7720146884321366e-08,
      "learning_rate": 0.007825973033905054,
      "loss": 0.0,
      "num_input_tokens_seen": 20373448,
      "step": 35870
    },
    {
      "epoch": 629.3893805309734,
      "grad_norm": 3.829319439319079e-08,
      "learning_rate": 0.007807205974080927,
      "loss": 0.0,
      "num_input_tokens_seen": 20376504,
      "step": 35875
    },
    {
      "epoch": 629.4778761061947,
      "grad_norm": 5.5878039972867555e-08,
      "learning_rate": 0.007788460842172551,
      "loss": 0.0,
      "num_input_tokens_seen": 20379592,
      "step": 35880
    },
    {
      "epoch": 629.566371681416,
      "grad_norm": 6.426935073022833e-08,
      "learning_rate": 0.0077697376410706285,
      "loss": 0.0,
      "num_input_tokens_seen": 20382168,
      "step": 35885
    },
    {
      "epoch": 629.6548672566372,
      "grad_norm": 3.922755098528796e-08,
      "learning_rate": 0.007751036373662567,
      "loss": 0.0,
      "num_input_tokens_seen": 20385176,
      "step": 35890
    },
    {
      "epoch": 629.7433628318585,
      "grad_norm": 5.2522608484650846e-08,
      "learning_rate": 0.00773235704283231,
      "loss": 0.0,
      "num_input_tokens_seen": 20388248,
      "step": 35895
    },
    {
      "epoch": 629.8318584070796,
      "grad_norm": 6.217045012135713e-08,
      "learning_rate": 0.007713699651460437,
      "loss": 0.0,
      "num_input_tokens_seen": 20390744,
      "step": 35900
    },
    {
      "epoch": 629.9203539823009,
      "grad_norm": 3.572539952756415e-08,
      "learning_rate": 0.007695064202424162,
      "loss": 0.0,
      "num_input_tokens_seen": 20393832,
      "step": 35905
    },
    {
      "epoch": 630.0,
      "grad_norm": 1.321115661312433e-07,
      "learning_rate": 0.007676450698597286,
      "loss": 0.0,
      "num_input_tokens_seen": 20395800,
      "step": 35910
    },
    {
      "epoch": 630.0884955752213,
      "grad_norm": 2.6222521043450797e-08,
      "learning_rate": 0.007657859142850265,
      "loss": 0.0,
      "num_input_tokens_seen": 20398424,
      "step": 35915
    },
    {
      "epoch": 630.1769911504425,
      "grad_norm": 2.0579996373726317e-08,
      "learning_rate": 0.0076392895380501535,
      "loss": 0.0,
      "num_input_tokens_seen": 20401128,
      "step": 35920
    },
    {
      "epoch": 630.2654867256637,
      "grad_norm": 2.975755108991507e-08,
      "learning_rate": 0.007620741887060611,
      "loss": 0.0,
      "num_input_tokens_seen": 20403816,
      "step": 35925
    },
    {
      "epoch": 630.3539823008849,
      "grad_norm": 1.7064756718809804e-08,
      "learning_rate": 0.007602216192741901,
      "loss": 0.0,
      "num_input_tokens_seen": 20406408,
      "step": 35930
    },
    {
      "epoch": 630.4424778761062,
      "grad_norm": 2.2949583566855836e-08,
      "learning_rate": 0.007583712457950969,
      "loss": 0.0,
      "num_input_tokens_seen": 20409496,
      "step": 35935
    },
    {
      "epoch": 630.5309734513274,
      "grad_norm": 2.6943999031914245e-08,
      "learning_rate": 0.007565230685541269,
      "loss": 0.0,
      "num_input_tokens_seen": 20412616,
      "step": 35940
    },
    {
      "epoch": 630.6194690265487,
      "grad_norm": 6.74678943823892e-08,
      "learning_rate": 0.007546770878362968,
      "loss": 0.0,
      "num_input_tokens_seen": 20415896,
      "step": 35945
    },
    {
      "epoch": 630.70796460177,
      "grad_norm": 2.810614851966875e-08,
      "learning_rate": 0.0075283330392627405,
      "loss": 0.0,
      "num_input_tokens_seen": 20418376,
      "step": 35950
    },
    {
      "epoch": 630.7964601769911,
      "grad_norm": 5.6004864745773375e-08,
      "learning_rate": 0.007509917171083979,
      "loss": 0.0,
      "num_input_tokens_seen": 20421608,
      "step": 35955
    },
    {
      "epoch": 630.8849557522124,
      "grad_norm": 3.7332029023673385e-08,
      "learning_rate": 0.007491523276666662,
      "loss": 0.0,
      "num_input_tokens_seen": 20424360,
      "step": 35960
    },
    {
      "epoch": 630.9734513274336,
      "grad_norm": 4.374618711722178e-08,
      "learning_rate": 0.007473151358847318,
      "loss": 0.0,
      "num_input_tokens_seen": 20427240,
      "step": 35965
    },
    {
      "epoch": 631.0530973451328,
      "grad_norm": 4.572062906049723e-08,
      "learning_rate": 0.007454801420459117,
      "loss": 0.0,
      "num_input_tokens_seen": 20429808,
      "step": 35970
    },
    {
      "epoch": 631.141592920354,
      "grad_norm": 3.8703774407622404e-08,
      "learning_rate": 0.0074364734643319105,
      "loss": 0.0,
      "num_input_tokens_seen": 20433088,
      "step": 35975
    },
    {
      "epoch": 631.2300884955753,
      "grad_norm": 1.8370185372873493e-08,
      "learning_rate": 0.007418167493292022,
      "loss": 0.0,
      "num_input_tokens_seen": 20435744,
      "step": 35980
    },
    {
      "epoch": 631.3185840707964,
      "grad_norm": 6.841931821099934e-08,
      "learning_rate": 0.0073998835101625245,
      "loss": 0.0,
      "num_input_tokens_seen": 20438528,
      "step": 35985
    },
    {
      "epoch": 631.4070796460177,
      "grad_norm": 4.119818797221342e-08,
      "learning_rate": 0.007381621517762998,
      "loss": 0.0,
      "num_input_tokens_seen": 20441504,
      "step": 35990
    },
    {
      "epoch": 631.4955752212389,
      "grad_norm": 5.981234352248066e-08,
      "learning_rate": 0.007363381518909689,
      "loss": 0.0,
      "num_input_tokens_seen": 20443952,
      "step": 35995
    },
    {
      "epoch": 631.5840707964602,
      "grad_norm": 1.4563193317940204e-08,
      "learning_rate": 0.007345163516415448,
      "loss": 0.0,
      "num_input_tokens_seen": 20446736,
      "step": 36000
    },
    {
      "epoch": 631.5840707964602,
      "eval_loss": 0.7858958840370178,
      "eval_runtime": 0.9419,
      "eval_samples_per_second": 26.542,
      "eval_steps_per_second": 13.802,
      "num_input_tokens_seen": 20446736,
      "step": 36000
    },
    {
      "epoch": 631.6725663716815,
      "grad_norm": 7.12771139887991e-08,
      "learning_rate": 0.007326967513089693,
      "loss": 0.0,
      "num_input_tokens_seen": 20449360,
      "step": 36005
    },
    {
      "epoch": 631.7610619469026,
      "grad_norm": 2.6567816391320775e-08,
      "learning_rate": 0.0073087935117384815,
      "loss": 0.0,
      "num_input_tokens_seen": 20452112,
      "step": 36010
    },
    {
      "epoch": 631.8495575221239,
      "grad_norm": 2.09283577135011e-08,
      "learning_rate": 0.007290641515164503,
      "loss": 0.0,
      "num_input_tokens_seen": 20455136,
      "step": 36015
    },
    {
      "epoch": 631.9380530973451,
      "grad_norm": 4.701249878280578e-08,
      "learning_rate": 0.007272511526166986,
      "loss": 0.0,
      "num_input_tokens_seen": 20458336,
      "step": 36020
    },
    {
      "epoch": 632.0176991150443,
      "grad_norm": 7.515038191741041e-08,
      "learning_rate": 0.0072544035475418265,
      "loss": 0.0,
      "num_input_tokens_seen": 20460712,
      "step": 36025
    },
    {
      "epoch": 632.1061946902655,
      "grad_norm": 2.0163794189898e-08,
      "learning_rate": 0.007236317582081475,
      "loss": 0.0,
      "num_input_tokens_seen": 20463080,
      "step": 36030
    },
    {
      "epoch": 632.1946902654868,
      "grad_norm": 3.0936693207195276e-08,
      "learning_rate": 0.007218253632575066,
      "loss": 0.0,
      "num_input_tokens_seen": 20465736,
      "step": 36035
    },
    {
      "epoch": 632.2831858407079,
      "grad_norm": 5.13783469102691e-08,
      "learning_rate": 0.007200211701808223,
      "loss": 0.0,
      "num_input_tokens_seen": 20468808,
      "step": 36040
    },
    {
      "epoch": 632.3716814159292,
      "grad_norm": 3.802066217417632e-08,
      "learning_rate": 0.007182191792563286,
      "loss": 0.0,
      "num_input_tokens_seen": 20471240,
      "step": 36045
    },
    {
      "epoch": 632.4601769911504,
      "grad_norm": 2.037796242859713e-08,
      "learning_rate": 0.0071641939076191145,
      "loss": 0.0,
      "num_input_tokens_seen": 20474536,
      "step": 36050
    },
    {
      "epoch": 632.5486725663717,
      "grad_norm": 6.437270627657199e-08,
      "learning_rate": 0.007146218049751257,
      "loss": 0.0,
      "num_input_tokens_seen": 20477512,
      "step": 36055
    },
    {
      "epoch": 632.637168141593,
      "grad_norm": 3.494285749638948e-08,
      "learning_rate": 0.0071282642217317775,
      "loss": 0.0,
      "num_input_tokens_seen": 20480408,
      "step": 36060
    },
    {
      "epoch": 632.7256637168142,
      "grad_norm": 1.5544388887178684e-08,
      "learning_rate": 0.007110332426329396,
      "loss": 0.0,
      "num_input_tokens_seen": 20483864,
      "step": 36065
    },
    {
      "epoch": 632.8141592920354,
      "grad_norm": 1.0496570723717014e-08,
      "learning_rate": 0.007092422666309417,
      "loss": 0.0,
      "num_input_tokens_seen": 20486632,
      "step": 36070
    },
    {
      "epoch": 632.9026548672566,
      "grad_norm": 4.546360088397705e-08,
      "learning_rate": 0.0070745349444337295,
      "loss": 0.0,
      "num_input_tokens_seen": 20489480,
      "step": 36075
    },
    {
      "epoch": 632.9911504424779,
      "grad_norm": 4.0464701811515624e-08,
      "learning_rate": 0.007056669263460913,
      "loss": 0.0,
      "num_input_tokens_seen": 20492392,
      "step": 36080
    },
    {
      "epoch": 633.070796460177,
      "grad_norm": 4.92800680262917e-08,
      "learning_rate": 0.007038825626145995,
      "loss": 0.0,
      "num_input_tokens_seen": 20495352,
      "step": 36085
    },
    {
      "epoch": 633.1592920353983,
      "grad_norm": 6.532734175834776e-08,
      "learning_rate": 0.007021004035240724,
      "loss": 0.0,
      "num_input_tokens_seen": 20498184,
      "step": 36090
    },
    {
      "epoch": 633.2477876106195,
      "grad_norm": 4.218840743419605e-08,
      "learning_rate": 0.007003204493493453,
      "loss": 0.0,
      "num_input_tokens_seen": 20500760,
      "step": 36095
    },
    {
      "epoch": 633.3362831858407,
      "grad_norm": 4.423942456810437e-08,
      "learning_rate": 0.006985427003649036,
      "loss": 0.0,
      "num_input_tokens_seen": 20503464,
      "step": 36100
    },
    {
      "epoch": 633.4247787610619,
      "grad_norm": 2.0102675080124754e-08,
      "learning_rate": 0.006967671568449013,
      "loss": 0.0,
      "num_input_tokens_seen": 20506168,
      "step": 36105
    },
    {
      "epoch": 633.5132743362832,
      "grad_norm": 3.36087957464315e-08,
      "learning_rate": 0.006949938190631511,
      "loss": 0.0,
      "num_input_tokens_seen": 20509784,
      "step": 36110
    },
    {
      "epoch": 633.6017699115044,
      "grad_norm": 7.435996707272352e-08,
      "learning_rate": 0.0069322268729311905,
      "loss": 0.0,
      "num_input_tokens_seen": 20512824,
      "step": 36115
    },
    {
      "epoch": 633.6902654867257,
      "grad_norm": 1.3678011612228147e-07,
      "learning_rate": 0.006914537618079403,
      "loss": 0.0,
      "num_input_tokens_seen": 20515640,
      "step": 36120
    },
    {
      "epoch": 633.7787610619469,
      "grad_norm": 1.4579753404575513e-08,
      "learning_rate": 0.006896870428804031,
      "loss": 0.0,
      "num_input_tokens_seen": 20518248,
      "step": 36125
    },
    {
      "epoch": 633.8672566371681,
      "grad_norm": 3.5141603405008937e-08,
      "learning_rate": 0.006879225307829595,
      "loss": 0.0,
      "num_input_tokens_seen": 20520824,
      "step": 36130
    },
    {
      "epoch": 633.9557522123894,
      "grad_norm": 3.6077253184885194e-08,
      "learning_rate": 0.00686160225787717,
      "loss": 0.0,
      "num_input_tokens_seen": 20523560,
      "step": 36135
    },
    {
      "epoch": 634.0353982300885,
      "grad_norm": 2.2600390892080213e-08,
      "learning_rate": 0.006844001281664463,
      "loss": 0.0,
      "num_input_tokens_seen": 20526128,
      "step": 36140
    },
    {
      "epoch": 634.1238938053098,
      "grad_norm": 2.9158233161297176e-08,
      "learning_rate": 0.006826422381905789,
      "loss": 0.0,
      "num_input_tokens_seen": 20528576,
      "step": 36145
    },
    {
      "epoch": 634.212389380531,
      "grad_norm": 2.156571277112107e-08,
      "learning_rate": 0.006808865561311994,
      "loss": 0.0,
      "num_input_tokens_seen": 20531360,
      "step": 36150
    },
    {
      "epoch": 634.3008849557522,
      "grad_norm": 2.795634301833161e-08,
      "learning_rate": 0.00679133082259058,
      "loss": 0.0,
      "num_input_tokens_seen": 20534064,
      "step": 36155
    },
    {
      "epoch": 634.3893805309734,
      "grad_norm": 3.473655496577521e-08,
      "learning_rate": 0.00677381816844565,
      "loss": 0.0,
      "num_input_tokens_seen": 20537040,
      "step": 36160
    },
    {
      "epoch": 634.4778761061947,
      "grad_norm": 2.2194667437247517e-08,
      "learning_rate": 0.0067563276015778434,
      "loss": 0.0,
      "num_input_tokens_seen": 20540064,
      "step": 36165
    },
    {
      "epoch": 634.566371681416,
      "grad_norm": 4.151344157321546e-08,
      "learning_rate": 0.006738859124684437,
      "loss": 0.0,
      "num_input_tokens_seen": 20542720,
      "step": 36170
    },
    {
      "epoch": 634.6548672566372,
      "grad_norm": 9.32192918412511e-08,
      "learning_rate": 0.006721412740459259,
      "loss": 0.0,
      "num_input_tokens_seen": 20545760,
      "step": 36175
    },
    {
      "epoch": 634.7433628318585,
      "grad_norm": 4.1363350078427175e-08,
      "learning_rate": 0.006703988451592824,
      "loss": 0.0,
      "num_input_tokens_seen": 20548464,
      "step": 36180
    },
    {
      "epoch": 634.8318584070796,
      "grad_norm": 1.1977131286755593e-08,
      "learning_rate": 0.006686586260772114,
      "loss": 0.0,
      "num_input_tokens_seen": 20551200,
      "step": 36185
    },
    {
      "epoch": 634.9203539823009,
      "grad_norm": 4.330090774828932e-08,
      "learning_rate": 0.006669206170680819,
      "loss": 0.0,
      "num_input_tokens_seen": 20554720,
      "step": 36190
    },
    {
      "epoch": 635.0,
      "grad_norm": 1.3480260463438754e-07,
      "learning_rate": 0.0066518481839991095,
      "loss": 0.0,
      "num_input_tokens_seen": 20557256,
      "step": 36195
    },
    {
      "epoch": 635.0884955752213,
      "grad_norm": 8.302558285322448e-08,
      "learning_rate": 0.006634512303403861,
      "loss": 0.0,
      "num_input_tokens_seen": 20560472,
      "step": 36200
    },
    {
      "epoch": 635.0884955752213,
      "eval_loss": 0.8261749744415283,
      "eval_runtime": 0.9396,
      "eval_samples_per_second": 26.606,
      "eval_steps_per_second": 13.835,
      "num_input_tokens_seen": 20560472,
      "step": 36200
    },
    {
      "epoch": 635.1769911504425,
      "grad_norm": 5.6968847417238067e-08,
      "learning_rate": 0.0066171985315684355,
      "loss": 0.0,
      "num_input_tokens_seen": 20563496,
      "step": 36205
    },
    {
      "epoch": 635.2654867256637,
      "grad_norm": 2.320011027734381e-08,
      "learning_rate": 0.0065999068711628806,
      "loss": 0.0,
      "num_input_tokens_seen": 20566216,
      "step": 36210
    },
    {
      "epoch": 635.3539823008849,
      "grad_norm": 3.72940185400239e-08,
      "learning_rate": 0.0065826373248537295,
      "loss": 0.0,
      "num_input_tokens_seen": 20569064,
      "step": 36215
    },
    {
      "epoch": 635.4424778761062,
      "grad_norm": 5.4479524891348774e-08,
      "learning_rate": 0.006565389895304218,
      "loss": 0.0,
      "num_input_tokens_seen": 20571896,
      "step": 36220
    },
    {
      "epoch": 635.5309734513274,
      "grad_norm": 1.0694590457660524e-07,
      "learning_rate": 0.006548164585174104,
      "loss": 0.0,
      "num_input_tokens_seen": 20574440,
      "step": 36225
    },
    {
      "epoch": 635.6194690265487,
      "grad_norm": 4.813339771203573e-08,
      "learning_rate": 0.006530961397119728,
      "loss": 0.0,
      "num_input_tokens_seen": 20577096,
      "step": 36230
    },
    {
      "epoch": 635.70796460177,
      "grad_norm": 3.248405278100108e-08,
      "learning_rate": 0.00651378033379405,
      "loss": 0.0,
      "num_input_tokens_seen": 20579912,
      "step": 36235
    },
    {
      "epoch": 635.7964601769911,
      "grad_norm": 2.991304981492249e-08,
      "learning_rate": 0.006496621397846619,
      "loss": 0.0,
      "num_input_tokens_seen": 20582664,
      "step": 36240
    },
    {
      "epoch": 635.8849557522124,
      "grad_norm": 3.336197806902419e-08,
      "learning_rate": 0.006479484591923518,
      "loss": 0.0,
      "num_input_tokens_seen": 20585496,
      "step": 36245
    },
    {
      "epoch": 635.9734513274336,
      "grad_norm": 3.866063025270705e-08,
      "learning_rate": 0.006462369918667515,
      "loss": 0.0,
      "num_input_tokens_seen": 20588536,
      "step": 36250
    },
    {
      "epoch": 636.0530973451328,
      "grad_norm": 3.4095020140512133e-08,
      "learning_rate": 0.006445277380717851,
      "loss": 0.0,
      "num_input_tokens_seen": 20591024,
      "step": 36255
    },
    {
      "epoch": 636.141592920354,
      "grad_norm": 3.5733556558170676e-08,
      "learning_rate": 0.006428206980710466,
      "loss": 0.0,
      "num_input_tokens_seen": 20593424,
      "step": 36260
    },
    {
      "epoch": 636.2300884955753,
      "grad_norm": 5.913215161967855e-08,
      "learning_rate": 0.006411158721277788,
      "loss": 0.0,
      "num_input_tokens_seen": 20596960,
      "step": 36265
    },
    {
      "epoch": 636.3185840707964,
      "grad_norm": 6.6669386455942e-08,
      "learning_rate": 0.00639413260504888,
      "loss": 0.0,
      "num_input_tokens_seen": 20599664,
      "step": 36270
    },
    {
      "epoch": 636.4070796460177,
      "grad_norm": 5.163292726706459e-08,
      "learning_rate": 0.006377128634649376,
      "loss": 0.0,
      "num_input_tokens_seen": 20602688,
      "step": 36275
    },
    {
      "epoch": 636.4955752212389,
      "grad_norm": 7.067172447250414e-08,
      "learning_rate": 0.006360146812701528,
      "loss": 0.0,
      "num_input_tokens_seen": 20605920,
      "step": 36280
    },
    {
      "epoch": 636.5840707964602,
      "grad_norm": 3.2114883197209565e-08,
      "learning_rate": 0.006343187141824125,
      "loss": 0.0,
      "num_input_tokens_seen": 20608800,
      "step": 36285
    },
    {
      "epoch": 636.6725663716815,
      "grad_norm": 6.1544611185127e-08,
      "learning_rate": 0.00632624962463259,
      "loss": 0.0,
      "num_input_tokens_seen": 20611552,
      "step": 36290
    },
    {
      "epoch": 636.7610619469026,
      "grad_norm": 3.9872062984613876e-08,
      "learning_rate": 0.006309334263738853,
      "loss": 0.0,
      "num_input_tokens_seen": 20614256,
      "step": 36295
    },
    {
      "epoch": 636.8495575221239,
      "grad_norm": 4.618738103090436e-08,
      "learning_rate": 0.006292441061751508,
      "loss": 0.0,
      "num_input_tokens_seen": 20617344,
      "step": 36300
    },
    {
      "epoch": 636.9380530973451,
      "grad_norm": 1.934855120566681e-08,
      "learning_rate": 0.0062755700212757054,
      "loss": 0.0,
      "num_input_tokens_seen": 20619888,
      "step": 36305
    },
    {
      "epoch": 637.0176991150443,
      "grad_norm": 2.0250588761427935e-08,
      "learning_rate": 0.006258721144913148,
      "loss": 0.0,
      "num_input_tokens_seen": 20622528,
      "step": 36310
    },
    {
      "epoch": 637.1061946902655,
      "grad_norm": 2.6440872602506715e-08,
      "learning_rate": 0.0062418944352621575,
      "loss": 0.0,
      "num_input_tokens_seen": 20624848,
      "step": 36315
    },
    {
      "epoch": 637.1946902654868,
      "grad_norm": 3.4927612801993746e-08,
      "learning_rate": 0.0062250898949176405,
      "loss": 0.0,
      "num_input_tokens_seen": 20628256,
      "step": 36320
    },
    {
      "epoch": 637.2831858407079,
      "grad_norm": 5.3100112751280903e-08,
      "learning_rate": 0.006208307526471041,
      "loss": 0.0,
      "num_input_tokens_seen": 20631040,
      "step": 36325
    },
    {
      "epoch": 637.3716814159292,
      "grad_norm": 3.679674520640219e-08,
      "learning_rate": 0.006191547332510405,
      "loss": 0.0,
      "num_input_tokens_seen": 20634352,
      "step": 36330
    },
    {
      "epoch": 637.4601769911504,
      "grad_norm": 1.298807372762667e-08,
      "learning_rate": 0.006174809315620416,
      "loss": 0.0,
      "num_input_tokens_seen": 20637104,
      "step": 36335
    },
    {
      "epoch": 637.5486725663717,
      "grad_norm": 3.204112175581031e-08,
      "learning_rate": 0.00615809347838221,
      "loss": 0.0,
      "num_input_tokens_seen": 20640240,
      "step": 36340
    },
    {
      "epoch": 637.637168141593,
      "grad_norm": 4.9341309704686864e-08,
      "learning_rate": 0.006141399823373655,
      "loss": 0.0,
      "num_input_tokens_seen": 20643136,
      "step": 36345
    },
    {
      "epoch": 637.7256637168142,
      "grad_norm": 5.9515787853570146e-08,
      "learning_rate": 0.0061247283531690455,
      "loss": 0.0,
      "num_input_tokens_seen": 20646000,
      "step": 36350
    },
    {
      "epoch": 637.8141592920354,
      "grad_norm": 3.4425436723495295e-08,
      "learning_rate": 0.0061080790703393895,
      "loss": 0.0,
      "num_input_tokens_seen": 20648592,
      "step": 36355
    },
    {
      "epoch": 637.9026548672566,
      "grad_norm": 5.353059862045484e-08,
      "learning_rate": 0.006091451977452217,
      "loss": 0.0,
      "num_input_tokens_seen": 20651536,
      "step": 36360
    },
    {
      "epoch": 637.9911504424779,
      "grad_norm": 6.83821212987823e-08,
      "learning_rate": 0.00607484707707161,
      "loss": 0.0,
      "num_input_tokens_seen": 20654448,
      "step": 36365
    },
    {
      "epoch": 638.070796460177,
      "grad_norm": 2.4757961014643115e-08,
      "learning_rate": 0.006058264371758254,
      "loss": 0.0,
      "num_input_tokens_seen": 20657120,
      "step": 36370
    },
    {
      "epoch": 638.1592920353983,
      "grad_norm": 1.192285026263562e-08,
      "learning_rate": 0.00604170386406942,
      "loss": 0.0,
      "num_input_tokens_seen": 20659856,
      "step": 36375
    },
    {
      "epoch": 638.2477876106195,
      "grad_norm": 4.9512177469068774e-08,
      "learning_rate": 0.006025165556558931,
      "loss": 0.0,
      "num_input_tokens_seen": 20662240,
      "step": 36380
    },
    {
      "epoch": 638.3362831858407,
      "grad_norm": 6.501687010995738e-08,
      "learning_rate": 0.006008649451777248,
      "loss": 0.0,
      "num_input_tokens_seen": 20664880,
      "step": 36385
    },
    {
      "epoch": 638.4247787610619,
      "grad_norm": 4.853299984119985e-08,
      "learning_rate": 0.005992155552271283,
      "loss": 0.0,
      "num_input_tokens_seen": 20668016,
      "step": 36390
    },
    {
      "epoch": 638.5132743362832,
      "grad_norm": 5.491632038001626e-08,
      "learning_rate": 0.005975683860584685,
      "loss": 0.0,
      "num_input_tokens_seen": 20670928,
      "step": 36395
    },
    {
      "epoch": 638.6017699115044,
      "grad_norm": 6.531324459047028e-08,
      "learning_rate": 0.0059592343792575385,
      "loss": 0.0,
      "num_input_tokens_seen": 20673984,
      "step": 36400
    },
    {
      "epoch": 638.6017699115044,
      "eval_loss": 0.7735335826873779,
      "eval_runtime": 0.9708,
      "eval_samples_per_second": 25.752,
      "eval_steps_per_second": 13.391,
      "num_input_tokens_seen": 20673984,
      "step": 36400
    },
    {
      "epoch": 638.6902654867257,
      "grad_norm": 3.994686181840734e-08,
      "learning_rate": 0.0059428071108265975,
      "loss": 0.0,
      "num_input_tokens_seen": 20677056,
      "step": 36405
    },
    {
      "epoch": 638.7787610619469,
      "grad_norm": 2.5417181248599263e-08,
      "learning_rate": 0.005926402057825136,
      "loss": 0.0,
      "num_input_tokens_seen": 20680048,
      "step": 36410
    },
    {
      "epoch": 638.8672566371681,
      "grad_norm": 3.6849403528549374e-08,
      "learning_rate": 0.005910019222782997,
      "loss": 0.0,
      "num_input_tokens_seen": 20682864,
      "step": 36415
    },
    {
      "epoch": 638.9557522123894,
      "grad_norm": 5.7376698947564364e-08,
      "learning_rate": 0.005893658608226643,
      "loss": 0.0,
      "num_input_tokens_seen": 20685248,
      "step": 36420
    },
    {
      "epoch": 639.0353982300885,
      "grad_norm": 3.239420109935054e-08,
      "learning_rate": 0.0058773202166791045,
      "loss": 0.0,
      "num_input_tokens_seen": 20687624,
      "step": 36425
    },
    {
      "epoch": 639.1238938053098,
      "grad_norm": 2.6459797908273686e-08,
      "learning_rate": 0.005861004050659918,
      "loss": 0.0,
      "num_input_tokens_seen": 20689992,
      "step": 36430
    },
    {
      "epoch": 639.212389380531,
      "grad_norm": 3.983943841490145e-08,
      "learning_rate": 0.005844710112685286,
      "loss": 0.0,
      "num_input_tokens_seen": 20692520,
      "step": 36435
    },
    {
      "epoch": 639.3008849557522,
      "grad_norm": 3.250666225085297e-08,
      "learning_rate": 0.005828438405267933,
      "loss": 0.0,
      "num_input_tokens_seen": 20695176,
      "step": 36440
    },
    {
      "epoch": 639.3893805309734,
      "grad_norm": 2.6324466162463978e-08,
      "learning_rate": 0.00581218893091715,
      "loss": 0.0,
      "num_input_tokens_seen": 20698584,
      "step": 36445
    },
    {
      "epoch": 639.4778761061947,
      "grad_norm": 3.0896060820850835e-08,
      "learning_rate": 0.005795961692138801,
      "loss": 0.0,
      "num_input_tokens_seen": 20701416,
      "step": 36450
    },
    {
      "epoch": 639.566371681416,
      "grad_norm": 2.607143301247561e-08,
      "learning_rate": 0.00577975669143535,
      "loss": 0.0,
      "num_input_tokens_seen": 20704680,
      "step": 36455
    },
    {
      "epoch": 639.6548672566372,
      "grad_norm": 3.230451284252922e-08,
      "learning_rate": 0.005763573931305782,
      "loss": 0.0,
      "num_input_tokens_seen": 20707304,
      "step": 36460
    },
    {
      "epoch": 639.7433628318585,
      "grad_norm": 3.335070175580768e-08,
      "learning_rate": 0.005747413414245733,
      "loss": 0.0,
      "num_input_tokens_seen": 20710280,
      "step": 36465
    },
    {
      "epoch": 639.8318584070796,
      "grad_norm": 2.6609704661950673e-08,
      "learning_rate": 0.005731275142747294,
      "loss": 0.0,
      "num_input_tokens_seen": 20713256,
      "step": 36470
    },
    {
      "epoch": 639.9203539823009,
      "grad_norm": 2.7382389689023512e-08,
      "learning_rate": 0.005715159119299256,
      "loss": 0.0,
      "num_input_tokens_seen": 20716728,
      "step": 36475
    },
    {
      "epoch": 640.0,
      "grad_norm": 2.1588364518265735e-07,
      "learning_rate": 0.005699065346386867,
      "loss": 0.0,
      "num_input_tokens_seen": 20718920,
      "step": 36480
    },
    {
      "epoch": 640.0884955752213,
      "grad_norm": 2.6973028255383724e-08,
      "learning_rate": 0.0056829938264919885,
      "loss": 0.0,
      "num_input_tokens_seen": 20721464,
      "step": 36485
    },
    {
      "epoch": 640.1769911504425,
      "grad_norm": 2.9187951611220342e-08,
      "learning_rate": 0.005666944562093074,
      "loss": 0.0,
      "num_input_tokens_seen": 20724536,
      "step": 36490
    },
    {
      "epoch": 640.2654867256637,
      "grad_norm": 4.424877175779329e-08,
      "learning_rate": 0.005650917555665108,
      "loss": 0.0,
      "num_input_tokens_seen": 20727496,
      "step": 36495
    },
    {
      "epoch": 640.3539823008849,
      "grad_norm": 7.109500899105115e-08,
      "learning_rate": 0.005634912809679632,
      "loss": 0.0,
      "num_input_tokens_seen": 20730328,
      "step": 36500
    },
    {
      "epoch": 640.4424778761062,
      "grad_norm": 3.3947369360021185e-08,
      "learning_rate": 0.005618930326604854,
      "loss": 0.0,
      "num_input_tokens_seen": 20732840,
      "step": 36505
    },
    {
      "epoch": 640.5309734513274,
      "grad_norm": 3.100324263982657e-08,
      "learning_rate": 0.005602970108905386,
      "loss": 0.0,
      "num_input_tokens_seen": 20735528,
      "step": 36510
    },
    {
      "epoch": 640.6194690265487,
      "grad_norm": 2.5697666217183723e-08,
      "learning_rate": 0.005587032159042543,
      "loss": 0.0,
      "num_input_tokens_seen": 20738520,
      "step": 36515
    },
    {
      "epoch": 640.70796460177,
      "grad_norm": 3.9029778520216496e-08,
      "learning_rate": 0.005571116479474158,
      "loss": 0.0,
      "num_input_tokens_seen": 20741624,
      "step": 36520
    },
    {
      "epoch": 640.7964601769911,
      "grad_norm": 4.9979703931057884e-08,
      "learning_rate": 0.005555223072654619,
      "loss": 0.0,
      "num_input_tokens_seen": 20745176,
      "step": 36525
    },
    {
      "epoch": 640.8849557522124,
      "grad_norm": 4.266974329425466e-08,
      "learning_rate": 0.005539351941034881,
      "loss": 0.0,
      "num_input_tokens_seen": 20748024,
      "step": 36530
    },
    {
      "epoch": 640.9734513274336,
      "grad_norm": 2.7497469190507218e-08,
      "learning_rate": 0.0055235030870624865,
      "loss": 0.0,
      "num_input_tokens_seen": 20750584,
      "step": 36535
    },
    {
      "epoch": 641.0530973451328,
      "grad_norm": 7.629714104950835e-08,
      "learning_rate": 0.005507676513181514,
      "loss": 0.0,
      "num_input_tokens_seen": 20752816,
      "step": 36540
    },
    {
      "epoch": 641.141592920354,
      "grad_norm": 6.2240289366855e-08,
      "learning_rate": 0.005491872221832628,
      "loss": 0.0,
      "num_input_tokens_seen": 20755440,
      "step": 36545
    },
    {
      "epoch": 641.2300884955753,
      "grad_norm": 3.2048593112676826e-08,
      "learning_rate": 0.005476090215453061,
      "loss": 0.0,
      "num_input_tokens_seen": 20757968,
      "step": 36550
    },
    {
      "epoch": 641.3185840707964,
      "grad_norm": 4.120362717685566e-08,
      "learning_rate": 0.0054603304964765675,
      "loss": 0.0,
      "num_input_tokens_seen": 20761344,
      "step": 36555
    },
    {
      "epoch": 641.4070796460177,
      "grad_norm": 3.456584707350885e-08,
      "learning_rate": 0.005444593067333519,
      "loss": 0.0,
      "num_input_tokens_seen": 20763728,
      "step": 36560
    },
    {
      "epoch": 641.4955752212389,
      "grad_norm": 2.3443591956606724e-08,
      "learning_rate": 0.00542887793045081,
      "loss": 0.0,
      "num_input_tokens_seen": 20766512,
      "step": 36565
    },
    {
      "epoch": 641.5840707964602,
      "grad_norm": 3.226651301702077e-08,
      "learning_rate": 0.005413185088251932,
      "loss": 0.0,
      "num_input_tokens_seen": 20769168,
      "step": 36570
    },
    {
      "epoch": 641.6725663716815,
      "grad_norm": 4.404321884976525e-08,
      "learning_rate": 0.005397514543156884,
      "loss": 0.0,
      "num_input_tokens_seen": 20771456,
      "step": 36575
    },
    {
      "epoch": 641.7610619469026,
      "grad_norm": 5.61456481307232e-08,
      "learning_rate": 0.0053818662975822825,
      "loss": 0.0,
      "num_input_tokens_seen": 20775248,
      "step": 36580
    },
    {
      "epoch": 641.8495575221239,
      "grad_norm": 3.587053143405683e-08,
      "learning_rate": 0.005366240353941315,
      "loss": 0.0,
      "num_input_tokens_seen": 20778432,
      "step": 36585
    },
    {
      "epoch": 641.9380530973451,
      "grad_norm": 1.9218337143911413e-08,
      "learning_rate": 0.005350636714643636,
      "loss": 0.0,
      "num_input_tokens_seen": 20781216,
      "step": 36590
    },
    {
      "epoch": 642.0176991150443,
      "grad_norm": 3.6878716969113157e-08,
      "learning_rate": 0.005335055382095555,
      "loss": 0.0,
      "num_input_tokens_seen": 20783856,
      "step": 36595
    },
    {
      "epoch": 642.1061946902655,
      "grad_norm": 4.0810462564877525e-08,
      "learning_rate": 0.005319496358699915,
      "loss": 0.0,
      "num_input_tokens_seen": 20786240,
      "step": 36600
    },
    {
      "epoch": 642.1061946902655,
      "eval_loss": 0.8091742992401123,
      "eval_runtime": 0.9421,
      "eval_samples_per_second": 26.537,
      "eval_steps_per_second": 13.799,
      "num_input_tokens_seen": 20786240,
      "step": 36600
    },
    {
      "epoch": 642.1946902654868,
      "grad_norm": 1.1611691519419765e-07,
      "learning_rate": 0.005303959646856099,
      "loss": 0.0,
      "num_input_tokens_seen": 20789136,
      "step": 36605
    },
    {
      "epoch": 642.2831858407079,
      "grad_norm": 5.699756044919013e-08,
      "learning_rate": 0.005288445248960089,
      "loss": 0.0,
      "num_input_tokens_seen": 20792432,
      "step": 36610
    },
    {
      "epoch": 642.3716814159292,
      "grad_norm": 3.16632871033562e-08,
      "learning_rate": 0.005272953167404354,
      "loss": 0.0,
      "num_input_tokens_seen": 20795424,
      "step": 36615
    },
    {
      "epoch": 642.4601769911504,
      "grad_norm": 3.9078649649582076e-08,
      "learning_rate": 0.005257483404578017,
      "loss": 0.0,
      "num_input_tokens_seen": 20798112,
      "step": 36620
    },
    {
      "epoch": 642.5486725663717,
      "grad_norm": 4.0164874093306935e-08,
      "learning_rate": 0.0052420359628666865,
      "loss": 0.0,
      "num_input_tokens_seen": 20801360,
      "step": 36625
    },
    {
      "epoch": 642.637168141593,
      "grad_norm": 8.027679854194503e-08,
      "learning_rate": 0.00522661084465254,
      "loss": 0.0,
      "num_input_tokens_seen": 20804464,
      "step": 36630
    },
    {
      "epoch": 642.7256637168142,
      "grad_norm": 4.532380870614361e-08,
      "learning_rate": 0.005211208052314326,
      "loss": 0.0,
      "num_input_tokens_seen": 20807040,
      "step": 36635
    },
    {
      "epoch": 642.8141592920354,
      "grad_norm": 4.3886618783517406e-08,
      "learning_rate": 0.005195827588227391,
      "loss": 0.0,
      "num_input_tokens_seen": 20809696,
      "step": 36640
    },
    {
      "epoch": 642.9026548672566,
      "grad_norm": 2.873183291285386e-08,
      "learning_rate": 0.0051804694547635255,
      "loss": 0.0,
      "num_input_tokens_seen": 20812320,
      "step": 36645
    },
    {
      "epoch": 642.9911504424779,
      "grad_norm": 3.3174178071249116e-08,
      "learning_rate": 0.005165133654291232,
      "loss": 0.0,
      "num_input_tokens_seen": 20814832,
      "step": 36650
    },
    {
      "epoch": 643.070796460177,
      "grad_norm": 2.834717882649329e-08,
      "learning_rate": 0.005149820189175402,
      "loss": 0.0,
      "num_input_tokens_seen": 20817328,
      "step": 36655
    },
    {
      "epoch": 643.1592920353983,
      "grad_norm": 3.9583760269579216e-08,
      "learning_rate": 0.005134529061777598,
      "loss": 0.0,
      "num_input_tokens_seen": 20820272,
      "step": 36660
    },
    {
      "epoch": 643.2477876106195,
      "grad_norm": 4.233273642739732e-08,
      "learning_rate": 0.005119260274455933,
      "loss": 0.0,
      "num_input_tokens_seen": 20822992,
      "step": 36665
    },
    {
      "epoch": 643.3362831858407,
      "grad_norm": 1.375044789142521e-08,
      "learning_rate": 0.005104013829565007,
      "loss": 0.0,
      "num_input_tokens_seen": 20825984,
      "step": 36670
    },
    {
      "epoch": 643.4247787610619,
      "grad_norm": 2.013017841306919e-08,
      "learning_rate": 0.005088789729456006,
      "loss": 0.0,
      "num_input_tokens_seen": 20828704,
      "step": 36675
    },
    {
      "epoch": 643.5132743362832,
      "grad_norm": 8.595691269874806e-08,
      "learning_rate": 0.005073587976476735,
      "loss": 0.0,
      "num_input_tokens_seen": 20831696,
      "step": 36680
    },
    {
      "epoch": 643.6017699115044,
      "grad_norm": 3.481757104850658e-08,
      "learning_rate": 0.005058408572971418,
      "loss": 0.0,
      "num_input_tokens_seen": 20834800,
      "step": 36685
    },
    {
      "epoch": 643.6902654867257,
      "grad_norm": 1.311202879605844e-07,
      "learning_rate": 0.005043251521280983,
      "loss": 0.0,
      "num_input_tokens_seen": 20837744,
      "step": 36690
    },
    {
      "epoch": 643.7787610619469,
      "grad_norm": 4.308834178345933e-08,
      "learning_rate": 0.005028116823742795,
      "loss": 0.0,
      "num_input_tokens_seen": 20840480,
      "step": 36695
    },
    {
      "epoch": 643.8672566371681,
      "grad_norm": 1.9494461156455145e-08,
      "learning_rate": 0.005013004482690819,
      "loss": 0.0,
      "num_input_tokens_seen": 20843088,
      "step": 36700
    },
    {
      "epoch": 643.9557522123894,
      "grad_norm": 3.034211815133858e-08,
      "learning_rate": 0.0049979145004555746,
      "loss": 0.0,
      "num_input_tokens_seen": 20846304,
      "step": 36705
    },
    {
      "epoch": 644.0353982300885,
      "grad_norm": 2.8696103271386164e-08,
      "learning_rate": 0.004982846879364116,
      "loss": 0.0,
      "num_input_tokens_seen": 20848320,
      "step": 36710
    },
    {
      "epoch": 644.1238938053098,
      "grad_norm": 1.838600738324203e-08,
      "learning_rate": 0.0049678016217400535,
      "loss": 0.0,
      "num_input_tokens_seen": 20851008,
      "step": 36715
    },
    {
      "epoch": 644.212389380531,
      "grad_norm": 1.07322559728118e-08,
      "learning_rate": 0.004952778729903595,
      "loss": 0.0,
      "num_input_tokens_seen": 20853856,
      "step": 36720
    },
    {
      "epoch": 644.3008849557522,
      "grad_norm": 2.2175894898168735e-08,
      "learning_rate": 0.004937778206171422,
      "loss": 0.0,
      "num_input_tokens_seen": 20857056,
      "step": 36725
    },
    {
      "epoch": 644.3893805309734,
      "grad_norm": 2.479312932734956e-08,
      "learning_rate": 0.004922800052856835,
      "loss": 0.0,
      "num_input_tokens_seen": 20860064,
      "step": 36730
    },
    {
      "epoch": 644.4778761061947,
      "grad_norm": 4.576204304385101e-08,
      "learning_rate": 0.004907844272269602,
      "loss": 0.0,
      "num_input_tokens_seen": 20863216,
      "step": 36735
    },
    {
      "epoch": 644.566371681416,
      "grad_norm": 3.5262956998849404e-08,
      "learning_rate": 0.004892910866716144,
      "loss": 0.0,
      "num_input_tokens_seen": 20866288,
      "step": 36740
    },
    {
      "epoch": 644.6548672566372,
      "grad_norm": 1.998389365098774e-08,
      "learning_rate": 0.004877999838499369,
      "loss": 0.0,
      "num_input_tokens_seen": 20868816,
      "step": 36745
    },
    {
      "epoch": 644.7433628318585,
      "grad_norm": 1.929228261587923e-07,
      "learning_rate": 0.0048631111899187065,
      "loss": 0.0,
      "num_input_tokens_seen": 20871056,
      "step": 36750
    },
    {
      "epoch": 644.8318584070796,
      "grad_norm": 3.7842976752244795e-08,
      "learning_rate": 0.0048482449232702335,
      "loss": 0.0,
      "num_input_tokens_seen": 20874096,
      "step": 36755
    },
    {
      "epoch": 644.9203539823009,
      "grad_norm": 7.028447157608753e-08,
      "learning_rate": 0.004833401040846469,
      "loss": 0.0,
      "num_input_tokens_seen": 20877168,
      "step": 36760
    },
    {
      "epoch": 645.0,
      "grad_norm": 8.119040550980117e-09,
      "learning_rate": 0.004818579544936546,
      "loss": 0.0,
      "num_input_tokens_seen": 20879464,
      "step": 36765
    },
    {
      "epoch": 645.0884955752213,
      "grad_norm": 2.2904336205442632e-08,
      "learning_rate": 0.004803780437826121,
      "loss": 0.0,
      "num_input_tokens_seen": 20882296,
      "step": 36770
    },
    {
      "epoch": 645.1769911504425,
      "grad_norm": 4.063452507807597e-08,
      "learning_rate": 0.004789003721797402,
      "loss": 0.0,
      "num_input_tokens_seen": 20885384,
      "step": 36775
    },
    {
      "epoch": 645.2654867256637,
      "grad_norm": 4.368028427848003e-08,
      "learning_rate": 0.004774249399129132,
      "loss": 0.0,
      "num_input_tokens_seen": 20887736,
      "step": 36780
    },
    {
      "epoch": 645.3539823008849,
      "grad_norm": 3.828874994837861e-08,
      "learning_rate": 0.004759517472096642,
      "loss": 0.0,
      "num_input_tokens_seen": 20890680,
      "step": 36785
    },
    {
      "epoch": 645.4424778761062,
      "grad_norm": 3.7803143726478083e-08,
      "learning_rate": 0.004744807942971746,
      "loss": 0.0,
      "num_input_tokens_seen": 20893480,
      "step": 36790
    },
    {
      "epoch": 645.5309734513274,
      "grad_norm": 7.840308313689093e-08,
      "learning_rate": 0.004730120814022881,
      "loss": 0.0,
      "num_input_tokens_seen": 20896312,
      "step": 36795
    },
    {
      "epoch": 645.6194690265487,
      "grad_norm": 3.8408423108649004e-08,
      "learning_rate": 0.004715456087514935,
      "loss": 0.0,
      "num_input_tokens_seen": 20899128,
      "step": 36800
    },
    {
      "epoch": 645.6194690265487,
      "eval_loss": 0.8278312683105469,
      "eval_runtime": 0.95,
      "eval_samples_per_second": 26.317,
      "eval_steps_per_second": 13.685,
      "num_input_tokens_seen": 20899128,
      "step": 36800
    },
    {
      "epoch": 645.70796460177,
      "grad_norm": 3.22693232135407e-08,
      "learning_rate": 0.004700813765709432,
      "loss": 0.0,
      "num_input_tokens_seen": 20901880,
      "step": 36805
    },
    {
      "epoch": 645.7964601769911,
      "grad_norm": 3.666595560503083e-08,
      "learning_rate": 0.004686193850864401,
      "loss": 0.0,
      "num_input_tokens_seen": 20904840,
      "step": 36810
    },
    {
      "epoch": 645.8849557522124,
      "grad_norm": 3.6521640822684276e-08,
      "learning_rate": 0.004671596345234385,
      "loss": 0.0,
      "num_input_tokens_seen": 20907688,
      "step": 36815
    },
    {
      "epoch": 645.9734513274336,
      "grad_norm": 4.0125627265297226e-08,
      "learning_rate": 0.00465702125107052,
      "loss": 0.0,
      "num_input_tokens_seen": 20911096,
      "step": 36820
    },
    {
      "epoch": 646.0530973451328,
      "grad_norm": 3.7802809771392276e-08,
      "learning_rate": 0.004642468570620506,
      "loss": 0.0,
      "num_input_tokens_seen": 20913912,
      "step": 36825
    },
    {
      "epoch": 646.141592920354,
      "grad_norm": 3.740027665344314e-08,
      "learning_rate": 0.004627938306128482,
      "loss": 0.0,
      "num_input_tokens_seen": 20916776,
      "step": 36830
    },
    {
      "epoch": 646.2300884955753,
      "grad_norm": 3.477663312878576e-08,
      "learning_rate": 0.004613430459835255,
      "loss": 0.0,
      "num_input_tokens_seen": 20919736,
      "step": 36835
    },
    {
      "epoch": 646.3185840707964,
      "grad_norm": 3.290728400884291e-08,
      "learning_rate": 0.004598945033978085,
      "loss": 0.0,
      "num_input_tokens_seen": 20922872,
      "step": 36840
    },
    {
      "epoch": 646.4070796460177,
      "grad_norm": 6.337757696428525e-08,
      "learning_rate": 0.004584482030790804,
      "loss": 0.0,
      "num_input_tokens_seen": 20925352,
      "step": 36845
    },
    {
      "epoch": 646.4955752212389,
      "grad_norm": 2.1806918937272712e-08,
      "learning_rate": 0.004570041452503826,
      "loss": 0.0,
      "num_input_tokens_seen": 20928024,
      "step": 36850
    },
    {
      "epoch": 646.5840707964602,
      "grad_norm": 1.4695407557496765e-08,
      "learning_rate": 0.004555623301344003,
      "loss": 0.0,
      "num_input_tokens_seen": 20930792,
      "step": 36855
    },
    {
      "epoch": 646.6725663716815,
      "grad_norm": 4.75331312088656e-08,
      "learning_rate": 0.004541227579534857,
      "loss": 0.0,
      "num_input_tokens_seen": 20933800,
      "step": 36860
    },
    {
      "epoch": 646.7610619469026,
      "grad_norm": 5.157863824933884e-08,
      "learning_rate": 0.004526854289296378,
      "loss": 0.0,
      "num_input_tokens_seen": 20936744,
      "step": 36865
    },
    {
      "epoch": 646.8495575221239,
      "grad_norm": 1.0859145582742258e-08,
      "learning_rate": 0.004512503432845078,
      "loss": 0.0,
      "num_input_tokens_seen": 20939656,
      "step": 36870
    },
    {
      "epoch": 646.9380530973451,
      "grad_norm": 1.117792169935683e-08,
      "learning_rate": 0.004498175012394068,
      "loss": 0.0,
      "num_input_tokens_seen": 20942680,
      "step": 36875
    },
    {
      "epoch": 647.0176991150443,
      "grad_norm": 6.540041397329333e-08,
      "learning_rate": 0.004483869030152965,
      "loss": 0.0,
      "num_input_tokens_seen": 20944744,
      "step": 36880
    },
    {
      "epoch": 647.1061946902655,
      "grad_norm": 3.9540399399129456e-08,
      "learning_rate": 0.004469585488327904,
      "loss": 0.0,
      "num_input_tokens_seen": 20947528,
      "step": 36885
    },
    {
      "epoch": 647.1946902654868,
      "grad_norm": 3.592704445054551e-08,
      "learning_rate": 0.0044553243891216395,
      "loss": 0.0,
      "num_input_tokens_seen": 20950200,
      "step": 36890
    },
    {
      "epoch": 647.2831858407079,
      "grad_norm": 3.462697506506629e-08,
      "learning_rate": 0.004441085734733363,
      "loss": 0.0,
      "num_input_tokens_seen": 20952680,
      "step": 36895
    },
    {
      "epoch": 647.3716814159292,
      "grad_norm": 3.129603598495123e-08,
      "learning_rate": 0.004426869527358884,
      "loss": 0.0,
      "num_input_tokens_seen": 20955272,
      "step": 36900
    },
    {
      "epoch": 647.4601769911504,
      "grad_norm": 4.8557048160091654e-08,
      "learning_rate": 0.0044126757691905156,
      "loss": 0.0,
      "num_input_tokens_seen": 20958232,
      "step": 36905
    },
    {
      "epoch": 647.5486725663717,
      "grad_norm": 4.991055391201371e-08,
      "learning_rate": 0.004398504462417107,
      "loss": 0.0,
      "num_input_tokens_seen": 20961640,
      "step": 36910
    },
    {
      "epoch": 647.637168141593,
      "grad_norm": 4.405602282986365e-08,
      "learning_rate": 0.0043843556092240605,
      "loss": 0.0,
      "num_input_tokens_seen": 20964552,
      "step": 36915
    },
    {
      "epoch": 647.7256637168142,
      "grad_norm": 2.5479900855884807e-08,
      "learning_rate": 0.004370229211793281,
      "loss": 0.0,
      "num_input_tokens_seen": 20967304,
      "step": 36920
    },
    {
      "epoch": 647.8141592920354,
      "grad_norm": 6.200297519853848e-08,
      "learning_rate": 0.0043561252723032405,
      "loss": 0.0,
      "num_input_tokens_seen": 20970504,
      "step": 36925
    },
    {
      "epoch": 647.9026548672566,
      "grad_norm": 2.9372975163255433e-08,
      "learning_rate": 0.004342043792929001,
      "loss": 0.0,
      "num_input_tokens_seen": 20973240,
      "step": 36930
    },
    {
      "epoch": 647.9911504424779,
      "grad_norm": 4.313697488100843e-08,
      "learning_rate": 0.004327984775842025,
      "loss": 0.0,
      "num_input_tokens_seen": 20976248,
      "step": 36935
    },
    {
      "epoch": 648.070796460177,
      "grad_norm": 5.492332988410453e-08,
      "learning_rate": 0.004313948223210428,
      "loss": 0.0,
      "num_input_tokens_seen": 20978880,
      "step": 36940
    },
    {
      "epoch": 648.1592920353983,
      "grad_norm": 1.4846460061335165e-08,
      "learning_rate": 0.004299934137198846,
      "loss": 0.0,
      "num_input_tokens_seen": 20981456,
      "step": 36945
    },
    {
      "epoch": 648.2477876106195,
      "grad_norm": 2.093704054573209e-08,
      "learning_rate": 0.004285942519968383,
      "loss": 0.0,
      "num_input_tokens_seen": 20984208,
      "step": 36950
    },
    {
      "epoch": 648.3362831858407,
      "grad_norm": 3.413774507521339e-08,
      "learning_rate": 0.004271973373676746,
      "loss": 0.0,
      "num_input_tokens_seen": 20986896,
      "step": 36955
    },
    {
      "epoch": 648.4247787610619,
      "grad_norm": 5.6070089016202473e-08,
      "learning_rate": 0.004258026700478146,
      "loss": 0.0,
      "num_input_tokens_seen": 20989504,
      "step": 36960
    },
    {
      "epoch": 648.5132743362832,
      "grad_norm": 4.3967393281718614e-08,
      "learning_rate": 0.004244102502523328,
      "loss": 0.0,
      "num_input_tokens_seen": 20992400,
      "step": 36965
    },
    {
      "epoch": 648.6017699115044,
      "grad_norm": 4.196034453229913e-08,
      "learning_rate": 0.004230200781959592,
      "loss": 0.0,
      "num_input_tokens_seen": 20995376,
      "step": 36970
    },
    {
      "epoch": 648.6902654867257,
      "grad_norm": 7.181028394143141e-08,
      "learning_rate": 0.004216321540930756,
      "loss": 0.0,
      "num_input_tokens_seen": 20998352,
      "step": 36975
    },
    {
      "epoch": 648.7787610619469,
      "grad_norm": 5.33842872130208e-08,
      "learning_rate": 0.004202464781577175,
      "loss": 0.0,
      "num_input_tokens_seen": 21001280,
      "step": 36980
    },
    {
      "epoch": 648.8672566371681,
      "grad_norm": 6.009101838344577e-08,
      "learning_rate": 0.00418863050603574,
      "loss": 0.0,
      "num_input_tokens_seen": 21004672,
      "step": 36985
    },
    {
      "epoch": 648.9557522123894,
      "grad_norm": 8.178159305316512e-08,
      "learning_rate": 0.004174818716439843,
      "loss": 0.0,
      "num_input_tokens_seen": 21007280,
      "step": 36990
    },
    {
      "epoch": 649.0353982300885,
      "grad_norm": 5.676412229149719e-08,
      "learning_rate": 0.004161029414919464,
      "loss": 0.0,
      "num_input_tokens_seen": 21009448,
      "step": 36995
    },
    {
      "epoch": 649.1238938053098,
      "grad_norm": 3.796812464429422e-08,
      "learning_rate": 0.004147262603601071,
      "loss": 0.0,
      "num_input_tokens_seen": 21011928,
      "step": 37000
    },
    {
      "epoch": 649.1238938053098,
      "eval_loss": 0.8117527365684509,
      "eval_runtime": 0.9349,
      "eval_samples_per_second": 26.741,
      "eval_steps_per_second": 13.905,
      "num_input_tokens_seen": 21011928,
      "step": 37000
    },
    {
      "epoch": 649.212389380531,
      "grad_norm": 2.9113389032886516e-08,
      "learning_rate": 0.004133518284607679,
      "loss": 0.0,
      "num_input_tokens_seen": 21014648,
      "step": 37005
    },
    {
      "epoch": 649.3008849557522,
      "grad_norm": 6.281534581376036e-08,
      "learning_rate": 0.004119796460058861,
      "loss": 0.0,
      "num_input_tokens_seen": 21017384,
      "step": 37010
    },
    {
      "epoch": 649.3893805309734,
      "grad_norm": 3.338741194625072e-08,
      "learning_rate": 0.00410609713207064,
      "loss": 0.0,
      "num_input_tokens_seen": 21020280,
      "step": 37015
    },
    {
      "epoch": 649.4778761061947,
      "grad_norm": 3.0811055040658175e-08,
      "learning_rate": 0.004092420302755678,
      "loss": 0.0,
      "num_input_tokens_seen": 21023752,
      "step": 37020
    },
    {
      "epoch": 649.566371681416,
      "grad_norm": 1.9337896617344086e-08,
      "learning_rate": 0.004078765974223103,
      "loss": 0.0,
      "num_input_tokens_seen": 21026904,
      "step": 37025
    },
    {
      "epoch": 649.6548672566372,
      "grad_norm": 4.727112568048142e-08,
      "learning_rate": 0.004065134148578564,
      "loss": 0.0,
      "num_input_tokens_seen": 21029224,
      "step": 37030
    },
    {
      "epoch": 649.7433628318585,
      "grad_norm": 2.212022742753561e-08,
      "learning_rate": 0.004051524827924279,
      "loss": 0.0,
      "num_input_tokens_seen": 21032296,
      "step": 37035
    },
    {
      "epoch": 649.8318584070796,
      "grad_norm": 2.0680026580066624e-08,
      "learning_rate": 0.004037938014358955,
      "loss": 0.0,
      "num_input_tokens_seen": 21035096,
      "step": 37040
    },
    {
      "epoch": 649.9203539823009,
      "grad_norm": 3.090278966055848e-08,
      "learning_rate": 0.004024373709977863,
      "loss": 0.0,
      "num_input_tokens_seen": 21037944,
      "step": 37045
    },
    {
      "epoch": 650.0,
      "grad_norm": 2.7358961318668662e-08,
      "learning_rate": 0.004010831916872814,
      "loss": 0.0,
      "num_input_tokens_seen": 21040632,
      "step": 37050
    },
    {
      "epoch": 650.0884955752213,
      "grad_norm": 3.2930039139955625e-08,
      "learning_rate": 0.003997312637132089,
      "loss": 0.0,
      "num_input_tokens_seen": 21043736,
      "step": 37055
    },
    {
      "epoch": 650.1769911504425,
      "grad_norm": 2.5299767614228585e-08,
      "learning_rate": 0.003983815872840535,
      "loss": 0.0,
      "num_input_tokens_seen": 21046376,
      "step": 37060
    },
    {
      "epoch": 650.2654867256637,
      "grad_norm": 3.8657571366229604e-08,
      "learning_rate": 0.003970341626079521,
      "loss": 0.0,
      "num_input_tokens_seen": 21049368,
      "step": 37065
    },
    {
      "epoch": 650.3539823008849,
      "grad_norm": 7.487333419931019e-08,
      "learning_rate": 0.003956889898926952,
      "loss": 0.0,
      "num_input_tokens_seen": 21051832,
      "step": 37070
    },
    {
      "epoch": 650.4424778761062,
      "grad_norm": 5.407892444964091e-08,
      "learning_rate": 0.0039434606934572675,
      "loss": 0.0,
      "num_input_tokens_seen": 21055064,
      "step": 37075
    },
    {
      "epoch": 650.5309734513274,
      "grad_norm": 4.9115886469053294e-08,
      "learning_rate": 0.003930054011741396,
      "loss": 0.0,
      "num_input_tokens_seen": 21057800,
      "step": 37080
    },
    {
      "epoch": 650.6194690265487,
      "grad_norm": 5.768522370885876e-08,
      "learning_rate": 0.0039166698558468155,
      "loss": 0.0,
      "num_input_tokens_seen": 21061336,
      "step": 37085
    },
    {
      "epoch": 650.70796460177,
      "grad_norm": 4.118613006198757e-08,
      "learning_rate": 0.0039033082278375594,
      "loss": 0.0,
      "num_input_tokens_seen": 21064280,
      "step": 37090
    },
    {
      "epoch": 650.7964601769911,
      "grad_norm": 1.8944835034062635e-08,
      "learning_rate": 0.003889969129774112,
      "loss": 0.0,
      "num_input_tokens_seen": 21066888,
      "step": 37095
    },
    {
      "epoch": 650.8849557522124,
      "grad_norm": 6.035667610149176e-08,
      "learning_rate": 0.0038766525637135784,
      "loss": 0.0,
      "num_input_tokens_seen": 21069544,
      "step": 37100
    },
    {
      "epoch": 650.9734513274336,
      "grad_norm": 2.5994147279106983e-08,
      "learning_rate": 0.0038633585317095318,
      "loss": 0.0,
      "num_input_tokens_seen": 21072312,
      "step": 37105
    },
    {
      "epoch": 651.0530973451328,
      "grad_norm": 5.194322838519838e-08,
      "learning_rate": 0.00385008703581205,
      "loss": 0.0,
      "num_input_tokens_seen": 21074672,
      "step": 37110
    },
    {
      "epoch": 651.141592920354,
      "grad_norm": 6.940526731114005e-08,
      "learning_rate": 0.0038368380780677944,
      "loss": 0.0,
      "num_input_tokens_seen": 21077696,
      "step": 37115
    },
    {
      "epoch": 651.2300884955753,
      "grad_norm": 7.362491771800705e-08,
      "learning_rate": 0.003823611660519882,
      "loss": 0.0,
      "num_input_tokens_seen": 21080592,
      "step": 37120
    },
    {
      "epoch": 651.3185840707964,
      "grad_norm": 6.333379332090772e-08,
      "learning_rate": 0.0038104077852080475,
      "loss": 0.0,
      "num_input_tokens_seen": 21083920,
      "step": 37125
    },
    {
      "epoch": 651.4070796460177,
      "grad_norm": 2.09202468681724e-08,
      "learning_rate": 0.003797226454168462,
      "loss": 0.0,
      "num_input_tokens_seen": 21086736,
      "step": 37130
    },
    {
      "epoch": 651.4955752212389,
      "grad_norm": 2.522790154557697e-08,
      "learning_rate": 0.003784067669433849,
      "loss": 0.0,
      "num_input_tokens_seen": 21088992,
      "step": 37135
    },
    {
      "epoch": 651.5840707964602,
      "grad_norm": 5.34563646681363e-08,
      "learning_rate": 0.0037709314330334528,
      "loss": 0.0,
      "num_input_tokens_seen": 21092416,
      "step": 37140
    },
    {
      "epoch": 651.6725663716815,
      "grad_norm": 4.249406515555165e-08,
      "learning_rate": 0.003757817746993086,
      "loss": 0.0,
      "num_input_tokens_seen": 21095232,
      "step": 37145
    },
    {
      "epoch": 651.7610619469026,
      "grad_norm": 8.579137755759803e-08,
      "learning_rate": 0.0037447266133349977,
      "loss": 0.0,
      "num_input_tokens_seen": 21098064,
      "step": 37150
    },
    {
      "epoch": 651.8495575221239,
      "grad_norm": 4.245572426953004e-08,
      "learning_rate": 0.003731658034078039,
      "loss": 0.0,
      "num_input_tokens_seen": 21101136,
      "step": 37155
    },
    {
      "epoch": 651.9380530973451,
      "grad_norm": 3.861057606968643e-08,
      "learning_rate": 0.0037186120112375153,
      "loss": 0.0,
      "num_input_tokens_seen": 21104112,
      "step": 37160
    },
    {
      "epoch": 652.0176991150443,
      "grad_norm": 7.755475373016907e-09,
      "learning_rate": 0.003705588546825317,
      "loss": 0.0,
      "num_input_tokens_seen": 21106448,
      "step": 37165
    },
    {
      "epoch": 652.1061946902655,
      "grad_norm": 3.408525728332279e-08,
      "learning_rate": 0.0036925876428498205,
      "loss": 0.0,
      "num_input_tokens_seen": 21109408,
      "step": 37170
    },
    {
      "epoch": 652.1946902654868,
      "grad_norm": 2.893963468864058e-08,
      "learning_rate": 0.0036796093013159057,
      "loss": 0.0,
      "num_input_tokens_seen": 21112464,
      "step": 37175
    },
    {
      "epoch": 652.2831858407079,
      "grad_norm": 2.30171597337403e-08,
      "learning_rate": 0.0036666535242250217,
      "loss": 0.0,
      "num_input_tokens_seen": 21115408,
      "step": 37180
    },
    {
      "epoch": 652.3716814159292,
      "grad_norm": 4.65489051748591e-08,
      "learning_rate": 0.003653720313575104,
      "loss": 0.0,
      "num_input_tokens_seen": 21118480,
      "step": 37185
    },
    {
      "epoch": 652.4601769911504,
      "grad_norm": 7.365077436816136e-08,
      "learning_rate": 0.003640809671360623,
      "loss": 0.0,
      "num_input_tokens_seen": 21121104,
      "step": 37190
    },
    {
      "epoch": 652.5486725663717,
      "grad_norm": 6.492869175644955e-08,
      "learning_rate": 0.003627921599572553,
      "loss": 0.0,
      "num_input_tokens_seen": 21124112,
      "step": 37195
    },
    {
      "epoch": 652.637168141593,
      "grad_norm": 3.0036300557867435e-08,
      "learning_rate": 0.003615056100198405,
      "loss": 0.0,
      "num_input_tokens_seen": 21126880,
      "step": 37200
    },
    {
      "epoch": 652.637168141593,
      "eval_loss": 0.7886305451393127,
      "eval_runtime": 0.9345,
      "eval_samples_per_second": 26.753,
      "eval_steps_per_second": 13.912,
      "num_input_tokens_seen": 21126880,
      "step": 37200
    },
    {
      "epoch": 652.7256637168142,
      "grad_norm": 6.062035140530497e-08,
      "learning_rate": 0.003602213175222174,
      "loss": 0.0,
      "num_input_tokens_seen": 21130272,
      "step": 37205
    },
    {
      "epoch": 652.8141592920354,
      "grad_norm": 3.5634577955079294e-08,
      "learning_rate": 0.0035893928266244432,
      "loss": 0.0,
      "num_input_tokens_seen": 21132608,
      "step": 37210
    },
    {
      "epoch": 652.9026548672566,
      "grad_norm": 3.227031086794341e-08,
      "learning_rate": 0.003576595056382248,
      "loss": 0.0,
      "num_input_tokens_seen": 21135184,
      "step": 37215
    },
    {
      "epoch": 652.9911504424779,
      "grad_norm": 2.6578204526117588e-08,
      "learning_rate": 0.0035638198664691423,
      "loss": 0.0,
      "num_input_tokens_seen": 21138096,
      "step": 37220
    },
    {
      "epoch": 653.070796460177,
      "grad_norm": 4.208596138255416e-08,
      "learning_rate": 0.003551067258855267,
      "loss": 0.0,
      "num_input_tokens_seen": 21140472,
      "step": 37225
    },
    {
      "epoch": 653.1592920353983,
      "grad_norm": 3.33659144757803e-08,
      "learning_rate": 0.0035383372355071996,
      "loss": 0.0,
      "num_input_tokens_seen": 21142856,
      "step": 37230
    },
    {
      "epoch": 653.2477876106195,
      "grad_norm": 2.7888562570410613e-08,
      "learning_rate": 0.0035256297983881023,
      "loss": 0.0,
      "num_input_tokens_seen": 21145688,
      "step": 37235
    },
    {
      "epoch": 653.3362831858407,
      "grad_norm": 1.0680620476932745e-07,
      "learning_rate": 0.0035129449494575747,
      "loss": 0.0,
      "num_input_tokens_seen": 21148728,
      "step": 37240
    },
    {
      "epoch": 653.4247787610619,
      "grad_norm": 6.677143460365187e-08,
      "learning_rate": 0.0035002826906718187,
      "loss": 0.0,
      "num_input_tokens_seen": 21151800,
      "step": 37245
    },
    {
      "epoch": 653.5132743362832,
      "grad_norm": 2.2372571351070292e-08,
      "learning_rate": 0.003487643023983522,
      "loss": 0.0,
      "num_input_tokens_seen": 21154456,
      "step": 37250
    },
    {
      "epoch": 653.6017699115044,
      "grad_norm": 4.016085242142253e-08,
      "learning_rate": 0.003475025951341842,
      "loss": 0.0,
      "num_input_tokens_seen": 21157192,
      "step": 37255
    },
    {
      "epoch": 653.6902654867257,
      "grad_norm": 3.2407420746949356e-08,
      "learning_rate": 0.00346243147469249,
      "loss": 0.0,
      "num_input_tokens_seen": 21160504,
      "step": 37260
    },
    {
      "epoch": 653.7787610619469,
      "grad_norm": 5.3330690974462414e-08,
      "learning_rate": 0.0034498595959777446,
      "loss": 0.0,
      "num_input_tokens_seen": 21163080,
      "step": 37265
    },
    {
      "epoch": 653.8672566371681,
      "grad_norm": 3.686852423356868e-08,
      "learning_rate": 0.003437310317136305,
      "loss": 0.0,
      "num_input_tokens_seen": 21166392,
      "step": 37270
    },
    {
      "epoch": 653.9557522123894,
      "grad_norm": 4.860536506612334e-08,
      "learning_rate": 0.0034247836401034236,
      "loss": 0.0,
      "num_input_tokens_seen": 21169320,
      "step": 37275
    },
    {
      "epoch": 654.0353982300885,
      "grad_norm": 4.481107396259176e-08,
      "learning_rate": 0.003412279566810905,
      "loss": 0.0,
      "num_input_tokens_seen": 21171848,
      "step": 37280
    },
    {
      "epoch": 654.1238938053098,
      "grad_norm": 3.59937111227282e-08,
      "learning_rate": 0.00339979809918699,
      "loss": 0.0,
      "num_input_tokens_seen": 21174520,
      "step": 37285
    },
    {
      "epoch": 654.212389380531,
      "grad_norm": 3.788504443491547e-08,
      "learning_rate": 0.0033873392391565228,
      "loss": 0.0,
      "num_input_tokens_seen": 21177896,
      "step": 37290
    },
    {
      "epoch": 654.3008849557522,
      "grad_norm": 9.270372913761094e-08,
      "learning_rate": 0.003374902988640782,
      "loss": 0.0,
      "num_input_tokens_seen": 21181176,
      "step": 37295
    },
    {
      "epoch": 654.3893805309734,
      "grad_norm": 3.3911117469642704e-08,
      "learning_rate": 0.0033624893495576014,
      "loss": 0.0,
      "num_input_tokens_seen": 21184296,
      "step": 37300
    },
    {
      "epoch": 654.4778761061947,
      "grad_norm": 7.00971654055138e-08,
      "learning_rate": 0.0033500983238213323,
      "loss": 0.0,
      "num_input_tokens_seen": 21186808,
      "step": 37305
    },
    {
      "epoch": 654.566371681416,
      "grad_norm": 5.063611752120778e-08,
      "learning_rate": 0.0033377299133428126,
      "loss": 0.0,
      "num_input_tokens_seen": 21189352,
      "step": 37310
    },
    {
      "epoch": 654.6548672566372,
      "grad_norm": 3.709910600946387e-08,
      "learning_rate": 0.003325384120029434,
      "loss": 0.0,
      "num_input_tokens_seen": 21192088,
      "step": 37315
    },
    {
      "epoch": 654.7433628318585,
      "grad_norm": 4.739462511338388e-08,
      "learning_rate": 0.0033130609457850233,
      "loss": 0.0,
      "num_input_tokens_seen": 21195112,
      "step": 37320
    },
    {
      "epoch": 654.8318584070796,
      "grad_norm": 2.5356511557106387e-08,
      "learning_rate": 0.0033007603925100104,
      "loss": 0.0,
      "num_input_tokens_seen": 21197720,
      "step": 37325
    },
    {
      "epoch": 654.9203539823009,
      "grad_norm": 2.0548434065403853e-08,
      "learning_rate": 0.003288482462101294,
      "loss": 0.0,
      "num_input_tokens_seen": 21200888,
      "step": 37330
    },
    {
      "epoch": 655.0,
      "grad_norm": 8.300532527982796e-08,
      "learning_rate": 0.0032762271564522605,
      "loss": 0.0,
      "num_input_tokens_seen": 21202968,
      "step": 37335
    },
    {
      "epoch": 655.0884955752213,
      "grad_norm": 4.202845005352174e-08,
      "learning_rate": 0.003263994477452864,
      "loss": 0.0,
      "num_input_tokens_seen": 21206328,
      "step": 37340
    },
    {
      "epoch": 655.1769911504425,
      "grad_norm": 6.618953563020114e-08,
      "learning_rate": 0.0032517844269895125,
      "loss": 0.0,
      "num_input_tokens_seen": 21209128,
      "step": 37345
    },
    {
      "epoch": 655.2654867256637,
      "grad_norm": 4.0008632851140646e-08,
      "learning_rate": 0.0032395970069451496,
      "loss": 0.0,
      "num_input_tokens_seen": 21211752,
      "step": 37350
    },
    {
      "epoch": 655.3539823008849,
      "grad_norm": 4.323899460700886e-08,
      "learning_rate": 0.0032274322191992388,
      "loss": 0.0,
      "num_input_tokens_seen": 21214664,
      "step": 37355
    },
    {
      "epoch": 655.4424778761062,
      "grad_norm": 1.05073318934501e-07,
      "learning_rate": 0.0032152900656277294,
      "loss": 0.0,
      "num_input_tokens_seen": 21217400,
      "step": 37360
    },
    {
      "epoch": 655.5309734513274,
      "grad_norm": 4.386232532738177e-08,
      "learning_rate": 0.0032031705481030902,
      "loss": 0.0,
      "num_input_tokens_seen": 21219896,
      "step": 37365
    },
    {
      "epoch": 655.6194690265487,
      "grad_norm": 4.4575337199148635e-08,
      "learning_rate": 0.0031910736684943428,
      "loss": 0.0,
      "num_input_tokens_seen": 21223032,
      "step": 37370
    },
    {
      "epoch": 655.70796460177,
      "grad_norm": 3.5059535719028645e-08,
      "learning_rate": 0.0031789994286669453,
      "loss": 0.0,
      "num_input_tokens_seen": 21225592,
      "step": 37375
    },
    {
      "epoch": 655.7964601769911,
      "grad_norm": 1.4152097271846742e-08,
      "learning_rate": 0.003166947830482908,
      "loss": 0.0,
      "num_input_tokens_seen": 21228680,
      "step": 37380
    },
    {
      "epoch": 655.8849557522124,
      "grad_norm": 6.626687820698862e-08,
      "learning_rate": 0.003154918875800727,
      "loss": 0.0,
      "num_input_tokens_seen": 21231832,
      "step": 37385
    },
    {
      "epoch": 655.9734513274336,
      "grad_norm": 3.7577326139626166e-08,
      "learning_rate": 0.00314291256647542,
      "loss": 0.0,
      "num_input_tokens_seen": 21234840,
      "step": 37390
    },
    {
      "epoch": 656.0530973451328,
      "grad_norm": 7.368196008883388e-08,
      "learning_rate": 0.0031309289043585375,
      "loss": 0.0,
      "num_input_tokens_seen": 21237344,
      "step": 37395
    },
    {
      "epoch": 656.141592920354,
      "grad_norm": 2.6667446917372217e-08,
      "learning_rate": 0.003118967891298069,
      "loss": 0.0,
      "num_input_tokens_seen": 21239760,
      "step": 37400
    },
    {
      "epoch": 656.141592920354,
      "eval_loss": 0.7863349914550781,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.588,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 21239760,
      "step": 37400
    },
    {
      "epoch": 656.2300884955753,
      "grad_norm": 3.121964553542966e-08,
      "learning_rate": 0.003107029529138572,
      "loss": 0.0,
      "num_input_tokens_seen": 21242448,
      "step": 37405
    },
    {
      "epoch": 656.3185840707964,
      "grad_norm": 9.19739875371306e-08,
      "learning_rate": 0.0030951138197211235,
      "loss": 0.0,
      "num_input_tokens_seen": 21245472,
      "step": 37410
    },
    {
      "epoch": 656.4070796460177,
      "grad_norm": 3.861311981268045e-08,
      "learning_rate": 0.0030832207648832377,
      "loss": 0.0,
      "num_input_tokens_seen": 21248432,
      "step": 37415
    },
    {
      "epoch": 656.4955752212389,
      "grad_norm": 3.2920649317702555e-08,
      "learning_rate": 0.0030713503664589635,
      "loss": 0.0,
      "num_input_tokens_seen": 21251152,
      "step": 37420
    },
    {
      "epoch": 656.5840707964602,
      "grad_norm": 2.0426202951284722e-08,
      "learning_rate": 0.0030595026262788872,
      "loss": 0.0,
      "num_input_tokens_seen": 21254528,
      "step": 37425
    },
    {
      "epoch": 656.6725663716815,
      "grad_norm": 3.511190982408152e-08,
      "learning_rate": 0.00304767754617008,
      "loss": 0.0,
      "num_input_tokens_seen": 21257216,
      "step": 37430
    },
    {
      "epoch": 656.7610619469026,
      "grad_norm": 2.987644265317613e-08,
      "learning_rate": 0.003035875127956117,
      "loss": 0.0,
      "num_input_tokens_seen": 21259872,
      "step": 37435
    },
    {
      "epoch": 656.8495575221239,
      "grad_norm": 4.80549395831531e-08,
      "learning_rate": 0.0030240953734570752,
      "loss": 0.0,
      "num_input_tokens_seen": 21262784,
      "step": 37440
    },
    {
      "epoch": 656.9380530973451,
      "grad_norm": 4.500115124983495e-08,
      "learning_rate": 0.003012338284489535,
      "loss": 0.0,
      "num_input_tokens_seen": 21265920,
      "step": 37445
    },
    {
      "epoch": 657.0176991150443,
      "grad_norm": 3.123362546375574e-08,
      "learning_rate": 0.0030006038628665964,
      "loss": 0.0,
      "num_input_tokens_seen": 21268192,
      "step": 37450
    },
    {
      "epoch": 657.1061946902655,
      "grad_norm": 3.082235267015676e-08,
      "learning_rate": 0.002988892110397845,
      "loss": 0.0,
      "num_input_tokens_seen": 21270656,
      "step": 37455
    },
    {
      "epoch": 657.1946902654868,
      "grad_norm": 9.903742181904818e-08,
      "learning_rate": 0.0029772030288894025,
      "loss": 0.0,
      "num_input_tokens_seen": 21274016,
      "step": 37460
    },
    {
      "epoch": 657.2831858407079,
      "grad_norm": 3.446801244422204e-08,
      "learning_rate": 0.0029655366201438438,
      "loss": 0.0,
      "num_input_tokens_seen": 21277312,
      "step": 37465
    },
    {
      "epoch": 657.3716814159292,
      "grad_norm": 1.30500424688762e-07,
      "learning_rate": 0.0029538928859602965,
      "loss": 0.0,
      "num_input_tokens_seen": 21280400,
      "step": 37470
    },
    {
      "epoch": 657.4601769911504,
      "grad_norm": 6.624212289807474e-08,
      "learning_rate": 0.002942271828134374,
      "loss": 0.0,
      "num_input_tokens_seen": 21283216,
      "step": 37475
    },
    {
      "epoch": 657.5486725663717,
      "grad_norm": 4.497356087540538e-08,
      "learning_rate": 0.00293067344845816,
      "loss": 0.0,
      "num_input_tokens_seen": 21285856,
      "step": 37480
    },
    {
      "epoch": 657.637168141593,
      "grad_norm": 6.281396025542563e-08,
      "learning_rate": 0.0029190977487202896,
      "loss": 0.0,
      "num_input_tokens_seen": 21288784,
      "step": 37485
    },
    {
      "epoch": 657.7256637168142,
      "grad_norm": 5.85421275900444e-08,
      "learning_rate": 0.0029075447307058853,
      "loss": 0.0,
      "num_input_tokens_seen": 21291312,
      "step": 37490
    },
    {
      "epoch": 657.8141592920354,
      "grad_norm": 2.2603668270448907e-08,
      "learning_rate": 0.0028960143961965722,
      "loss": 0.0,
      "num_input_tokens_seen": 21294640,
      "step": 37495
    },
    {
      "epoch": 657.9026548672566,
      "grad_norm": 3.44041026778541e-08,
      "learning_rate": 0.002884506746970461,
      "loss": 0.0,
      "num_input_tokens_seen": 21297344,
      "step": 37500
    },
    {
      "epoch": 657.9911504424779,
      "grad_norm": 5.6824134730959486e-08,
      "learning_rate": 0.0028730217848021654,
      "loss": 0.0,
      "num_input_tokens_seen": 21299936,
      "step": 37505
    },
    {
      "epoch": 658.070796460177,
      "grad_norm": 2.8667901830203846e-08,
      "learning_rate": 0.0028615595114628188,
      "loss": 0.0,
      "num_input_tokens_seen": 21302256,
      "step": 37510
    },
    {
      "epoch": 658.1592920353983,
      "grad_norm": 1.5941381548145728e-08,
      "learning_rate": 0.002850119928720074,
      "loss": 0.0,
      "num_input_tokens_seen": 21305648,
      "step": 37515
    },
    {
      "epoch": 658.2477876106195,
      "grad_norm": 2.1498847146972366e-08,
      "learning_rate": 0.0028387030383380195,
      "loss": 0.0,
      "num_input_tokens_seen": 21308576,
      "step": 37520
    },
    {
      "epoch": 658.3362831858407,
      "grad_norm": 5.689162208000198e-08,
      "learning_rate": 0.0028273088420772974,
      "loss": 0.0,
      "num_input_tokens_seen": 21311584,
      "step": 37525
    },
    {
      "epoch": 658.4247787610619,
      "grad_norm": 6.221060289135494e-08,
      "learning_rate": 0.002815937341695068,
      "loss": 0.0,
      "num_input_tokens_seen": 21314480,
      "step": 37530
    },
    {
      "epoch": 658.5132743362832,
      "grad_norm": 2.7359016385730683e-08,
      "learning_rate": 0.0028045885389448963,
      "loss": 0.0,
      "num_input_tokens_seen": 21317104,
      "step": 37535
    },
    {
      "epoch": 658.6017699115044,
      "grad_norm": 2.864973325245046e-08,
      "learning_rate": 0.002793262435576965,
      "loss": 0.0,
      "num_input_tokens_seen": 21319792,
      "step": 37540
    },
    {
      "epoch": 658.6902654867257,
      "grad_norm": 6.323876533542716e-08,
      "learning_rate": 0.0027819590333378772,
      "loss": 0.0,
      "num_input_tokens_seen": 21322816,
      "step": 37545
    },
    {
      "epoch": 658.7787610619469,
      "grad_norm": 9.240433485047106e-08,
      "learning_rate": 0.002770678333970755,
      "loss": 0.0,
      "num_input_tokens_seen": 21325536,
      "step": 37550
    },
    {
      "epoch": 658.8672566371681,
      "grad_norm": 2.3633747403550842e-08,
      "learning_rate": 0.0027594203392152573,
      "loss": 0.0,
      "num_input_tokens_seen": 21328272,
      "step": 37555
    },
    {
      "epoch": 658.9557522123894,
      "grad_norm": 5.0810282203883617e-08,
      "learning_rate": 0.002748185050807478,
      "loss": 0.0,
      "num_input_tokens_seen": 21331344,
      "step": 37560
    },
    {
      "epoch": 659.0353982300885,
      "grad_norm": 3.7327133384224e-08,
      "learning_rate": 0.002736972470480031,
      "loss": 0.0,
      "num_input_tokens_seen": 21333712,
      "step": 37565
    },
    {
      "epoch": 659.1238938053098,
      "grad_norm": 1.9719260890838086e-08,
      "learning_rate": 0.002725782599962068,
      "loss": 0.0,
      "num_input_tokens_seen": 21336400,
      "step": 37570
    },
    {
      "epoch": 659.212389380531,
      "grad_norm": 3.127897230115195e-08,
      "learning_rate": 0.0027146154409791734,
      "loss": 0.0,
      "num_input_tokens_seen": 21339440,
      "step": 37575
    },
    {
      "epoch": 659.3008849557522,
      "grad_norm": 1.7234295768275842e-08,
      "learning_rate": 0.002703470995253504,
      "loss": 0.0,
      "num_input_tokens_seen": 21342160,
      "step": 37580
    },
    {
      "epoch": 659.3893805309734,
      "grad_norm": 7.686649183824557e-08,
      "learning_rate": 0.0026923492645036184,
      "loss": 0.0,
      "num_input_tokens_seen": 21345168,
      "step": 37585
    },
    {
      "epoch": 659.4778761061947,
      "grad_norm": 8.048530020232647e-08,
      "learning_rate": 0.0026812502504446776,
      "loss": 0.0,
      "num_input_tokens_seen": 21347808,
      "step": 37590
    },
    {
      "epoch": 659.566371681416,
      "grad_norm": 2.920110375725926e-08,
      "learning_rate": 0.0026701739547882798,
      "loss": 0.0,
      "num_input_tokens_seen": 21350704,
      "step": 37595
    },
    {
      "epoch": 659.6548672566372,
      "grad_norm": 3.558504602096946e-08,
      "learning_rate": 0.0026591203792425077,
      "loss": 0.0,
      "num_input_tokens_seen": 21353776,
      "step": 37600
    },
    {
      "epoch": 659.6548672566372,
      "eval_loss": 0.7899188995361328,
      "eval_runtime": 0.9402,
      "eval_samples_per_second": 26.59,
      "eval_steps_per_second": 13.827,
      "num_input_tokens_seen": 21353776,
      "step": 37600
    },
    {
      "epoch": 659.7433628318585,
      "grad_norm": 3.45776527410635e-08,
      "learning_rate": 0.0026480895255119818,
      "loss": 0.0,
      "num_input_tokens_seen": 21356960,
      "step": 37605
    },
    {
      "epoch": 659.8318584070796,
      "grad_norm": 5.158298321816801e-08,
      "learning_rate": 0.002637081395297791,
      "loss": 0.0,
      "num_input_tokens_seen": 21359680,
      "step": 37610
    },
    {
      "epoch": 659.9203539823009,
      "grad_norm": 2.8147780994913774e-08,
      "learning_rate": 0.0026260959902975113,
      "loss": 0.0,
      "num_input_tokens_seen": 21362736,
      "step": 37615
    },
    {
      "epoch": 660.0,
      "grad_norm": 5.1334104966827e-08,
      "learning_rate": 0.00261513331220527,
      "loss": 0.0,
      "num_input_tokens_seen": 21365112,
      "step": 37620
    },
    {
      "epoch": 660.0884955752213,
      "grad_norm": 4.7880980957870634e-08,
      "learning_rate": 0.0026041933627116154,
      "loss": 0.0,
      "num_input_tokens_seen": 21368568,
      "step": 37625
    },
    {
      "epoch": 660.1769911504425,
      "grad_norm": 1.650272629660776e-08,
      "learning_rate": 0.0025932761435036476,
      "loss": 0.0,
      "num_input_tokens_seen": 21371352,
      "step": 37630
    },
    {
      "epoch": 660.2654867256637,
      "grad_norm": 2.599655068991069e-08,
      "learning_rate": 0.002582381656264904,
      "loss": 0.0,
      "num_input_tokens_seen": 21373912,
      "step": 37635
    },
    {
      "epoch": 660.3539823008849,
      "grad_norm": 1.4284664118235924e-08,
      "learning_rate": 0.0025715099026754895,
      "loss": 0.0,
      "num_input_tokens_seen": 21376952,
      "step": 37640
    },
    {
      "epoch": 660.4424778761062,
      "grad_norm": 3.1396012900586356e-08,
      "learning_rate": 0.002560660884411947,
      "loss": 0.0,
      "num_input_tokens_seen": 21379656,
      "step": 37645
    },
    {
      "epoch": 660.5309734513274,
      "grad_norm": 2.3053406295048262e-08,
      "learning_rate": 0.0025498346031473385,
      "loss": 0.0,
      "num_input_tokens_seen": 21381944,
      "step": 37650
    },
    {
      "epoch": 660.6194690265487,
      "grad_norm": 5.413679815546857e-08,
      "learning_rate": 0.0025390310605511945,
      "loss": 0.0,
      "num_input_tokens_seen": 21384792,
      "step": 37655
    },
    {
      "epoch": 660.70796460177,
      "grad_norm": 6.238354899323895e-08,
      "learning_rate": 0.0025282502582895995,
      "loss": 0.0,
      "num_input_tokens_seen": 21387288,
      "step": 37660
    },
    {
      "epoch": 660.7964601769911,
      "grad_norm": 8.262998107966268e-08,
      "learning_rate": 0.002517492198025023,
      "loss": 0.0,
      "num_input_tokens_seen": 21390072,
      "step": 37665
    },
    {
      "epoch": 660.8849557522124,
      "grad_norm": 2.4502419648797513e-08,
      "learning_rate": 0.0025067568814165554,
      "loss": 0.0,
      "num_input_tokens_seen": 21393304,
      "step": 37670
    },
    {
      "epoch": 660.9734513274336,
      "grad_norm": 1.6573912020589887e-08,
      "learning_rate": 0.0024960443101196884,
      "loss": 0.0,
      "num_input_tokens_seen": 21396600,
      "step": 37675
    },
    {
      "epoch": 661.0530973451328,
      "grad_norm": 2.2239980523863778e-08,
      "learning_rate": 0.002485354485786434,
      "loss": 0.0,
      "num_input_tokens_seen": 21398984,
      "step": 37680
    },
    {
      "epoch": 661.141592920354,
      "grad_norm": 2.081779015838947e-08,
      "learning_rate": 0.002474687410065307,
      "loss": 0.0,
      "num_input_tokens_seen": 21401896,
      "step": 37685
    },
    {
      "epoch": 661.2300884955753,
      "grad_norm": 4.423341692927352e-08,
      "learning_rate": 0.002464043084601308,
      "loss": 0.0,
      "num_input_tokens_seen": 21404488,
      "step": 37690
    },
    {
      "epoch": 661.3185840707964,
      "grad_norm": 3.7595281554558824e-08,
      "learning_rate": 0.0024534215110358915,
      "loss": 0.0,
      "num_input_tokens_seen": 21407368,
      "step": 37695
    },
    {
      "epoch": 661.4070796460177,
      "grad_norm": 3.756954569666959e-08,
      "learning_rate": 0.002442822691007096,
      "loss": 0.0,
      "num_input_tokens_seen": 21409768,
      "step": 37700
    },
    {
      "epoch": 661.4955752212389,
      "grad_norm": 2.200377302585821e-08,
      "learning_rate": 0.002432246626149348,
      "loss": 0.0,
      "num_input_tokens_seen": 21412952,
      "step": 37705
    },
    {
      "epoch": 661.5840707964602,
      "grad_norm": 2.1044506581802125e-08,
      "learning_rate": 0.002421693318093626,
      "loss": 0.0,
      "num_input_tokens_seen": 21415720,
      "step": 37710
    },
    {
      "epoch": 661.6725663716815,
      "grad_norm": 6.366853710915166e-08,
      "learning_rate": 0.0024111627684673784,
      "loss": 0.0,
      "num_input_tokens_seen": 21418968,
      "step": 37715
    },
    {
      "epoch": 661.7610619469026,
      "grad_norm": 4.862944891215193e-08,
      "learning_rate": 0.0024006549788945395,
      "loss": 0.0,
      "num_input_tokens_seen": 21421832,
      "step": 37720
    },
    {
      "epoch": 661.8495575221239,
      "grad_norm": 1.1695328083760614e-07,
      "learning_rate": 0.0023901699509955463,
      "loss": 0.0,
      "num_input_tokens_seen": 21424504,
      "step": 37725
    },
    {
      "epoch": 661.9380530973451,
      "grad_norm": 2.575555235750926e-08,
      "learning_rate": 0.0023797076863873554,
      "loss": 0.0,
      "num_input_tokens_seen": 21427624,
      "step": 37730
    },
    {
      "epoch": 662.0176991150443,
      "grad_norm": 3.465128273205664e-08,
      "learning_rate": 0.0023692681866833262,
      "loss": 0.0,
      "num_input_tokens_seen": 21430248,
      "step": 37735
    },
    {
      "epoch": 662.1061946902655,
      "grad_norm": 4.376929041427502e-08,
      "learning_rate": 0.0023588514534934046,
      "loss": 0.0,
      "num_input_tokens_seen": 21433352,
      "step": 37740
    },
    {
      "epoch": 662.1946902654868,
      "grad_norm": 3.0929502514709384e-08,
      "learning_rate": 0.002348457488423955,
      "loss": 0.0,
      "num_input_tokens_seen": 21436120,
      "step": 37745
    },
    {
      "epoch": 662.2831858407079,
      "grad_norm": 3.898851730355091e-08,
      "learning_rate": 0.0023380862930778624,
      "loss": 0.0,
      "num_input_tokens_seen": 21438856,
      "step": 37750
    },
    {
      "epoch": 662.3716814159292,
      "grad_norm": 1.7803250429437867e-08,
      "learning_rate": 0.0023277378690545135,
      "loss": 0.0,
      "num_input_tokens_seen": 21441576,
      "step": 37755
    },
    {
      "epoch": 662.4601769911504,
      "grad_norm": 3.583700447506999e-08,
      "learning_rate": 0.0023174122179497325,
      "loss": 0.0,
      "num_input_tokens_seen": 21444552,
      "step": 37760
    },
    {
      "epoch": 662.5486725663717,
      "grad_norm": 4.353854166083693e-08,
      "learning_rate": 0.0023071093413558784,
      "loss": 0.0,
      "num_input_tokens_seen": 21447208,
      "step": 37765
    },
    {
      "epoch": 662.637168141593,
      "grad_norm": 4.518368612593804e-08,
      "learning_rate": 0.002296829240861814,
      "loss": 0.0,
      "num_input_tokens_seen": 21450504,
      "step": 37770
    },
    {
      "epoch": 662.7256637168142,
      "grad_norm": 5.4693071405154114e-08,
      "learning_rate": 0.002286571918052821,
      "loss": 0.0,
      "num_input_tokens_seen": 21453704,
      "step": 37775
    },
    {
      "epoch": 662.8141592920354,
      "grad_norm": 2.2519618170235844e-08,
      "learning_rate": 0.0022763373745107174,
      "loss": 0.0,
      "num_input_tokens_seen": 21456856,
      "step": 37780
    },
    {
      "epoch": 662.9026548672566,
      "grad_norm": 4.508081019594101e-08,
      "learning_rate": 0.0022661256118138074,
      "loss": 0.0,
      "num_input_tokens_seen": 21459224,
      "step": 37785
    },
    {
      "epoch": 662.9911504424779,
      "grad_norm": 1.9781850824074354e-08,
      "learning_rate": 0.0022559366315368645,
      "loss": 0.0,
      "num_input_tokens_seen": 21462200,
      "step": 37790
    },
    {
      "epoch": 663.070796460177,
      "grad_norm": 2.007508648205203e-08,
      "learning_rate": 0.002245770435251182,
      "loss": 0.0,
      "num_input_tokens_seen": 21464776,
      "step": 37795
    },
    {
      "epoch": 663.1592920353983,
      "grad_norm": 4.4238692709086536e-08,
      "learning_rate": 0.002235627024524456,
      "loss": 0.0,
      "num_input_tokens_seen": 21467368,
      "step": 37800
    },
    {
      "epoch": 663.1592920353983,
      "eval_loss": 0.7847999930381775,
      "eval_runtime": 0.9529,
      "eval_samples_per_second": 26.235,
      "eval_steps_per_second": 13.642,
      "num_input_tokens_seen": 21467368,
      "step": 37800
    },
    {
      "epoch": 663.2477876106195,
      "grad_norm": 2.649870012305655e-08,
      "learning_rate": 0.0022255064009209847,
      "loss": 0.0,
      "num_input_tokens_seen": 21469960,
      "step": 37805
    },
    {
      "epoch": 663.3362831858407,
      "grad_norm": 1.9268748374656752e-08,
      "learning_rate": 0.0022154085660014864,
      "loss": 0.0,
      "num_input_tokens_seen": 21473000,
      "step": 37810
    },
    {
      "epoch": 663.4247787610619,
      "grad_norm": 2.5421618587984085e-08,
      "learning_rate": 0.0022053335213231494,
      "loss": 0.0,
      "num_input_tokens_seen": 21475912,
      "step": 37815
    },
    {
      "epoch": 663.5132743362832,
      "grad_norm": 4.5663337999712894e-08,
      "learning_rate": 0.002195281268439697,
      "loss": 0.0,
      "num_input_tokens_seen": 21479224,
      "step": 37820
    },
    {
      "epoch": 663.6017699115044,
      "grad_norm": 1.003513006025969e-07,
      "learning_rate": 0.002185251808901306,
      "loss": 0.0,
      "num_input_tokens_seen": 21482200,
      "step": 37825
    },
    {
      "epoch": 663.6902654867257,
      "grad_norm": 4.103519302134373e-08,
      "learning_rate": 0.0021752451442546227,
      "loss": 0.0,
      "num_input_tokens_seen": 21485160,
      "step": 37830
    },
    {
      "epoch": 663.7787610619469,
      "grad_norm": 2.9529447331810843e-08,
      "learning_rate": 0.0021652612760428456,
      "loss": 0.0,
      "num_input_tokens_seen": 21488104,
      "step": 37835
    },
    {
      "epoch": 663.8672566371681,
      "grad_norm": 3.1629589614112774e-08,
      "learning_rate": 0.0021553002058055603,
      "loss": 0.0,
      "num_input_tokens_seen": 21490952,
      "step": 37840
    },
    {
      "epoch": 663.9557522123894,
      "grad_norm": 5.90975872682975e-08,
      "learning_rate": 0.0021453619350789376,
      "loss": 0.0,
      "num_input_tokens_seen": 21493880,
      "step": 37845
    },
    {
      "epoch": 664.0353982300885,
      "grad_norm": 4.1690952912176726e-08,
      "learning_rate": 0.0021354464653955516,
      "loss": 0.0,
      "num_input_tokens_seen": 21496280,
      "step": 37850
    },
    {
      "epoch": 664.1238938053098,
      "grad_norm": 2.60836010568255e-08,
      "learning_rate": 0.002125553798284513,
      "loss": 0.0,
      "num_input_tokens_seen": 21499048,
      "step": 37855
    },
    {
      "epoch": 664.212389380531,
      "grad_norm": 2.418865641118373e-08,
      "learning_rate": 0.002115683935271384,
      "loss": 0.0,
      "num_input_tokens_seen": 21502184,
      "step": 37860
    },
    {
      "epoch": 664.3008849557522,
      "grad_norm": 1.6188881346579365e-08,
      "learning_rate": 0.0021058368778782144,
      "loss": 0.0,
      "num_input_tokens_seen": 21504856,
      "step": 37865
    },
    {
      "epoch": 664.3893805309734,
      "grad_norm": 2.4827597755461284e-08,
      "learning_rate": 0.002096012627623539,
      "loss": 0.0,
      "num_input_tokens_seen": 21508200,
      "step": 37870
    },
    {
      "epoch": 664.4778761061947,
      "grad_norm": 5.0405848384116325e-08,
      "learning_rate": 0.00208621118602243,
      "loss": 0.0,
      "num_input_tokens_seen": 21511096,
      "step": 37875
    },
    {
      "epoch": 664.566371681416,
      "grad_norm": 5.8483117015839525e-08,
      "learning_rate": 0.002076432554586327,
      "loss": 0.0,
      "num_input_tokens_seen": 21513864,
      "step": 37880
    },
    {
      "epoch": 664.6548672566372,
      "grad_norm": 3.786524160886984e-08,
      "learning_rate": 0.002066676734823258,
      "loss": 0.0,
      "num_input_tokens_seen": 21516584,
      "step": 37885
    },
    {
      "epoch": 664.7433628318585,
      "grad_norm": 1.0048010778973548e-07,
      "learning_rate": 0.0020569437282376866,
      "loss": 0.0,
      "num_input_tokens_seen": 21519432,
      "step": 37890
    },
    {
      "epoch": 664.8318584070796,
      "grad_norm": 4.706843625967849e-08,
      "learning_rate": 0.002047233536330545,
      "loss": 0.0,
      "num_input_tokens_seen": 21522168,
      "step": 37895
    },
    {
      "epoch": 664.9203539823009,
      "grad_norm": 8.064790080197781e-08,
      "learning_rate": 0.0020375461605993015,
      "loss": 0.0,
      "num_input_tokens_seen": 21525368,
      "step": 37900
    },
    {
      "epoch": 665.0,
      "grad_norm": 1.9485643321104362e-08,
      "learning_rate": 0.002027881602537845,
      "loss": 0.0,
      "num_input_tokens_seen": 21527784,
      "step": 37905
    },
    {
      "epoch": 665.0884955752213,
      "grad_norm": 2.4993514813331785e-08,
      "learning_rate": 0.002018239863636567,
      "loss": 0.0,
      "num_input_tokens_seen": 21530072,
      "step": 37910
    },
    {
      "epoch": 665.1769911504425,
      "grad_norm": 7.861197559577704e-08,
      "learning_rate": 0.002008620945382378,
      "loss": 0.0,
      "num_input_tokens_seen": 21532872,
      "step": 37915
    },
    {
      "epoch": 665.2654867256637,
      "grad_norm": 1.2556266426599905e-07,
      "learning_rate": 0.001999024849258607,
      "loss": 0.0,
      "num_input_tokens_seen": 21535672,
      "step": 37920
    },
    {
      "epoch": 665.3539823008849,
      "grad_norm": 5.137002645483335e-08,
      "learning_rate": 0.001989451576745105,
      "loss": 0.0,
      "num_input_tokens_seen": 21538840,
      "step": 37925
    },
    {
      "epoch": 665.4424778761062,
      "grad_norm": 3.264521808432619e-08,
      "learning_rate": 0.00197990112931819,
      "loss": 0.0,
      "num_input_tokens_seen": 21541464,
      "step": 37930
    },
    {
      "epoch": 665.5309734513274,
      "grad_norm": 6.84740228962255e-08,
      "learning_rate": 0.0019703735084506345,
      "loss": 0.0,
      "num_input_tokens_seen": 21544216,
      "step": 37935
    },
    {
      "epoch": 665.6194690265487,
      "grad_norm": 9.753097884868112e-08,
      "learning_rate": 0.001960868715611763,
      "loss": 0.0,
      "num_input_tokens_seen": 21547576,
      "step": 37940
    },
    {
      "epoch": 665.70796460177,
      "grad_norm": 2.3686125061317398e-08,
      "learning_rate": 0.0019513867522673034,
      "loss": 0.0,
      "num_input_tokens_seen": 21550392,
      "step": 37945
    },
    {
      "epoch": 665.7964601769911,
      "grad_norm": 3.4306431473396515e-08,
      "learning_rate": 0.001941927619879502,
      "loss": 0.0,
      "num_input_tokens_seen": 21553416,
      "step": 37950
    },
    {
      "epoch": 665.8849557522124,
      "grad_norm": 3.636456824551715e-08,
      "learning_rate": 0.0019324913199070758,
      "loss": 0.0,
      "num_input_tokens_seen": 21556792,
      "step": 37955
    },
    {
      "epoch": 665.9734513274336,
      "grad_norm": 4.344121151689251e-08,
      "learning_rate": 0.0019230778538052106,
      "loss": 0.0,
      "num_input_tokens_seen": 21559416,
      "step": 37960
    },
    {
      "epoch": 666.0530973451328,
      "grad_norm": 7.866967166592076e-08,
      "learning_rate": 0.0019136872230255952,
      "loss": 0.0,
      "num_input_tokens_seen": 21561768,
      "step": 37965
    },
    {
      "epoch": 666.141592920354,
      "grad_norm": 6.551740483473623e-08,
      "learning_rate": 0.0019043194290164045,
      "loss": 0.0,
      "num_input_tokens_seen": 21564568,
      "step": 37970
    },
    {
      "epoch": 666.2300884955753,
      "grad_norm": 2.868379844755964e-08,
      "learning_rate": 0.0018949744732222162,
      "loss": 0.0,
      "num_input_tokens_seen": 21567640,
      "step": 37975
    },
    {
      "epoch": 666.3185840707964,
      "grad_norm": 2.7418002090939808e-08,
      "learning_rate": 0.0018856523570841776,
      "loss": 0.0,
      "num_input_tokens_seen": 21570568,
      "step": 37980
    },
    {
      "epoch": 666.4070796460177,
      "grad_norm": 5.185851748024106e-08,
      "learning_rate": 0.0018763530820398555,
      "loss": 0.0,
      "num_input_tokens_seen": 21573176,
      "step": 37985
    },
    {
      "epoch": 666.4955752212389,
      "grad_norm": 6.124844986743483e-08,
      "learning_rate": 0.0018670766495233525,
      "loss": 0.0,
      "num_input_tokens_seen": 21576184,
      "step": 37990
    },
    {
      "epoch": 666.5840707964602,
      "grad_norm": 1.6464724694742472e-08,
      "learning_rate": 0.001857823060965158,
      "loss": 0.0,
      "num_input_tokens_seen": 21578840,
      "step": 37995
    },
    {
      "epoch": 666.6725663716815,
      "grad_norm": 3.086255162543239e-08,
      "learning_rate": 0.0018485923177923467,
      "loss": 0.0,
      "num_input_tokens_seen": 21581512,
      "step": 38000
    },
    {
      "epoch": 666.6725663716815,
      "eval_loss": 0.7813501954078674,
      "eval_runtime": 0.9546,
      "eval_samples_per_second": 26.189,
      "eval_steps_per_second": 13.618,
      "num_input_tokens_seen": 21581512,
      "step": 38000
    },
    {
      "epoch": 666.7610619469026,
      "grad_norm": 5.436567462879793e-08,
      "learning_rate": 0.001839384421428364,
      "loss": 0.0,
      "num_input_tokens_seen": 21584920,
      "step": 38005
    },
    {
      "epoch": 666.8495575221239,
      "grad_norm": 2.973313861787119e-08,
      "learning_rate": 0.0018301993732932065,
      "loss": 0.0,
      "num_input_tokens_seen": 21587512,
      "step": 38010
    },
    {
      "epoch": 666.9380530973451,
      "grad_norm": 5.375415312869336e-08,
      "learning_rate": 0.0018210371748033248,
      "loss": 0.0,
      "num_input_tokens_seen": 21590312,
      "step": 38015
    },
    {
      "epoch": 667.0176991150443,
      "grad_norm": 8.453049815670965e-08,
      "learning_rate": 0.0018118978273716556,
      "loss": 0.0,
      "num_input_tokens_seen": 21592400,
      "step": 38020
    },
    {
      "epoch": 667.1061946902655,
      "grad_norm": 1.632611379420723e-08,
      "learning_rate": 0.001802781332407588,
      "loss": 0.0,
      "num_input_tokens_seen": 21595792,
      "step": 38025
    },
    {
      "epoch": 667.1946902654868,
      "grad_norm": 6.664970442216145e-08,
      "learning_rate": 0.0017936876913169806,
      "loss": 0.0,
      "num_input_tokens_seen": 21598464,
      "step": 38030
    },
    {
      "epoch": 667.2831858407079,
      "grad_norm": 2.0106444509337962e-08,
      "learning_rate": 0.0017846169055022287,
      "loss": 0.0,
      "num_input_tokens_seen": 21601408,
      "step": 38035
    },
    {
      "epoch": 667.3716814159292,
      "grad_norm": 1.1351018258665135e-07,
      "learning_rate": 0.0017755689763621295,
      "loss": 0.0,
      "num_input_tokens_seen": 21604192,
      "step": 38040
    },
    {
      "epoch": 667.4601769911504,
      "grad_norm": 3.233967760252199e-08,
      "learning_rate": 0.0017665439052920173,
      "loss": 0.0,
      "num_input_tokens_seen": 21607360,
      "step": 38045
    },
    {
      "epoch": 667.5486725663717,
      "grad_norm": 4.887234084094416e-08,
      "learning_rate": 0.0017575416936836286,
      "loss": 0.0,
      "num_input_tokens_seen": 21610000,
      "step": 38050
    },
    {
      "epoch": 667.637168141593,
      "grad_norm": 3.4989188435474716e-08,
      "learning_rate": 0.0017485623429252528,
      "loss": 0.0,
      "num_input_tokens_seen": 21613296,
      "step": 38055
    },
    {
      "epoch": 667.7256637168142,
      "grad_norm": 3.648264268463208e-08,
      "learning_rate": 0.0017396058544016156,
      "loss": 0.0,
      "num_input_tokens_seen": 21615648,
      "step": 38060
    },
    {
      "epoch": 667.8141592920354,
      "grad_norm": 3.75050248635489e-08,
      "learning_rate": 0.0017306722294938958,
      "loss": 0.0,
      "num_input_tokens_seen": 21618352,
      "step": 38065
    },
    {
      "epoch": 667.9026548672566,
      "grad_norm": 1.761000945066371e-08,
      "learning_rate": 0.0017217614695798078,
      "loss": 0.0,
      "num_input_tokens_seen": 21621520,
      "step": 38070
    },
    {
      "epoch": 667.9911504424779,
      "grad_norm": 1.8809098278893543e-08,
      "learning_rate": 0.001712873576033469,
      "loss": 0.0,
      "num_input_tokens_seen": 21624368,
      "step": 38075
    },
    {
      "epoch": 668.070796460177,
      "grad_norm": 4.82365010157082e-08,
      "learning_rate": 0.0017040085502255163,
      "loss": 0.0,
      "num_input_tokens_seen": 21626816,
      "step": 38080
    },
    {
      "epoch": 668.1592920353983,
      "grad_norm": 6.67929143105539e-08,
      "learning_rate": 0.0016951663935230565,
      "loss": 0.0,
      "num_input_tokens_seen": 21629504,
      "step": 38085
    },
    {
      "epoch": 668.2477876106195,
      "grad_norm": 2.9938732382106537e-08,
      "learning_rate": 0.0016863471072896485,
      "loss": 0.0,
      "num_input_tokens_seen": 21632416,
      "step": 38090
    },
    {
      "epoch": 668.3362831858407,
      "grad_norm": 2.9025962078321754e-08,
      "learning_rate": 0.0016775506928853377,
      "loss": 0.0,
      "num_input_tokens_seen": 21635264,
      "step": 38095
    },
    {
      "epoch": 668.4247787610619,
      "grad_norm": 4.615922577499987e-08,
      "learning_rate": 0.001668777151666656,
      "loss": 0.0,
      "num_input_tokens_seen": 21638192,
      "step": 38100
    },
    {
      "epoch": 668.5132743362832,
      "grad_norm": 3.4986676666903804e-08,
      "learning_rate": 0.0016600264849865709,
      "loss": 0.0,
      "num_input_tokens_seen": 21641504,
      "step": 38105
    },
    {
      "epoch": 668.6017699115044,
      "grad_norm": 3.149193261720029e-08,
      "learning_rate": 0.0016512986941945695,
      "loss": 0.0,
      "num_input_tokens_seen": 21644672,
      "step": 38110
    },
    {
      "epoch": 668.6902654867257,
      "grad_norm": 3.8721751138837135e-08,
      "learning_rate": 0.0016425937806365753,
      "loss": 0.0,
      "num_input_tokens_seen": 21647200,
      "step": 38115
    },
    {
      "epoch": 668.7787610619469,
      "grad_norm": 3.9269011153919564e-08,
      "learning_rate": 0.0016339117456549979,
      "loss": 0.0,
      "num_input_tokens_seen": 21649808,
      "step": 38120
    },
    {
      "epoch": 668.8672566371681,
      "grad_norm": 3.5385632202178385e-08,
      "learning_rate": 0.0016252525905886995,
      "loss": 0.0,
      "num_input_tokens_seen": 21652784,
      "step": 38125
    },
    {
      "epoch": 668.9557522123894,
      "grad_norm": 3.6932092939423455e-08,
      "learning_rate": 0.0016166163167730617,
      "loss": 0.0,
      "num_input_tokens_seen": 21655456,
      "step": 38130
    },
    {
      "epoch": 669.0353982300885,
      "grad_norm": 1.7174599520330958e-08,
      "learning_rate": 0.0016080029255398864,
      "loss": 0.0,
      "num_input_tokens_seen": 21657464,
      "step": 38135
    },
    {
      "epoch": 669.1238938053098,
      "grad_norm": 2.0777921605485972e-08,
      "learning_rate": 0.0015994124182174606,
      "loss": 0.0,
      "num_input_tokens_seen": 21660184,
      "step": 38140
    },
    {
      "epoch": 669.212389380531,
      "grad_norm": 5.614894504901713e-08,
      "learning_rate": 0.001590844796130575,
      "loss": 0.0,
      "num_input_tokens_seen": 21662776,
      "step": 38145
    },
    {
      "epoch": 669.3008849557522,
      "grad_norm": 4.0771780618342746e-08,
      "learning_rate": 0.001582300060600439,
      "loss": 0.0,
      "num_input_tokens_seen": 21665384,
      "step": 38150
    },
    {
      "epoch": 669.3893805309734,
      "grad_norm": 3.5777517837232153e-08,
      "learning_rate": 0.0015737782129447652,
      "loss": 0.0,
      "num_input_tokens_seen": 21668760,
      "step": 38155
    },
    {
      "epoch": 669.4778761061947,
      "grad_norm": 4.314702906071943e-08,
      "learning_rate": 0.0015652792544777361,
      "loss": 0.0,
      "num_input_tokens_seen": 21671592,
      "step": 38160
    },
    {
      "epoch": 669.566371681416,
      "grad_norm": 3.5235938611322126e-08,
      "learning_rate": 0.0015568031865099863,
      "loss": 0.0,
      "num_input_tokens_seen": 21674664,
      "step": 38165
    },
    {
      "epoch": 669.6548672566372,
      "grad_norm": 2.869182402776005e-08,
      "learning_rate": 0.0015483500103486369,
      "loss": 0.0,
      "num_input_tokens_seen": 21677672,
      "step": 38170
    },
    {
      "epoch": 669.7433628318585,
      "grad_norm": 4.279128873463378e-08,
      "learning_rate": 0.0015399197272972787,
      "loss": 0.0,
      "num_input_tokens_seen": 21680728,
      "step": 38175
    },
    {
      "epoch": 669.8318584070796,
      "grad_norm": 2.407829491346547e-08,
      "learning_rate": 0.0015315123386559714,
      "loss": 0.0,
      "num_input_tokens_seen": 21683464,
      "step": 38180
    },
    {
      "epoch": 669.9203539823009,
      "grad_norm": 2.7190120377440508e-08,
      "learning_rate": 0.0015231278457212283,
      "loss": 0.0,
      "num_input_tokens_seen": 21686568,
      "step": 38185
    },
    {
      "epoch": 670.0,
      "grad_norm": 1.3715221314214432e-07,
      "learning_rate": 0.001514766249786048,
      "loss": 0.0,
      "num_input_tokens_seen": 21688728,
      "step": 38190
    },
    {
      "epoch": 670.0884955752213,
      "grad_norm": 5.9099864557765613e-08,
      "learning_rate": 0.0015064275521398994,
      "loss": 0.0,
      "num_input_tokens_seen": 21691496,
      "step": 38195
    },
    {
      "epoch": 670.1769911504425,
      "grad_norm": 3.1588239579605215e-08,
      "learning_rate": 0.0014981117540686872,
      "loss": 0.0,
      "num_input_tokens_seen": 21694376,
      "step": 38200
    },
    {
      "epoch": 670.1769911504425,
      "eval_loss": 0.7832437753677368,
      "eval_runtime": 0.9531,
      "eval_samples_per_second": 26.23,
      "eval_steps_per_second": 13.64,
      "num_input_tokens_seen": 21694376,
      "step": 38200
    },
    {
      "epoch": 670.2654867256637,
      "grad_norm": 3.499761191960715e-08,
      "learning_rate": 0.0014898188568548687,
      "loss": 0.0,
      "num_input_tokens_seen": 21697336,
      "step": 38205
    },
    {
      "epoch": 670.3539823008849,
      "grad_norm": 4.272216003187168e-08,
      "learning_rate": 0.0014815488617772542,
      "loss": 0.0,
      "num_input_tokens_seen": 21700440,
      "step": 38210
    },
    {
      "epoch": 670.4424778761062,
      "grad_norm": 2.989823855159557e-08,
      "learning_rate": 0.0014733017701112072,
      "loss": 0.0,
      "num_input_tokens_seen": 21703448,
      "step": 38215
    },
    {
      "epoch": 670.5309734513274,
      "grad_norm": 7.983527439137106e-08,
      "learning_rate": 0.0014650775831285435,
      "loss": 0.0,
      "num_input_tokens_seen": 21706408,
      "step": 38220
    },
    {
      "epoch": 670.6194690265487,
      "grad_norm": 5.3262475319115765e-08,
      "learning_rate": 0.001456876302097515,
      "loss": 0.0,
      "num_input_tokens_seen": 21709368,
      "step": 38225
    },
    {
      "epoch": 670.70796460177,
      "grad_norm": 3.7635768279642434e-08,
      "learning_rate": 0.0014486979282828604,
      "loss": 0.0,
      "num_input_tokens_seen": 21712104,
      "step": 38230
    },
    {
      "epoch": 670.7964601769911,
      "grad_norm": 3.2032481556143466e-08,
      "learning_rate": 0.001440542462945804,
      "loss": 0.0,
      "num_input_tokens_seen": 21715304,
      "step": 38235
    },
    {
      "epoch": 670.8849557522124,
      "grad_norm": 7.429822090898597e-08,
      "learning_rate": 0.0014324099073440232,
      "loss": 0.0,
      "num_input_tokens_seen": 21718184,
      "step": 38240
    },
    {
      "epoch": 670.9734513274336,
      "grad_norm": 7.701578397245612e-08,
      "learning_rate": 0.0014243002627316482,
      "loss": 0.0,
      "num_input_tokens_seen": 21720712,
      "step": 38245
    },
    {
      "epoch": 671.0530973451328,
      "grad_norm": 6.362460425179961e-08,
      "learning_rate": 0.0014162135303592781,
      "loss": 0.0,
      "num_input_tokens_seen": 21723472,
      "step": 38250
    },
    {
      "epoch": 671.141592920354,
      "grad_norm": 4.059573655013082e-08,
      "learning_rate": 0.001408149711474016,
      "loss": 0.0,
      "num_input_tokens_seen": 21726368,
      "step": 38255
    },
    {
      "epoch": 671.2300884955753,
      "grad_norm": 3.114764624001509e-08,
      "learning_rate": 0.0014001088073193834,
      "loss": 0.0,
      "num_input_tokens_seen": 21729168,
      "step": 38260
    },
    {
      "epoch": 671.3185840707964,
      "grad_norm": 6.347609371459839e-08,
      "learning_rate": 0.0013920908191354052,
      "loss": 0.0,
      "num_input_tokens_seen": 21731824,
      "step": 38265
    },
    {
      "epoch": 671.4070796460177,
      "grad_norm": 8.312285082467952e-09,
      "learning_rate": 0.001384095748158526,
      "loss": 0.0,
      "num_input_tokens_seen": 21734896,
      "step": 38270
    },
    {
      "epoch": 671.4955752212389,
      "grad_norm": 3.9911959959226806e-08,
      "learning_rate": 0.0013761235956217255,
      "loss": 0.0,
      "num_input_tokens_seen": 21737392,
      "step": 38275
    },
    {
      "epoch": 671.5840707964602,
      "grad_norm": 1.497691570762072e-08,
      "learning_rate": 0.0013681743627543873,
      "loss": 0.0,
      "num_input_tokens_seen": 21740080,
      "step": 38280
    },
    {
      "epoch": 671.6725663716815,
      "grad_norm": 4.503295514268757e-08,
      "learning_rate": 0.001360248050782381,
      "loss": 0.0,
      "num_input_tokens_seen": 21743056,
      "step": 38285
    },
    {
      "epoch": 671.7610619469026,
      "grad_norm": 5.029926342103863e-08,
      "learning_rate": 0.001352344660928062,
      "loss": 0.0,
      "num_input_tokens_seen": 21745856,
      "step": 38290
    },
    {
      "epoch": 671.8495575221239,
      "grad_norm": 1.0662675720141124e-07,
      "learning_rate": 0.0013444641944102052,
      "loss": 0.0,
      "num_input_tokens_seen": 21748784,
      "step": 38295
    },
    {
      "epoch": 671.9380530973451,
      "grad_norm": 4.2400344568704895e-08,
      "learning_rate": 0.0013366066524441056,
      "loss": 0.0,
      "num_input_tokens_seen": 21751360,
      "step": 38300
    },
    {
      "epoch": 672.0176991150443,
      "grad_norm": 5.79097800823547e-08,
      "learning_rate": 0.0013287720362414768,
      "loss": 0.0,
      "num_input_tokens_seen": 21754216,
      "step": 38305
    },
    {
      "epoch": 672.1061946902655,
      "grad_norm": 2.3854747510654306e-08,
      "learning_rate": 0.0013209603470105025,
      "loss": 0.0,
      "num_input_tokens_seen": 21757096,
      "step": 38310
    },
    {
      "epoch": 672.1946902654868,
      "grad_norm": 3.186459807125175e-08,
      "learning_rate": 0.0013131715859558857,
      "loss": 0.0,
      "num_input_tokens_seen": 21759656,
      "step": 38315
    },
    {
      "epoch": 672.2831858407079,
      "grad_norm": 4.6508581874604715e-08,
      "learning_rate": 0.001305405754278699,
      "loss": 0.0,
      "num_input_tokens_seen": 21762728,
      "step": 38320
    },
    {
      "epoch": 672.3716814159292,
      "grad_norm": 3.7767328819882096e-08,
      "learning_rate": 0.0012976628531765843,
      "loss": 0.0,
      "num_input_tokens_seen": 21765384,
      "step": 38325
    },
    {
      "epoch": 672.4601769911504,
      "grad_norm": 2.351332639705106e-08,
      "learning_rate": 0.0012899428838435533,
      "loss": 0.0,
      "num_input_tokens_seen": 21768072,
      "step": 38330
    },
    {
      "epoch": 672.5486725663717,
      "grad_norm": 3.540122506251464e-08,
      "learning_rate": 0.001282245847470137,
      "loss": 0.0,
      "num_input_tokens_seen": 21770776,
      "step": 38335
    },
    {
      "epoch": 672.637168141593,
      "grad_norm": 2.3661606007863156e-08,
      "learning_rate": 0.001274571745243319,
      "loss": 0.0,
      "num_input_tokens_seen": 21773688,
      "step": 38340
    },
    {
      "epoch": 672.7256637168142,
      "grad_norm": 3.064680953457355e-08,
      "learning_rate": 0.0012669205783465364,
      "loss": 0.0,
      "num_input_tokens_seen": 21776520,
      "step": 38345
    },
    {
      "epoch": 672.8141592920354,
      "grad_norm": 5.077304976452979e-08,
      "learning_rate": 0.001259292347959695,
      "loss": 0.0,
      "num_input_tokens_seen": 21779912,
      "step": 38350
    },
    {
      "epoch": 672.9026548672566,
      "grad_norm": 3.058561759416989e-08,
      "learning_rate": 0.0012516870552591707,
      "loss": 0.0,
      "num_input_tokens_seen": 21783064,
      "step": 38355
    },
    {
      "epoch": 672.9911504424779,
      "grad_norm": 4.562546607189688e-08,
      "learning_rate": 0.001244104701417792,
      "loss": 0.0,
      "num_input_tokens_seen": 21785752,
      "step": 38360
    },
    {
      "epoch": 673.070796460177,
      "grad_norm": 4.18063059726137e-08,
      "learning_rate": 0.0012365452876048565,
      "loss": 0.0,
      "num_input_tokens_seen": 21788280,
      "step": 38365
    },
    {
      "epoch": 673.1592920353983,
      "grad_norm": 5.460309893123849e-08,
      "learning_rate": 0.001229008814986099,
      "loss": 0.0,
      "num_input_tokens_seen": 21790888,
      "step": 38370
    },
    {
      "epoch": 673.2477876106195,
      "grad_norm": 6.222386161880422e-08,
      "learning_rate": 0.0012214952847237725,
      "loss": 0.0,
      "num_input_tokens_seen": 21793880,
      "step": 38375
    },
    {
      "epoch": 673.3362831858407,
      "grad_norm": 7.134742929792992e-08,
      "learning_rate": 0.0012140046979765339,
      "loss": 0.0,
      "num_input_tokens_seen": 21796712,
      "step": 38380
    },
    {
      "epoch": 673.4247787610619,
      "grad_norm": 3.7434595867580356e-08,
      "learning_rate": 0.0012065370558995258,
      "loss": 0.0,
      "num_input_tokens_seen": 21799816,
      "step": 38385
    },
    {
      "epoch": 673.5132743362832,
      "grad_norm": 3.837060447153817e-08,
      "learning_rate": 0.0011990923596443602,
      "loss": 0.0,
      "num_input_tokens_seen": 21802776,
      "step": 38390
    },
    {
      "epoch": 673.6017699115044,
      "grad_norm": 2.7394044366246817e-08,
      "learning_rate": 0.001191670610359119,
      "loss": 0.0,
      "num_input_tokens_seen": 21805768,
      "step": 38395
    },
    {
      "epoch": 673.6902654867257,
      "grad_norm": 3.340823440112217e-08,
      "learning_rate": 0.0011842718091882865,
      "loss": 0.0,
      "num_input_tokens_seen": 21808568,
      "step": 38400
    },
    {
      "epoch": 673.6902654867257,
      "eval_loss": 0.7953433990478516,
      "eval_runtime": 0.9416,
      "eval_samples_per_second": 26.552,
      "eval_steps_per_second": 13.807,
      "num_input_tokens_seen": 21808568,
      "step": 38400
    },
    {
      "epoch": 673.7787610619469,
      "grad_norm": 3.83823142158235e-08,
      "learning_rate": 0.0011768959572729,
      "loss": 0.0,
      "num_input_tokens_seen": 21811496,
      "step": 38405
    },
    {
      "epoch": 673.8672566371681,
      "grad_norm": 3.2094856550202167e-08,
      "learning_rate": 0.001169543055750366,
      "loss": 0.0,
      "num_input_tokens_seen": 21813784,
      "step": 38410
    },
    {
      "epoch": 673.9557522123894,
      "grad_norm": 3.650057323056899e-08,
      "learning_rate": 0.0011622131057546115,
      "loss": 0.0,
      "num_input_tokens_seen": 21817224,
      "step": 38415
    },
    {
      "epoch": 674.0353982300885,
      "grad_norm": 3.2848998188228506e-08,
      "learning_rate": 0.0011549061084160316,
      "loss": 0.0,
      "num_input_tokens_seen": 21819688,
      "step": 38420
    },
    {
      "epoch": 674.1238938053098,
      "grad_norm": 6.140194130921373e-08,
      "learning_rate": 0.0011476220648614088,
      "loss": 0.0,
      "num_input_tokens_seen": 21822312,
      "step": 38425
    },
    {
      "epoch": 674.212389380531,
      "grad_norm": 2.2812741917732637e-08,
      "learning_rate": 0.0011403609762140777,
      "loss": 0.0,
      "num_input_tokens_seen": 21825144,
      "step": 38430
    },
    {
      "epoch": 674.3008849557522,
      "grad_norm": 7.511926725101148e-08,
      "learning_rate": 0.0011331228435937756,
      "loss": 0.0,
      "num_input_tokens_seen": 21828008,
      "step": 38435
    },
    {
      "epoch": 674.3893805309734,
      "grad_norm": 1.225963899287308e-07,
      "learning_rate": 0.0011259076681166935,
      "loss": 0.0,
      "num_input_tokens_seen": 21831432,
      "step": 38440
    },
    {
      "epoch": 674.4778761061947,
      "grad_norm": 5.003910530376743e-08,
      "learning_rate": 0.0011187154508955244,
      "loss": 0.0,
      "num_input_tokens_seen": 21834472,
      "step": 38445
    },
    {
      "epoch": 674.566371681416,
      "grad_norm": 6.247152839478076e-08,
      "learning_rate": 0.001111546193039381,
      "loss": 0.0,
      "num_input_tokens_seen": 21836840,
      "step": 38450
    },
    {
      "epoch": 674.6548672566372,
      "grad_norm": 2.8499313131646886e-08,
      "learning_rate": 0.0011043998956538792,
      "loss": 0.0,
      "num_input_tokens_seen": 21839992,
      "step": 38455
    },
    {
      "epoch": 674.7433628318585,
      "grad_norm": 1.9929581540623076e-08,
      "learning_rate": 0.0010972765598410538,
      "loss": 0.0,
      "num_input_tokens_seen": 21842968,
      "step": 38460
    },
    {
      "epoch": 674.8318584070796,
      "grad_norm": 2.3524480141645654e-08,
      "learning_rate": 0.0010901761866993931,
      "loss": 0.0,
      "num_input_tokens_seen": 21845400,
      "step": 38465
    },
    {
      "epoch": 674.9203539823009,
      "grad_norm": 2.9118906397229694e-08,
      "learning_rate": 0.0010830987773238876,
      "loss": 0.0,
      "num_input_tokens_seen": 21848168,
      "step": 38470
    },
    {
      "epoch": 675.0,
      "grad_norm": 1.0766945024442975e-07,
      "learning_rate": 0.0010760443328059644,
      "loss": 0.0,
      "num_input_tokens_seen": 21850760,
      "step": 38475
    },
    {
      "epoch": 675.0884955752213,
      "grad_norm": 3.585607544209779e-08,
      "learning_rate": 0.001069012854233503,
      "loss": 0.0,
      "num_input_tokens_seen": 21853336,
      "step": 38480
    },
    {
      "epoch": 675.1769911504425,
      "grad_norm": 8.061828538075133e-08,
      "learning_rate": 0.0010620043426908365,
      "loss": 0.0,
      "num_input_tokens_seen": 21856296,
      "step": 38485
    },
    {
      "epoch": 675.2654867256637,
      "grad_norm": 6.746235214905028e-08,
      "learning_rate": 0.0010550187992587833,
      "loss": 0.0,
      "num_input_tokens_seen": 21858936,
      "step": 38490
    },
    {
      "epoch": 675.3539823008849,
      "grad_norm": 4.288154187293003e-08,
      "learning_rate": 0.0010480562250145653,
      "loss": 0.0,
      "num_input_tokens_seen": 21862184,
      "step": 38495
    },
    {
      "epoch": 675.4424778761062,
      "grad_norm": 3.8811350577816484e-08,
      "learning_rate": 0.0010411166210319567,
      "loss": 0.0,
      "num_input_tokens_seen": 21864792,
      "step": 38500
    },
    {
      "epoch": 675.5309734513274,
      "grad_norm": 5.484825038593044e-08,
      "learning_rate": 0.0010341999883810848,
      "loss": 0.0,
      "num_input_tokens_seen": 21867800,
      "step": 38505
    },
    {
      "epoch": 675.6194690265487,
      "grad_norm": 1.035331642640358e-08,
      "learning_rate": 0.0010273063281285965,
      "loss": 0.0,
      "num_input_tokens_seen": 21870968,
      "step": 38510
    },
    {
      "epoch": 675.70796460177,
      "grad_norm": 2.7131127566804025e-08,
      "learning_rate": 0.0010204356413375747,
      "loss": 0.0,
      "num_input_tokens_seen": 21873736,
      "step": 38515
    },
    {
      "epoch": 675.7964601769911,
      "grad_norm": 5.689189563895525e-08,
      "learning_rate": 0.001013587929067572,
      "loss": 0.0,
      "num_input_tokens_seen": 21876808,
      "step": 38520
    },
    {
      "epoch": 675.8849557522124,
      "grad_norm": 4.6168974421334497e-08,
      "learning_rate": 0.00100676319237461,
      "loss": 0.0,
      "num_input_tokens_seen": 21879944,
      "step": 38525
    },
    {
      "epoch": 675.9734513274336,
      "grad_norm": 3.850312779718479e-08,
      "learning_rate": 0.0009999614323110972,
      "loss": 0.0,
      "num_input_tokens_seen": 21882920,
      "step": 38530
    },
    {
      "epoch": 676.0530973451328,
      "grad_norm": 2.1913125536343614e-08,
      "learning_rate": 0.000993182649926011,
      "loss": 0.0,
      "num_input_tokens_seen": 21885056,
      "step": 38535
    },
    {
      "epoch": 676.141592920354,
      "grad_norm": 3.7638457683897286e-08,
      "learning_rate": 0.000986426846264682,
      "loss": 0.0,
      "num_input_tokens_seen": 21887664,
      "step": 38540
    },
    {
      "epoch": 676.2300884955753,
      "grad_norm": 5.807854108752508e-08,
      "learning_rate": 0.00097969402236896,
      "loss": 0.0,
      "num_input_tokens_seen": 21890864,
      "step": 38545
    },
    {
      "epoch": 676.3185840707964,
      "grad_norm": 4.466122049961996e-08,
      "learning_rate": 0.0009729841792771143,
      "loss": 0.0,
      "num_input_tokens_seen": 21893232,
      "step": 38550
    },
    {
      "epoch": 676.4070796460177,
      "grad_norm": 3.050596930620486e-08,
      "learning_rate": 0.0009662973180239176,
      "loss": 0.0,
      "num_input_tokens_seen": 21896128,
      "step": 38555
    },
    {
      "epoch": 676.4955752212389,
      "grad_norm": 3.402989889877972e-08,
      "learning_rate": 0.0009596334396405448,
      "loss": 0.0,
      "num_input_tokens_seen": 21898768,
      "step": 38560
    },
    {
      "epoch": 676.5840707964602,
      "grad_norm": 8.500694548274623e-08,
      "learning_rate": 0.0009529925451546406,
      "loss": 0.0,
      "num_input_tokens_seen": 21901824,
      "step": 38565
    },
    {
      "epoch": 676.6725663716815,
      "grad_norm": 3.272971937917646e-08,
      "learning_rate": 0.0009463746355903357,
      "loss": 0.0,
      "num_input_tokens_seen": 21904816,
      "step": 38570
    },
    {
      "epoch": 676.7610619469026,
      "grad_norm": 5.4881912348037076e-08,
      "learning_rate": 0.0009397797119681971,
      "loss": 0.0,
      "num_input_tokens_seen": 21908208,
      "step": 38575
    },
    {
      "epoch": 676.8495575221239,
      "grad_norm": 7.060258866431468e-08,
      "learning_rate": 0.0009332077753052281,
      "loss": 0.0,
      "num_input_tokens_seen": 21911136,
      "step": 38580
    },
    {
      "epoch": 676.9380530973451,
      "grad_norm": 4.4273139820916185e-08,
      "learning_rate": 0.0009266588266149011,
      "loss": 0.0,
      "num_input_tokens_seen": 21913952,
      "step": 38585
    },
    {
      "epoch": 677.0176991150443,
      "grad_norm": 3.794923131295036e-08,
      "learning_rate": 0.0009201328669071584,
      "loss": 0.0,
      "num_input_tokens_seen": 21916616,
      "step": 38590
    },
    {
      "epoch": 677.1061946902655,
      "grad_norm": 2.179042901673256e-08,
      "learning_rate": 0.0009136298971883949,
      "loss": 0.0,
      "num_input_tokens_seen": 21919464,
      "step": 38595
    },
    {
      "epoch": 677.1946902654868,
      "grad_norm": 6.344411929148919e-08,
      "learning_rate": 0.0009071499184614251,
      "loss": 0.0,
      "num_input_tokens_seen": 21922424,
      "step": 38600
    },
    {
      "epoch": 677.1946902654868,
      "eval_loss": 0.8041082620620728,
      "eval_runtime": 0.936,
      "eval_samples_per_second": 26.709,
      "eval_steps_per_second": 13.889,
      "num_input_tokens_seen": 21922424,
      "step": 38600
    },
    {
      "epoch": 677.2831858407079,
      "grad_norm": 2.624825157226951e-08,
      "learning_rate": 0.0009006929317255663,
      "loss": 0.0,
      "num_input_tokens_seen": 21925352,
      "step": 38605
    },
    {
      "epoch": 677.3716814159292,
      "grad_norm": 7.456537076677705e-08,
      "learning_rate": 0.0008942589379765387,
      "loss": 0.0,
      "num_input_tokens_seen": 21928232,
      "step": 38610
    },
    {
      "epoch": 677.4601769911504,
      "grad_norm": 1.7836745414001598e-08,
      "learning_rate": 0.0008878479382065817,
      "loss": 0.0,
      "num_input_tokens_seen": 21930936,
      "step": 38615
    },
    {
      "epoch": 677.5486725663717,
      "grad_norm": 3.617888921780832e-08,
      "learning_rate": 0.0008814599334043215,
      "loss": 0.0,
      "num_input_tokens_seen": 21933832,
      "step": 38620
    },
    {
      "epoch": 677.637168141593,
      "grad_norm": 3.4327193532135425e-08,
      "learning_rate": 0.0008750949245548866,
      "loss": 0.0,
      "num_input_tokens_seen": 21936584,
      "step": 38625
    },
    {
      "epoch": 677.7256637168142,
      "grad_norm": 9.644985965451269e-08,
      "learning_rate": 0.0008687529126398252,
      "loss": 0.0,
      "num_input_tokens_seen": 21938936,
      "step": 38630
    },
    {
      "epoch": 677.8141592920354,
      "grad_norm": 4.290393107453383e-08,
      "learning_rate": 0.0008624338986371715,
      "loss": 0.0,
      "num_input_tokens_seen": 21941928,
      "step": 38635
    },
    {
      "epoch": 677.9026548672566,
      "grad_norm": 3.877931931128842e-08,
      "learning_rate": 0.0008561378835213962,
      "loss": 0.0,
      "num_input_tokens_seen": 21944952,
      "step": 38640
    },
    {
      "epoch": 677.9911504424779,
      "grad_norm": 2.95348510093163e-08,
      "learning_rate": 0.0008498648682634058,
      "loss": 0.0,
      "num_input_tokens_seen": 21947768,
      "step": 38645
    },
    {
      "epoch": 678.070796460177,
      "grad_norm": 4.983577284178864e-08,
      "learning_rate": 0.0008436148538306099,
      "loss": 0.0,
      "num_input_tokens_seen": 21950296,
      "step": 38650
    },
    {
      "epoch": 678.1592920353983,
      "grad_norm": 5.8301171890207115e-08,
      "learning_rate": 0.0008373878411868041,
      "loss": 0.0,
      "num_input_tokens_seen": 21953080,
      "step": 38655
    },
    {
      "epoch": 678.2477876106195,
      "grad_norm": 2.352419592455135e-08,
      "learning_rate": 0.000831183831292287,
      "loss": 0.0,
      "num_input_tokens_seen": 21955736,
      "step": 38660
    },
    {
      "epoch": 678.3362831858407,
      "grad_norm": 3.525707015228363e-08,
      "learning_rate": 0.0008250028251037933,
      "loss": 0.0,
      "num_input_tokens_seen": 21958952,
      "step": 38665
    },
    {
      "epoch": 678.4247787610619,
      "grad_norm": 4.336259706860801e-08,
      "learning_rate": 0.0008188448235745271,
      "loss": 0.0,
      "num_input_tokens_seen": 21962056,
      "step": 38670
    },
    {
      "epoch": 678.5132743362832,
      "grad_norm": 4.345683635165187e-08,
      "learning_rate": 0.0008127098276541122,
      "loss": 0.0,
      "num_input_tokens_seen": 21964808,
      "step": 38675
    },
    {
      "epoch": 678.6017699115044,
      "grad_norm": 1.5270691378077572e-08,
      "learning_rate": 0.0008065978382886418,
      "loss": 0.0,
      "num_input_tokens_seen": 21967928,
      "step": 38680
    },
    {
      "epoch": 678.6902654867257,
      "grad_norm": 5.0121695238658504e-08,
      "learning_rate": 0.0008005088564206785,
      "loss": 0.0,
      "num_input_tokens_seen": 21970376,
      "step": 38685
    },
    {
      "epoch": 678.7787610619469,
      "grad_norm": 6.852297929071938e-08,
      "learning_rate": 0.0007944428829891881,
      "loss": 0.0,
      "num_input_tokens_seen": 21973208,
      "step": 38690
    },
    {
      "epoch": 678.8672566371681,
      "grad_norm": 3.20256212660297e-08,
      "learning_rate": 0.0007883999189296386,
      "loss": 0.0,
      "num_input_tokens_seen": 21976712,
      "step": 38695
    },
    {
      "epoch": 678.9557522123894,
      "grad_norm": 1.823904582920477e-08,
      "learning_rate": 0.0007823799651739515,
      "loss": 0.0,
      "num_input_tokens_seen": 21979784,
      "step": 38700
    },
    {
      "epoch": 679.0353982300885,
      "grad_norm": 2.4705203216512928e-08,
      "learning_rate": 0.0007763830226504509,
      "loss": 0.0,
      "num_input_tokens_seen": 21982024,
      "step": 38705
    },
    {
      "epoch": 679.1238938053098,
      "grad_norm": 3.205364507152808e-08,
      "learning_rate": 0.0007704090922839468,
      "loss": 0.0,
      "num_input_tokens_seen": 21984936,
      "step": 38710
    },
    {
      "epoch": 679.212389380531,
      "grad_norm": 8.22729901983621e-08,
      "learning_rate": 0.0007644581749957025,
      "loss": 0.0,
      "num_input_tokens_seen": 21988008,
      "step": 38715
    },
    {
      "epoch": 679.3008849557522,
      "grad_norm": 5.181451356861544e-08,
      "learning_rate": 0.000758530271703417,
      "loss": 0.0,
      "num_input_tokens_seen": 21990760,
      "step": 38720
    },
    {
      "epoch": 679.3893805309734,
      "grad_norm": 4.284996535375285e-08,
      "learning_rate": 0.0007526253833212426,
      "loss": 0.0,
      "num_input_tokens_seen": 21993464,
      "step": 38725
    },
    {
      "epoch": 679.4778761061947,
      "grad_norm": 5.3267630306663705e-08,
      "learning_rate": 0.0007467435107598008,
      "loss": 0.0,
      "num_input_tokens_seen": 21996376,
      "step": 38730
    },
    {
      "epoch": 679.566371681416,
      "grad_norm": 3.3001622767869776e-08,
      "learning_rate": 0.0007408846549261328,
      "loss": 0.0,
      "num_input_tokens_seen": 21998856,
      "step": 38735
    },
    {
      "epoch": 679.6548672566372,
      "grad_norm": 2.6317604095993374e-08,
      "learning_rate": 0.0007350488167237656,
      "loss": 0.0,
      "num_input_tokens_seen": 22001784,
      "step": 38740
    },
    {
      "epoch": 679.7433628318585,
      "grad_norm": 3.01145348657883e-08,
      "learning_rate": 0.0007292359970526629,
      "loss": 0.0,
      "num_input_tokens_seen": 22004616,
      "step": 38745
    },
    {
      "epoch": 679.8318584070796,
      "grad_norm": 2.8702325849394583e-08,
      "learning_rate": 0.0007234461968092076,
      "loss": 0.0,
      "num_input_tokens_seen": 22007400,
      "step": 38750
    },
    {
      "epoch": 679.9203539823009,
      "grad_norm": 5.4236743096680584e-08,
      "learning_rate": 0.0007176794168862854,
      "loss": 0.0,
      "num_input_tokens_seen": 22010792,
      "step": 38755
    },
    {
      "epoch": 680.0,
      "grad_norm": 5.833883420791608e-08,
      "learning_rate": 0.000711935658173185,
      "loss": 0.0,
      "num_input_tokens_seen": 22013496,
      "step": 38760
    },
    {
      "epoch": 680.0884955752213,
      "grad_norm": 3.7171886901887774e-08,
      "learning_rate": 0.0007062149215556812,
      "loss": 0.0,
      "num_input_tokens_seen": 22016200,
      "step": 38765
    },
    {
      "epoch": 680.1769911504425,
      "grad_norm": 2.7656378520646285e-08,
      "learning_rate": 0.0007005172079159849,
      "loss": 0.0,
      "num_input_tokens_seen": 22019496,
      "step": 38770
    },
    {
      "epoch": 680.2654867256637,
      "grad_norm": 7.105413857289022e-08,
      "learning_rate": 0.0006948425181327267,
      "loss": 0.0,
      "num_input_tokens_seen": 22022168,
      "step": 38775
    },
    {
      "epoch": 680.3539823008849,
      "grad_norm": 2.7950550318678324e-08,
      "learning_rate": 0.000689190853081073,
      "loss": 0.0,
      "num_input_tokens_seen": 22025416,
      "step": 38780
    },
    {
      "epoch": 680.4424778761062,
      "grad_norm": 3.921374869264582e-08,
      "learning_rate": 0.000683562213632527,
      "loss": 0.0,
      "num_input_tokens_seen": 22028696,
      "step": 38785
    },
    {
      "epoch": 680.5309734513274,
      "grad_norm": 5.5112046482008736e-08,
      "learning_rate": 0.0006779566006551108,
      "loss": 0.0,
      "num_input_tokens_seen": 22031016,
      "step": 38790
    },
    {
      "epoch": 680.6194690265487,
      "grad_norm": 4.7365080746430976e-08,
      "learning_rate": 0.0006723740150132995,
      "loss": 0.0,
      "num_input_tokens_seen": 22033752,
      "step": 38795
    },
    {
      "epoch": 680.70796460177,
      "grad_norm": 4.727286295747035e-08,
      "learning_rate": 0.0006668144575679713,
      "loss": 0.0,
      "num_input_tokens_seen": 22036600,
      "step": 38800
    },
    {
      "epoch": 680.70796460177,
      "eval_loss": 0.8143786191940308,
      "eval_runtime": 0.9416,
      "eval_samples_per_second": 26.552,
      "eval_steps_per_second": 13.807,
      "num_input_tokens_seen": 22036600,
      "step": 38800
    },
    {
      "epoch": 680.7964601769911,
      "grad_norm": 4.839253264776744e-08,
      "learning_rate": 0.0006612779291765069,
      "loss": 0.0,
      "num_input_tokens_seen": 22039480,
      "step": 38805
    },
    {
      "epoch": 680.8849557522124,
      "grad_norm": 3.82449947267105e-08,
      "learning_rate": 0.0006557644306926736,
      "loss": 0.0,
      "num_input_tokens_seen": 22042696,
      "step": 38810
    },
    {
      "epoch": 680.9734513274336,
      "grad_norm": 4.112534313094329e-08,
      "learning_rate": 0.0006502739629667575,
      "loss": 0.0,
      "num_input_tokens_seen": 22045272,
      "step": 38815
    },
    {
      "epoch": 681.0530973451328,
      "grad_norm": 4.6685702415061314e-08,
      "learning_rate": 0.0006448065268454317,
      "loss": 0.0,
      "num_input_tokens_seen": 22047824,
      "step": 38820
    },
    {
      "epoch": 681.141592920354,
      "grad_norm": 2.5546706083900972e-08,
      "learning_rate": 0.0006393621231718549,
      "loss": 0.0,
      "num_input_tokens_seen": 22051280,
      "step": 38825
    },
    {
      "epoch": 681.2300884955753,
      "grad_norm": 3.749482146986338e-08,
      "learning_rate": 0.0006339407527856389,
      "loss": 0.0,
      "num_input_tokens_seen": 22054256,
      "step": 38830
    },
    {
      "epoch": 681.3185840707964,
      "grad_norm": 3.546698934542292e-08,
      "learning_rate": 0.0006285424165227982,
      "loss": 0.0,
      "num_input_tokens_seen": 22056848,
      "step": 38835
    },
    {
      "epoch": 681.4070796460177,
      "grad_norm": 3.837685724761286e-08,
      "learning_rate": 0.0006231671152158169,
      "loss": 0.0,
      "num_input_tokens_seen": 22059824,
      "step": 38840
    },
    {
      "epoch": 681.4955752212389,
      "grad_norm": 3.1770746033998876e-08,
      "learning_rate": 0.0006178148496936819,
      "loss": 0.0,
      "num_input_tokens_seen": 22062544,
      "step": 38845
    },
    {
      "epoch": 681.5840707964602,
      "grad_norm": 5.629536659057521e-08,
      "learning_rate": 0.000612485620781733,
      "loss": 0.0,
      "num_input_tokens_seen": 22065504,
      "step": 38850
    },
    {
      "epoch": 681.6725663716815,
      "grad_norm": 3.8152784043177235e-08,
      "learning_rate": 0.0006071794293018296,
      "loss": 0.0,
      "num_input_tokens_seen": 22068304,
      "step": 38855
    },
    {
      "epoch": 681.7610619469026,
      "grad_norm": 2.201691984282661e-08,
      "learning_rate": 0.0006018962760722501,
      "loss": 0.0,
      "num_input_tokens_seen": 22070832,
      "step": 38860
    },
    {
      "epoch": 681.8495575221239,
      "grad_norm": 2.9852547100972515e-08,
      "learning_rate": 0.0005966361619077098,
      "loss": 0.0,
      "num_input_tokens_seen": 22073968,
      "step": 38865
    },
    {
      "epoch": 681.9380530973451,
      "grad_norm": 3.642498924705251e-08,
      "learning_rate": 0.000591399087619393,
      "loss": 0.0,
      "num_input_tokens_seen": 22076576,
      "step": 38870
    },
    {
      "epoch": 682.0176991150443,
      "grad_norm": 8.753843872000289e-08,
      "learning_rate": 0.0005861850540149371,
      "loss": 0.0,
      "num_input_tokens_seen": 22078808,
      "step": 38875
    },
    {
      "epoch": 682.1061946902655,
      "grad_norm": 2.790616626668907e-08,
      "learning_rate": 0.0005809940618983822,
      "loss": 0.0,
      "num_input_tokens_seen": 22082264,
      "step": 38880
    },
    {
      "epoch": 682.1946902654868,
      "grad_norm": 3.7128252472484746e-08,
      "learning_rate": 0.0005758261120702712,
      "loss": 0.0,
      "num_input_tokens_seen": 22085080,
      "step": 38885
    },
    {
      "epoch": 682.2831858407079,
      "grad_norm": 5.128852720304167e-08,
      "learning_rate": 0.0005706812053275501,
      "loss": 0.0,
      "num_input_tokens_seen": 22088376,
      "step": 38890
    },
    {
      "epoch": 682.3716814159292,
      "grad_norm": 4.9408686919605316e-08,
      "learning_rate": 0.0005655593424636173,
      "loss": 0.0,
      "num_input_tokens_seen": 22091160,
      "step": 38895
    },
    {
      "epoch": 682.4601769911504,
      "grad_norm": 3.523823721707231e-08,
      "learning_rate": 0.0005604605242683746,
      "loss": 0.0,
      "num_input_tokens_seen": 22094216,
      "step": 38900
    },
    {
      "epoch": 682.5486725663717,
      "grad_norm": 2.7729063489800865e-08,
      "learning_rate": 0.0005553847515280596,
      "loss": 0.0,
      "num_input_tokens_seen": 22097736,
      "step": 38905
    },
    {
      "epoch": 682.637168141593,
      "grad_norm": 5.659629920273801e-08,
      "learning_rate": 0.0005503320250254795,
      "loss": 0.0,
      "num_input_tokens_seen": 22100328,
      "step": 38910
    },
    {
      "epoch": 682.7256637168142,
      "grad_norm": 6.466908786251224e-08,
      "learning_rate": 0.0005453023455397943,
      "loss": 0.0,
      "num_input_tokens_seen": 22103096,
      "step": 38915
    },
    {
      "epoch": 682.8141592920354,
      "grad_norm": 1.208355815407458e-08,
      "learning_rate": 0.0005402957138466502,
      "loss": 0.0,
      "num_input_tokens_seen": 22105624,
      "step": 38920
    },
    {
      "epoch": 682.9026548672566,
      "grad_norm": 4.38168044070153e-08,
      "learning_rate": 0.0005353121307181463,
      "loss": 0.0,
      "num_input_tokens_seen": 22108104,
      "step": 38925
    },
    {
      "epoch": 682.9911504424779,
      "grad_norm": 3.073882481885448e-08,
      "learning_rate": 0.0005303515969227845,
      "loss": 0.0,
      "num_input_tokens_seen": 22110808,
      "step": 38930
    },
    {
      "epoch": 683.070796460177,
      "grad_norm": 3.6609819176192104e-08,
      "learning_rate": 0.0005254141132255862,
      "loss": 0.0,
      "num_input_tokens_seen": 22112896,
      "step": 38935
    },
    {
      "epoch": 683.1592920353983,
      "grad_norm": 5.318355889016857e-08,
      "learning_rate": 0.0005204996803879258,
      "loss": 0.0,
      "num_input_tokens_seen": 22116272,
      "step": 38940
    },
    {
      "epoch": 683.2477876106195,
      "grad_norm": 1.9793343852825274e-08,
      "learning_rate": 0.0005156082991676969,
      "loss": 0.0,
      "num_input_tokens_seen": 22119456,
      "step": 38945
    },
    {
      "epoch": 683.3362831858407,
      "grad_norm": 2.009167054950467e-08,
      "learning_rate": 0.0005107399703192127,
      "loss": 0.0,
      "num_input_tokens_seen": 22122544,
      "step": 38950
    },
    {
      "epoch": 683.4247787610619,
      "grad_norm": 4.4817646482897544e-08,
      "learning_rate": 0.0005058946945932063,
      "loss": 0.0,
      "num_input_tokens_seen": 22125728,
      "step": 38955
    },
    {
      "epoch": 683.5132743362832,
      "grad_norm": 3.899962308651084e-08,
      "learning_rate": 0.0005010724727369131,
      "loss": 0.0,
      "num_input_tokens_seen": 22128864,
      "step": 38960
    },
    {
      "epoch": 683.6017699115044,
      "grad_norm": 7.598786311291406e-08,
      "learning_rate": 0.000496273305493955,
      "loss": 0.0,
      "num_input_tokens_seen": 22131408,
      "step": 38965
    },
    {
      "epoch": 683.6902654867257,
      "grad_norm": 3.3749458339116245e-08,
      "learning_rate": 0.0004914971936044399,
      "loss": 0.0,
      "num_input_tokens_seen": 22133920,
      "step": 38970
    },
    {
      "epoch": 683.7787610619469,
      "grad_norm": 3.862858477532427e-08,
      "learning_rate": 0.00048674413780491196,
      "loss": 0.0,
      "num_input_tokens_seen": 22136384,
      "step": 38975
    },
    {
      "epoch": 683.8672566371681,
      "grad_norm": 3.1215524387562255e-08,
      "learning_rate": 0.0004820141388283183,
      "loss": 0.0,
      "num_input_tokens_seen": 22139024,
      "step": 38980
    },
    {
      "epoch": 683.9557522123894,
      "grad_norm": 4.2310009717994035e-08,
      "learning_rate": 0.00047730719740410874,
      "loss": 0.0,
      "num_input_tokens_seen": 22142064,
      "step": 38985
    },
    {
      "epoch": 684.0353982300885,
      "grad_norm": 3.1045278348074135e-08,
      "learning_rate": 0.00047262331425816927,
      "loss": 0.0,
      "num_input_tokens_seen": 22144496,
      "step": 38990
    },
    {
      "epoch": 684.1238938053098,
      "grad_norm": 6.161729970699525e-08,
      "learning_rate": 0.00046796249011277213,
      "loss": 0.0,
      "num_input_tokens_seen": 22147328,
      "step": 38995
    },
    {
      "epoch": 684.212389380531,
      "grad_norm": 3.166390172282263e-08,
      "learning_rate": 0.00046332472568669236,
      "loss": 0.0,
      "num_input_tokens_seen": 22150992,
      "step": 39000
    },
    {
      "epoch": 684.212389380531,
      "eval_loss": 0.7954248189926147,
      "eval_runtime": 0.9462,
      "eval_samples_per_second": 26.423,
      "eval_steps_per_second": 13.74,
      "num_input_tokens_seen": 22150992,
      "step": 39000
    },
    {
      "epoch": 684.3008849557522,
      "grad_norm": 2.5294500716199764e-08,
      "learning_rate": 0.0004587100216951578,
      "loss": 0.0,
      "num_input_tokens_seen": 22153408,
      "step": 39005
    },
    {
      "epoch": 684.3893805309734,
      "grad_norm": 2.2927425291641157e-08,
      "learning_rate": 0.00045411837884978265,
      "loss": 0.0,
      "num_input_tokens_seen": 22156784,
      "step": 39010
    },
    {
      "epoch": 684.4778761061947,
      "grad_norm": 4.89205653764202e-08,
      "learning_rate": 0.00044954979785865045,
      "loss": 0.0,
      "num_input_tokens_seen": 22159968,
      "step": 39015
    },
    {
      "epoch": 684.566371681416,
      "grad_norm": 1.7083859660260714e-08,
      "learning_rate": 0.00044500427942631426,
      "loss": 0.0,
      "num_input_tokens_seen": 22162464,
      "step": 39020
    },
    {
      "epoch": 684.6548672566372,
      "grad_norm": 7.462561057991479e-08,
      "learning_rate": 0.0004404818242537467,
      "loss": 0.0,
      "num_input_tokens_seen": 22164880,
      "step": 39025
    },
    {
      "epoch": 684.7433628318585,
      "grad_norm": 2.689155920165831e-08,
      "learning_rate": 0.00043598243303837324,
      "loss": 0.0,
      "num_input_tokens_seen": 22167280,
      "step": 39030
    },
    {
      "epoch": 684.8318584070796,
      "grad_norm": 1.694288975784275e-08,
      "learning_rate": 0.00043150610647403885,
      "loss": 0.0,
      "num_input_tokens_seen": 22170144,
      "step": 39035
    },
    {
      "epoch": 684.9203539823009,
      "grad_norm": 4.301513811810764e-08,
      "learning_rate": 0.00042705284525104134,
      "loss": 0.0,
      "num_input_tokens_seen": 22172912,
      "step": 39040
    },
    {
      "epoch": 685.0,
      "grad_norm": 1.0134002081940707e-07,
      "learning_rate": 0.0004226226500561647,
      "loss": 0.0,
      "num_input_tokens_seen": 22175264,
      "step": 39045
    },
    {
      "epoch": 685.0884955752213,
      "grad_norm": 2.2385096443144903e-08,
      "learning_rate": 0.0004182155215725791,
      "loss": 0.0,
      "num_input_tokens_seen": 22178128,
      "step": 39050
    },
    {
      "epoch": 685.1769911504425,
      "grad_norm": 1.546244909889083e-08,
      "learning_rate": 0.00041383146047992424,
      "loss": 0.0,
      "num_input_tokens_seen": 22180960,
      "step": 39055
    },
    {
      "epoch": 685.2654867256637,
      "grad_norm": 3.231144063420288e-08,
      "learning_rate": 0.00040947046745427597,
      "loss": 0.0,
      "num_input_tokens_seen": 22183760,
      "step": 39060
    },
    {
      "epoch": 685.3539823008849,
      "grad_norm": 6.349777237346643e-08,
      "learning_rate": 0.00040513254316814625,
      "loss": 0.0,
      "num_input_tokens_seen": 22186400,
      "step": 39065
    },
    {
      "epoch": 685.4424778761062,
      "grad_norm": 7.876209195956108e-08,
      "learning_rate": 0.0004008176882905168,
      "loss": 0.0,
      "num_input_tokens_seen": 22189040,
      "step": 39070
    },
    {
      "epoch": 685.5309734513274,
      "grad_norm": 7.956150938071005e-08,
      "learning_rate": 0.00039652590348677184,
      "loss": 0.0,
      "num_input_tokens_seen": 22191728,
      "step": 39075
    },
    {
      "epoch": 685.6194690265487,
      "grad_norm": 2.8812442209869005e-08,
      "learning_rate": 0.00039225718941878206,
      "loss": 0.0,
      "num_input_tokens_seen": 22194880,
      "step": 39080
    },
    {
      "epoch": 685.70796460177,
      "grad_norm": 3.713010698902508e-08,
      "learning_rate": 0.00038801154674480417,
      "loss": 0.0,
      "num_input_tokens_seen": 22198032,
      "step": 39085
    },
    {
      "epoch": 685.7964601769911,
      "grad_norm": 1.9561419151159498e-08,
      "learning_rate": 0.00038378897611959784,
      "loss": 0.0,
      "num_input_tokens_seen": 22201184,
      "step": 39090
    },
    {
      "epoch": 685.8849557522124,
      "grad_norm": 1.0735693933838775e-07,
      "learning_rate": 0.00037958947819430875,
      "loss": 0.0,
      "num_input_tokens_seen": 22204144,
      "step": 39095
    },
    {
      "epoch": 685.9734513274336,
      "grad_norm": 3.138675808145308e-08,
      "learning_rate": 0.0003754130536165856,
      "loss": 0.0,
      "num_input_tokens_seen": 22206912,
      "step": 39100
    },
    {
      "epoch": 686.0530973451328,
      "grad_norm": 3.4077654476050157e-08,
      "learning_rate": 0.0003712597030304632,
      "loss": 0.0,
      "num_input_tokens_seen": 22209152,
      "step": 39105
    },
    {
      "epoch": 686.141592920354,
      "grad_norm": 6.462130386353238e-08,
      "learning_rate": 0.00036712942707646247,
      "loss": 0.0,
      "num_input_tokens_seen": 22211904,
      "step": 39110
    },
    {
      "epoch": 686.2300884955753,
      "grad_norm": 2.8266196494541873e-08,
      "learning_rate": 0.00036302222639149063,
      "loss": 0.0,
      "num_input_tokens_seen": 22214864,
      "step": 39115
    },
    {
      "epoch": 686.3185840707964,
      "grad_norm": 3.488814215302227e-08,
      "learning_rate": 0.000358938101608941,
      "loss": 0.0,
      "num_input_tokens_seen": 22217424,
      "step": 39120
    },
    {
      "epoch": 686.4070796460177,
      "grad_norm": 6.724427237259079e-08,
      "learning_rate": 0.0003548770533586598,
      "loss": 0.0,
      "num_input_tokens_seen": 22219936,
      "step": 39125
    },
    {
      "epoch": 686.4955752212389,
      "grad_norm": 3.0814636176046406e-08,
      "learning_rate": 0.0003508390822668961,
      "loss": 0.0,
      "num_input_tokens_seen": 22222432,
      "step": 39130
    },
    {
      "epoch": 686.5840707964602,
      "grad_norm": 3.6699528749295496e-08,
      "learning_rate": 0.00034682418895633503,
      "loss": 0.0,
      "num_input_tokens_seen": 22225456,
      "step": 39135
    },
    {
      "epoch": 686.6725663716815,
      "grad_norm": 3.6772746625501895e-08,
      "learning_rate": 0.0003428323740461647,
      "loss": 0.0,
      "num_input_tokens_seen": 22228304,
      "step": 39140
    },
    {
      "epoch": 686.7610619469026,
      "grad_norm": 5.0563148334958896e-08,
      "learning_rate": 0.00033886363815194276,
      "loss": 0.0,
      "num_input_tokens_seen": 22231824,
      "step": 39145
    },
    {
      "epoch": 686.8495575221239,
      "grad_norm": 2.0795443589349816e-08,
      "learning_rate": 0.0003349179818857129,
      "loss": 0.0,
      "num_input_tokens_seen": 22234688,
      "step": 39150
    },
    {
      "epoch": 686.9380530973451,
      "grad_norm": 3.447995666761017e-08,
      "learning_rate": 0.0003309954058559383,
      "loss": 0.0,
      "num_input_tokens_seen": 22237632,
      "step": 39155
    },
    {
      "epoch": 687.0176991150443,
      "grad_norm": 3.10585903662286e-08,
      "learning_rate": 0.0003270959106675186,
      "loss": 0.0,
      "num_input_tokens_seen": 22240208,
      "step": 39160
    },
    {
      "epoch": 687.1061946902655,
      "grad_norm": 4.719969126654178e-08,
      "learning_rate": 0.0003232194969218227,
      "loss": 0.0,
      "num_input_tokens_seen": 22243472,
      "step": 39165
    },
    {
      "epoch": 687.1946902654868,
      "grad_norm": 4.73492569597056e-08,
      "learning_rate": 0.00031936616521663905,
      "loss": 0.0,
      "num_input_tokens_seen": 22246384,
      "step": 39170
    },
    {
      "epoch": 687.2831858407079,
      "grad_norm": 7.638634968998304e-08,
      "learning_rate": 0.00031553591614619236,
      "loss": 0.0,
      "num_input_tokens_seen": 22249520,
      "step": 39175
    },
    {
      "epoch": 687.3716814159292,
      "grad_norm": 5.606282016401565e-08,
      "learning_rate": 0.00031172875030117676,
      "loss": 0.0,
      "num_input_tokens_seen": 22252336,
      "step": 39180
    },
    {
      "epoch": 687.4601769911504,
      "grad_norm": 6.103462624196254e-08,
      "learning_rate": 0.0003079446682686726,
      "loss": 0.0,
      "num_input_tokens_seen": 22254912,
      "step": 39185
    },
    {
      "epoch": 687.5486725663717,
      "grad_norm": 9.037844961312658e-08,
      "learning_rate": 0.0003041836706322465,
      "loss": 0.0,
      "num_input_tokens_seen": 22258080,
      "step": 39190
    },
    {
      "epoch": 687.637168141593,
      "grad_norm": 3.013387228634201e-08,
      "learning_rate": 0.0003004457579719011,
      "loss": 0.0,
      "num_input_tokens_seen": 22261072,
      "step": 39195
    },
    {
      "epoch": 687.7256637168142,
      "grad_norm": 3.391904002114643e-08,
      "learning_rate": 0.00029673093086405867,
      "loss": 0.0,
      "num_input_tokens_seen": 22263616,
      "step": 39200
    },
    {
      "epoch": 687.7256637168142,
      "eval_loss": 0.7884461879730225,
      "eval_runtime": 0.9674,
      "eval_samples_per_second": 25.844,
      "eval_steps_per_second": 13.439,
      "num_input_tokens_seen": 22263616,
      "step": 39200
    },
    {
      "epoch": 687.8141592920354,
      "grad_norm": 4.0504449572154044e-08,
      "learning_rate": 0.00029303918988159426,
      "loss": 0.0,
      "num_input_tokens_seen": 22266288,
      "step": 39205
    },
    {
      "epoch": 687.9026548672566,
      "grad_norm": 5.4145072425626495e-08,
      "learning_rate": 0.0002893705355938192,
      "loss": 0.0,
      "num_input_tokens_seen": 22269216,
      "step": 39210
    },
    {
      "epoch": 687.9911504424779,
      "grad_norm": 2.4260192077463216e-08,
      "learning_rate": 0.0002857249685664975,
      "loss": 0.0,
      "num_input_tokens_seen": 22271984,
      "step": 39215
    },
    {
      "epoch": 688.070796460177,
      "grad_norm": 1.9482490287714427e-08,
      "learning_rate": 0.0002821024893618129,
      "loss": 0.0,
      "num_input_tokens_seen": 22274032,
      "step": 39220
    },
    {
      "epoch": 688.1592920353983,
      "grad_norm": 4.3404824623394234e-08,
      "learning_rate": 0.0002785030985383852,
      "loss": 0.0,
      "num_input_tokens_seen": 22276752,
      "step": 39225
    },
    {
      "epoch": 688.2477876106195,
      "grad_norm": 4.691457178296332e-08,
      "learning_rate": 0.00027492679665130356,
      "loss": 0.0,
      "num_input_tokens_seen": 22279616,
      "step": 39230
    },
    {
      "epoch": 688.3362831858407,
      "grad_norm": 4.8156756804473844e-08,
      "learning_rate": 0.000271373584252077,
      "loss": 0.0,
      "num_input_tokens_seen": 22282272,
      "step": 39235
    },
    {
      "epoch": 688.4247787610619,
      "grad_norm": 3.7486842074940796e-08,
      "learning_rate": 0.00026784346188865046,
      "loss": 0.0,
      "num_input_tokens_seen": 22285376,
      "step": 39240
    },
    {
      "epoch": 688.5132743362832,
      "grad_norm": 2.3325609888047438e-08,
      "learning_rate": 0.0002643364301054218,
      "loss": 0.0,
      "num_input_tokens_seen": 22288400,
      "step": 39245
    },
    {
      "epoch": 688.6017699115044,
      "grad_norm": 3.916523994007548e-08,
      "learning_rate": 0.0002608524894431918,
      "loss": 0.0,
      "num_input_tokens_seen": 22291408,
      "step": 39250
    },
    {
      "epoch": 688.6902654867257,
      "grad_norm": 2.3816271621512897e-08,
      "learning_rate": 0.000257391640439264,
      "loss": 0.0,
      "num_input_tokens_seen": 22294608,
      "step": 39255
    },
    {
      "epoch": 688.7787610619469,
      "grad_norm": 5.47835696806942e-08,
      "learning_rate": 0.00025395388362732806,
      "loss": 0.0,
      "num_input_tokens_seen": 22297488,
      "step": 39260
    },
    {
      "epoch": 688.8672566371681,
      "grad_norm": 5.003320779906062e-08,
      "learning_rate": 0.00025053921953751,
      "loss": 0.0,
      "num_input_tokens_seen": 22300544,
      "step": 39265
    },
    {
      "epoch": 688.9557522123894,
      "grad_norm": 4.44091661222501e-08,
      "learning_rate": 0.00024714764869643855,
      "loss": 0.0,
      "num_input_tokens_seen": 22303408,
      "step": 39270
    },
    {
      "epoch": 689.0353982300885,
      "grad_norm": 5.192984886548402e-08,
      "learning_rate": 0.0002437791716270954,
      "loss": 0.0,
      "num_input_tokens_seen": 22305656,
      "step": 39275
    },
    {
      "epoch": 689.1238938053098,
      "grad_norm": 2.8609550284386387e-08,
      "learning_rate": 0.00024043378884896493,
      "loss": 0.0,
      "num_input_tokens_seen": 22308152,
      "step": 39280
    },
    {
      "epoch": 689.212389380531,
      "grad_norm": 2.953695421581415e-08,
      "learning_rate": 0.00023711150087793453,
      "loss": 0.0,
      "num_input_tokens_seen": 22311112,
      "step": 39285
    },
    {
      "epoch": 689.3008849557522,
      "grad_norm": 2.3997474229986437e-08,
      "learning_rate": 0.000233812308226361,
      "loss": 0.0,
      "num_input_tokens_seen": 22313784,
      "step": 39290
    },
    {
      "epoch": 689.3893805309734,
      "grad_norm": 7.912215238548015e-08,
      "learning_rate": 0.00023053621140300406,
      "loss": 0.0,
      "num_input_tokens_seen": 22316552,
      "step": 39295
    },
    {
      "epoch": 689.4778761061947,
      "grad_norm": 6.392016160816638e-08,
      "learning_rate": 0.00022728321091307623,
      "loss": 0.0,
      "num_input_tokens_seen": 22319544,
      "step": 39300
    },
    {
      "epoch": 689.566371681416,
      "grad_norm": 2.7768567889552287e-08,
      "learning_rate": 0.0002240533072582429,
      "loss": 0.0,
      "num_input_tokens_seen": 22322152,
      "step": 39305
    },
    {
      "epoch": 689.6548672566372,
      "grad_norm": 1.0118569804262734e-07,
      "learning_rate": 0.00022084650093658897,
      "loss": 0.0,
      "num_input_tokens_seen": 22325224,
      "step": 39310
    },
    {
      "epoch": 689.7433628318585,
      "grad_norm": 3.061665054815421e-08,
      "learning_rate": 0.0002176627924426522,
      "loss": 0.0,
      "num_input_tokens_seen": 22328312,
      "step": 39315
    },
    {
      "epoch": 689.8318584070796,
      "grad_norm": 3.447567564762721e-08,
      "learning_rate": 0.0002145021822673898,
      "loss": 0.0,
      "num_input_tokens_seen": 22330952,
      "step": 39320
    },
    {
      "epoch": 689.9203539823009,
      "grad_norm": 5.8727916751877274e-08,
      "learning_rate": 0.00021136467089822862,
      "loss": 0.0,
      "num_input_tokens_seen": 22334248,
      "step": 39325
    },
    {
      "epoch": 690.0,
      "grad_norm": 4.292676791806116e-08,
      "learning_rate": 0.00020825025881898162,
      "loss": 0.0,
      "num_input_tokens_seen": 22336792,
      "step": 39330
    },
    {
      "epoch": 690.0884955752213,
      "grad_norm": 2.8850090316723254e-08,
      "learning_rate": 0.0002051589465099479,
      "loss": 0.0,
      "num_input_tokens_seen": 22339912,
      "step": 39335
    },
    {
      "epoch": 690.1769911504425,
      "grad_norm": 4.51353443509106e-08,
      "learning_rate": 0.0002020907344478462,
      "loss": 0.0,
      "num_input_tokens_seen": 22342344,
      "step": 39340
    },
    {
      "epoch": 690.2654867256637,
      "grad_norm": 4.7052793661350734e-08,
      "learning_rate": 0.0001990456231058313,
      "loss": 0.0,
      "num_input_tokens_seen": 22345512,
      "step": 39345
    },
    {
      "epoch": 690.3539823008849,
      "grad_norm": 2.159246115240876e-08,
      "learning_rate": 0.00019602361295349423,
      "loss": 0.0,
      "num_input_tokens_seen": 22348264,
      "step": 39350
    },
    {
      "epoch": 690.4424778761062,
      "grad_norm": 6.89950780952131e-08,
      "learning_rate": 0.0001930247044568789,
      "loss": 0.0,
      "num_input_tokens_seen": 22351208,
      "step": 39355
    },
    {
      "epoch": 690.5309734513274,
      "grad_norm": 5.8653483847592724e-08,
      "learning_rate": 0.00019004889807843205,
      "loss": 0.0,
      "num_input_tokens_seen": 22354824,
      "step": 39360
    },
    {
      "epoch": 690.6194690265487,
      "grad_norm": 3.791862823732117e-08,
      "learning_rate": 0.00018709619427708656,
      "loss": 0.0,
      "num_input_tokens_seen": 22357864,
      "step": 39365
    },
    {
      "epoch": 690.70796460177,
      "grad_norm": 5.385053825079922e-08,
      "learning_rate": 0.00018416659350817822,
      "loss": 0.0,
      "num_input_tokens_seen": 22360520,
      "step": 39370
    },
    {
      "epoch": 690.7964601769911,
      "grad_norm": 1.9329338130091855e-08,
      "learning_rate": 0.00018126009622346229,
      "loss": 0.0,
      "num_input_tokens_seen": 22363416,
      "step": 39375
    },
    {
      "epoch": 690.8849557522124,
      "grad_norm": 7.318239170217566e-08,
      "learning_rate": 0.00017837670287119687,
      "loss": 0.0,
      "num_input_tokens_seen": 22366376,
      "step": 39380
    },
    {
      "epoch": 690.9734513274336,
      "grad_norm": 5.104568501224094e-08,
      "learning_rate": 0.00017551641389602633,
      "loss": 0.0,
      "num_input_tokens_seen": 22368984,
      "step": 39385
    },
    {
      "epoch": 691.0530973451328,
      "grad_norm": 2.625835193725834e-08,
      "learning_rate": 0.00017267922973903115,
      "loss": 0.0,
      "num_input_tokens_seen": 22371184,
      "step": 39390
    },
    {
      "epoch": 691.141592920354,
      "grad_norm": 8.548322938395359e-08,
      "learning_rate": 0.00016986515083774467,
      "loss": 0.0,
      "num_input_tokens_seen": 22374272,
      "step": 39395
    },
    {
      "epoch": 691.2300884955753,
      "grad_norm": 1.1660207022146096e-08,
      "learning_rate": 0.00016707417762611975,
      "loss": 0.0,
      "num_input_tokens_seen": 22377936,
      "step": 39400
    },
    {
      "epoch": 691.2300884955753,
      "eval_loss": 0.7761130928993225,
      "eval_runtime": 0.9458,
      "eval_samples_per_second": 26.434,
      "eval_steps_per_second": 13.745,
      "num_input_tokens_seen": 22377936,
      "step": 39400
    },
    {
      "epoch": 691.3185840707964,
      "grad_norm": 2.7841338123835158e-08,
      "learning_rate": 0.00016430631053459543,
      "loss": 0.0,
      "num_input_tokens_seen": 22380656,
      "step": 39405
    },
    {
      "epoch": 691.4070796460177,
      "grad_norm": 1.5615123416523602e-08,
      "learning_rate": 0.0001615615499899803,
      "loss": 0.0,
      "num_input_tokens_seen": 22383424,
      "step": 39410
    },
    {
      "epoch": 691.4955752212389,
      "grad_norm": 3.784317215149713e-08,
      "learning_rate": 0.00015883989641556905,
      "loss": 0.0,
      "num_input_tokens_seen": 22385840,
      "step": 39415
    },
    {
      "epoch": 691.5840707964602,
      "grad_norm": 1.4327982356121538e-08,
      "learning_rate": 0.00015614135023105934,
      "loss": 0.0,
      "num_input_tokens_seen": 22388720,
      "step": 39420
    },
    {
      "epoch": 691.6725663716815,
      "grad_norm": 2.5693637439871964e-08,
      "learning_rate": 0.00015346591185261827,
      "loss": 0.0,
      "num_input_tokens_seen": 22391520,
      "step": 39425
    },
    {
      "epoch": 691.7610619469026,
      "grad_norm": 4.7748908826861225e-08,
      "learning_rate": 0.00015081358169281576,
      "loss": 0.0,
      "num_input_tokens_seen": 22394432,
      "step": 39430
    },
    {
      "epoch": 691.8495575221239,
      "grad_norm": 4.116213858651463e-08,
      "learning_rate": 0.00014818436016069135,
      "loss": 0.0,
      "num_input_tokens_seen": 22397264,
      "step": 39435
    },
    {
      "epoch": 691.9380530973451,
      "grad_norm": 1.687740436295826e-08,
      "learning_rate": 0.00014557824766168735,
      "loss": 0.0,
      "num_input_tokens_seen": 22400176,
      "step": 39440
    },
    {
      "epoch": 692.0176991150443,
      "grad_norm": 1.637687851996361e-08,
      "learning_rate": 0.00014299524459769896,
      "loss": 0.0,
      "num_input_tokens_seen": 22402368,
      "step": 39445
    },
    {
      "epoch": 692.1061946902655,
      "grad_norm": 3.1421546253795896e-08,
      "learning_rate": 0.0001404353513670742,
      "loss": 0.0,
      "num_input_tokens_seen": 22404960,
      "step": 39450
    },
    {
      "epoch": 692.1946902654868,
      "grad_norm": 6.436471977622205e-08,
      "learning_rate": 0.0001378985683645806,
      "loss": 0.0,
      "num_input_tokens_seen": 22407568,
      "step": 39455
    },
    {
      "epoch": 692.2831858407079,
      "grad_norm": 4.29671160873113e-08,
      "learning_rate": 0.0001353848959813886,
      "loss": 0.0,
      "num_input_tokens_seen": 22410464,
      "step": 39460
    },
    {
      "epoch": 692.3716814159292,
      "grad_norm": 5.950649750730008e-08,
      "learning_rate": 0.00013289433460517142,
      "loss": 0.0,
      "num_input_tokens_seen": 22413088,
      "step": 39465
    },
    {
      "epoch": 692.4601769911504,
      "grad_norm": 4.947712994862741e-08,
      "learning_rate": 0.00013042688462000518,
      "loss": 0.0,
      "num_input_tokens_seen": 22416144,
      "step": 39470
    },
    {
      "epoch": 692.5486725663717,
      "grad_norm": 4.763738559177e-08,
      "learning_rate": 0.0001279825464063855,
      "loss": 0.0,
      "num_input_tokens_seen": 22418752,
      "step": 39475
    },
    {
      "epoch": 692.637168141593,
      "grad_norm": 5.572302796963413e-08,
      "learning_rate": 0.00012556132034126087,
      "loss": 0.0,
      "num_input_tokens_seen": 22422064,
      "step": 39480
    },
    {
      "epoch": 692.7256637168142,
      "grad_norm": 3.2798055826788186e-08,
      "learning_rate": 0.0001231632067980326,
      "loss": 0.0,
      "num_input_tokens_seen": 22425712,
      "step": 39485
    },
    {
      "epoch": 692.8141592920354,
      "grad_norm": 7.21892945421132e-08,
      "learning_rate": 0.00012078820614650486,
      "loss": 0.0,
      "num_input_tokens_seen": 22428624,
      "step": 39490
    },
    {
      "epoch": 692.9026548672566,
      "grad_norm": 5.744853481814971e-08,
      "learning_rate": 0.00011843631875291804,
      "loss": 0.0,
      "num_input_tokens_seen": 22431344,
      "step": 39495
    },
    {
      "epoch": 692.9911504424779,
      "grad_norm": 1.4601060804864119e-08,
      "learning_rate": 0.00011610754497999863,
      "loss": 0.0,
      "num_input_tokens_seen": 22434000,
      "step": 39500
    },
    {
      "epoch": 693.070796460177,
      "grad_norm": 6.641523242478797e-08,
      "learning_rate": 0.0001138018851868594,
      "loss": 0.0,
      "num_input_tokens_seen": 22436352,
      "step": 39505
    },
    {
      "epoch": 693.1592920353983,
      "grad_norm": 3.7329954238884966e-08,
      "learning_rate": 0.0001115193397290326,
      "loss": 0.0,
      "num_input_tokens_seen": 22439056,
      "step": 39510
    },
    {
      "epoch": 693.2477876106195,
      "grad_norm": 4.250429341823292e-08,
      "learning_rate": 0.00010925990895856996,
      "loss": 0.0,
      "num_input_tokens_seen": 22442384,
      "step": 39515
    },
    {
      "epoch": 693.3362831858407,
      "grad_norm": 4.654124197145393e-08,
      "learning_rate": 0.00010702359322385946,
      "loss": 0.0,
      "num_input_tokens_seen": 22445760,
      "step": 39520
    },
    {
      "epoch": 693.4247787610619,
      "grad_norm": 3.140776172472215e-08,
      "learning_rate": 0.00010481039286977523,
      "loss": 0.0,
      "num_input_tokens_seen": 22448672,
      "step": 39525
    },
    {
      "epoch": 693.5132743362832,
      "grad_norm": 2.029189261065767e-08,
      "learning_rate": 0.00010262030823764423,
      "loss": 0.0,
      "num_input_tokens_seen": 22451120,
      "step": 39530
    },
    {
      "epoch": 693.6017699115044,
      "grad_norm": 1.1787520293182752e-08,
      "learning_rate": 0.00010045333966517966,
      "loss": 0.0,
      "num_input_tokens_seen": 22453728,
      "step": 39535
    },
    {
      "epoch": 693.6902654867257,
      "grad_norm": 2.5953648119525496e-08,
      "learning_rate": 9.83094874865642e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22456832,
      "step": 39540
    },
    {
      "epoch": 693.7787610619469,
      "grad_norm": 2.941468224548771e-08,
      "learning_rate": 9.618875203241672e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22459456,
      "step": 39545
    },
    {
      "epoch": 693.8672566371681,
      "grad_norm": 8.18252701151323e-08,
      "learning_rate": 9.409113362977561e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22462416,
      "step": 39550
    },
    {
      "epoch": 693.9557522123894,
      "grad_norm": 3.658012559526469e-08,
      "learning_rate": 9.20166326020988e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22465168,
      "step": 39555
    },
    {
      "epoch": 694.0353982300885,
      "grad_norm": 7.01834395044898e-08,
      "learning_rate": 8.996524926933035e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22467592,
      "step": 39560
    },
    {
      "epoch": 694.1238938053098,
      "grad_norm": 5.982777651070137e-08,
      "learning_rate": 8.793698394781723e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22470280,
      "step": 39565
    },
    {
      "epoch": 694.212389380531,
      "grad_norm": 6.003021724154678e-08,
      "learning_rate": 8.593183695030926e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22473576,
      "step": 39570
    },
    {
      "epoch": 694.3008849557522,
      "grad_norm": 5.730890606514549e-08,
      "learning_rate": 8.39498085860757e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22475960,
      "step": 39575
    },
    {
      "epoch": 694.3893805309734,
      "grad_norm": 2.559386480527337e-08,
      "learning_rate": 8.199089916072211e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22478872,
      "step": 39580
    },
    {
      "epoch": 694.4778761061947,
      "grad_norm": 2.241480778764071e-08,
      "learning_rate": 8.005510897637346e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22481368,
      "step": 39585
    },
    {
      "epoch": 694.566371681416,
      "grad_norm": 2.5472360221101553e-08,
      "learning_rate": 7.8142438331541e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22484472,
      "step": 39590
    },
    {
      "epoch": 694.6548672566372,
      "grad_norm": 3.077883903301881e-08,
      "learning_rate": 7.625288752117209e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22487672,
      "step": 39595
    },
    {
      "epoch": 694.7433628318585,
      "grad_norm": 3.762900746551168e-08,
      "learning_rate": 7.4386456836667e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22490328,
      "step": 39600
    },
    {
      "epoch": 694.7433628318585,
      "eval_loss": 0.7776755690574646,
      "eval_runtime": 0.9339,
      "eval_samples_per_second": 26.77,
      "eval_steps_per_second": 13.92,
      "num_input_tokens_seen": 22490328,
      "step": 39600
    },
    {
      "epoch": 694.8318584070796,
      "grad_norm": 5.624158916361921e-08,
      "learning_rate": 7.254314656586214e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22493160,
      "step": 39605
    },
    {
      "epoch": 694.9203539823009,
      "grad_norm": 5.875026332091693e-08,
      "learning_rate": 7.07229569929968e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22496040,
      "step": 39610
    },
    {
      "epoch": 695.0,
      "grad_norm": 4.4388656306182384e-07,
      "learning_rate": 6.892588839879643e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22498816,
      "step": 39615
    },
    {
      "epoch": 695.0884955752213,
      "grad_norm": 1.5649906259795898e-08,
      "learning_rate": 6.71519410603727e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22501760,
      "step": 39620
    },
    {
      "epoch": 695.1769911504425,
      "grad_norm": 2.513178465335386e-08,
      "learning_rate": 6.540111525129011e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22504800,
      "step": 39625
    },
    {
      "epoch": 695.2654867256637,
      "grad_norm": 3.471575027447216e-08,
      "learning_rate": 6.367341124154934e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22507888,
      "step": 39630
    },
    {
      "epoch": 695.3539823008849,
      "grad_norm": 7.431927429024654e-08,
      "learning_rate": 6.19688292975873e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22510160,
      "step": 39635
    },
    {
      "epoch": 695.4424778761062,
      "grad_norm": 1.5086335736214096e-08,
      "learning_rate": 6.0287369682260336e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22512816,
      "step": 39640
    },
    {
      "epoch": 695.5309734513274,
      "grad_norm": 3.8050337991535343e-08,
      "learning_rate": 5.8629032654894384e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22515760,
      "step": 39645
    },
    {
      "epoch": 695.6194690265487,
      "grad_norm": 5.582302975426501e-08,
      "learning_rate": 5.699381847120155e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22518624,
      "step": 39650
    },
    {
      "epoch": 695.70796460177,
      "grad_norm": 5.59450157311403e-08,
      "learning_rate": 5.5381727383380094e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22521344,
      "step": 39655
    },
    {
      "epoch": 695.7964601769911,
      "grad_norm": 1.4608381171399287e-08,
      "learning_rate": 5.379275964001451e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22524096,
      "step": 39660
    },
    {
      "epoch": 695.8849557522124,
      "grad_norm": 2.8321748501980437e-08,
      "learning_rate": 5.222691548614211e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22527280,
      "step": 39665
    },
    {
      "epoch": 695.9734513274336,
      "grad_norm": 3.973908846433005e-08,
      "learning_rate": 5.068419516323641e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22530672,
      "step": 39670
    },
    {
      "epoch": 696.0530973451328,
      "grad_norm": 8.909488968811274e-08,
      "learning_rate": 4.91645989092071e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22532944,
      "step": 39675
    },
    {
      "epoch": 696.141592920354,
      "grad_norm": 1.9300210851724842e-07,
      "learning_rate": 4.7668126958400056e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22536400,
      "step": 39680
    },
    {
      "epoch": 696.2300884955753,
      "grad_norm": 3.585479646517342e-08,
      "learning_rate": 4.619477954159734e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22538688,
      "step": 39685
    },
    {
      "epoch": 696.3185840707964,
      "grad_norm": 3.232300471722738e-08,
      "learning_rate": 4.4744556885983884e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22542336,
      "step": 39690
    },
    {
      "epoch": 696.4070796460177,
      "grad_norm": 1.0651888260326814e-07,
      "learning_rate": 4.331745921523078e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22545184,
      "step": 39695
    },
    {
      "epoch": 696.4955752212389,
      "grad_norm": 1.773559610285247e-08,
      "learning_rate": 4.191348674937867e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22547920,
      "step": 39700
    },
    {
      "epoch": 696.5840707964602,
      "grad_norm": 2.0159424352073074e-08,
      "learning_rate": 4.0532639704971006e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22550592,
      "step": 39705
    },
    {
      "epoch": 696.6725663716815,
      "grad_norm": 3.4930959458279176e-08,
      "learning_rate": 3.917491829493747e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22553216,
      "step": 39710
    },
    {
      "epoch": 696.7610619469026,
      "grad_norm": 3.904590784031825e-08,
      "learning_rate": 3.78403227286439e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22556144,
      "step": 39715
    },
    {
      "epoch": 696.8495575221239,
      "grad_norm": 3.688143479507744e-08,
      "learning_rate": 3.652885321192567e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22559264,
      "step": 39720
    },
    {
      "epoch": 696.9380530973451,
      "grad_norm": 3.0796755368101e-08,
      "learning_rate": 3.524050994702099e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22562016,
      "step": 39725
    },
    {
      "epoch": 697.0176991150443,
      "grad_norm": 3.0661539085485856e-08,
      "learning_rate": 3.3975293132604276e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22564280,
      "step": 39730
    },
    {
      "epoch": 697.1061946902655,
      "grad_norm": 2.907765761506198e-08,
      "learning_rate": 3.2733202963786125e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22567400,
      "step": 39735
    },
    {
      "epoch": 697.1946902654868,
      "grad_norm": 1.3817262001225572e-08,
      "learning_rate": 3.15142396321133e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22570520,
      "step": 39740
    },
    {
      "epoch": 697.2831858407079,
      "grad_norm": 3.027406236810748e-08,
      "learning_rate": 3.0318403325552132e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22573336,
      "step": 39745
    },
    {
      "epoch": 697.3716814159292,
      "grad_norm": 4.938921804864549e-08,
      "learning_rate": 2.914569422855506e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22576152,
      "step": 39750
    },
    {
      "epoch": 697.4601769911504,
      "grad_norm": 3.282753979760855e-08,
      "learning_rate": 2.7996112521927462e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22578920,
      "step": 39755
    },
    {
      "epoch": 697.5486725663717,
      "grad_norm": 5.4546340777505975e-08,
      "learning_rate": 2.68696583829775e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22581352,
      "step": 39760
    },
    {
      "epoch": 697.637168141593,
      "grad_norm": 1.259524662344802e-08,
      "learning_rate": 2.576633198539957e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22584152,
      "step": 39765
    },
    {
      "epoch": 697.7256637168142,
      "grad_norm": 5.1487862862131806e-08,
      "learning_rate": 2.46861334993409e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22587384,
      "step": 39770
    },
    {
      "epoch": 697.8141592920354,
      "grad_norm": 2.9510447419056618e-08,
      "learning_rate": 2.3629063091384903e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22590296,
      "step": 39775
    },
    {
      "epoch": 697.9026548672566,
      "grad_norm": 4.303086598156369e-08,
      "learning_rate": 2.2595120924567834e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22593032,
      "step": 39780
    },
    {
      "epoch": 697.9911504424779,
      "grad_norm": 1.0244991521801694e-08,
      "learning_rate": 2.158430715829551e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22596136,
      "step": 39785
    },
    {
      "epoch": 698.070796460177,
      "grad_norm": 1.7763627013778205e-08,
      "learning_rate": 2.059662194849321e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22598352,
      "step": 39790
    },
    {
      "epoch": 698.1592920353983,
      "grad_norm": 9.447489901504014e-08,
      "learning_rate": 1.9632065447422463e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22601024,
      "step": 39795
    },
    {
      "epoch": 698.2477876106195,
      "grad_norm": 2.7160782067880973e-08,
      "learning_rate": 1.8690637803880916e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22604096,
      "step": 39800
    },
    {
      "epoch": 698.2477876106195,
      "eval_loss": 0.794979989528656,
      "eval_runtime": 0.9438,
      "eval_samples_per_second": 26.488,
      "eval_steps_per_second": 13.774,
      "num_input_tokens_seen": 22604096,
      "step": 39800
    },
    {
      "epoch": 698.3362831858407,
      "grad_norm": 2.6614801029722912e-08,
      "learning_rate": 1.7772339163019123e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22607040,
      "step": 39805
    },
    {
      "epoch": 698.4247787610619,
      "grad_norm": 3.505069301468211e-08,
      "learning_rate": 1.6877169666457138e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22609568,
      "step": 39810
    },
    {
      "epoch": 698.5132743362832,
      "grad_norm": 3.931443259830303e-08,
      "learning_rate": 1.6005129452234532e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22612320,
      "step": 39815
    },
    {
      "epoch": 698.6017699115044,
      "grad_norm": 4.5311438157114026e-08,
      "learning_rate": 1.5156218654843733e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22615328,
      "step": 39820
    },
    {
      "epoch": 698.6902654867257,
      "grad_norm": 4.4661319975602964e-08,
      "learning_rate": 1.4330437405196683e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22618784,
      "step": 39825
    },
    {
      "epoch": 698.7787610619469,
      "grad_norm": 1.9648030757934976e-08,
      "learning_rate": 1.352778583062486e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22621376,
      "step": 39830
    },
    {
      "epoch": 698.8672566371681,
      "grad_norm": 2.96684827816307e-08,
      "learning_rate": 1.2748264054929237e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22623936,
      "step": 39835
    },
    {
      "epoch": 698.9557522123894,
      "grad_norm": 5.997301855131809e-08,
      "learning_rate": 1.1991872198297004e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22627120,
      "step": 39840
    },
    {
      "epoch": 699.0353982300885,
      "grad_norm": 5.395724755885567e-08,
      "learning_rate": 1.1258610377384847e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22629960,
      "step": 39845
    },
    {
      "epoch": 699.1238938053098,
      "grad_norm": 4.8975341115919946e-08,
      "learning_rate": 1.0548478705268982e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 22632840,
      "step": 39850
    },
    {
      "epoch": 699.212389380531,
      "grad_norm": 3.48762050350615e-08,
      "learning_rate": 9.86147729147846e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22635656,
      "step": 39855
    },
    {
      "epoch": 699.3008849557522,
      "grad_norm": 4.512328999339843e-08,
      "learning_rate": 9.197606241928557e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22638632,
      "step": 39860
    },
    {
      "epoch": 699.3893805309734,
      "grad_norm": 4.064805736447852e-08,
      "learning_rate": 8.556865659004042e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22641400,
      "step": 39865
    },
    {
      "epoch": 699.4778761061947,
      "grad_norm": 1.6698278315629977e-08,
      "learning_rate": 7.939255641525867e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22644264,
      "step": 39870
    },
    {
      "epoch": 699.566371681416,
      "grad_norm": 5.893304333426386e-08,
      "learning_rate": 7.344776284751164e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22647128,
      "step": 39875
    },
    {
      "epoch": 699.6548672566372,
      "grad_norm": 3.594860586986215e-08,
      "learning_rate": 6.773427680323296e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22649928,
      "step": 39880
    },
    {
      "epoch": 699.7433628318585,
      "grad_norm": 4.4241421193191854e-08,
      "learning_rate": 6.225209916355112e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22652840,
      "step": 39885
    },
    {
      "epoch": 699.8318584070796,
      "grad_norm": 3.267879478130453e-08,
      "learning_rate": 5.7001230774123e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22655384,
      "step": 39890
    },
    {
      "epoch": 699.9203539823009,
      "grad_norm": 2.6052697776890454e-08,
      "learning_rate": 5.198167244446772e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22658232,
      "step": 39895
    },
    {
      "epoch": 700.0,
      "grad_norm": 1.6486517040448234e-07,
      "learning_rate": 4.71934249487993e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22660488,
      "step": 39900
    },
    {
      "epoch": 700.0884955752213,
      "grad_norm": 4.879915138644719e-08,
      "learning_rate": 4.2636489025527075e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22663336,
      "step": 39905
    },
    {
      "epoch": 700.1769911504425,
      "grad_norm": 3.912854396048715e-08,
      "learning_rate": 3.831086537742223e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22666120,
      "step": 39910
    },
    {
      "epoch": 700.2654867256637,
      "grad_norm": 2.035062074412508e-08,
      "learning_rate": 3.4216554671451236e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22668648,
      "step": 39915
    },
    {
      "epoch": 700.3539823008849,
      "grad_norm": 1.4439789808307069e-08,
      "learning_rate": 3.035355753894242e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22671464,
      "step": 39920
    },
    {
      "epoch": 700.4424778761062,
      "grad_norm": 3.807232928920712e-08,
      "learning_rate": 2.6721874575752477e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22674360,
      "step": 39925
    },
    {
      "epoch": 700.5309734513274,
      "grad_norm": 6.602136437550143e-08,
      "learning_rate": 2.3321506341933418e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22677432,
      "step": 39930
    },
    {
      "epoch": 700.6194690265487,
      "grad_norm": 7.97234704918992e-08,
      "learning_rate": 2.0152453361732546e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22680296,
      "step": 39935
    },
    {
      "epoch": 700.70796460177,
      "grad_norm": 2.34460344472609e-08,
      "learning_rate": 1.7214716123925554e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22683304,
      "step": 39940
    },
    {
      "epoch": 700.7964601769911,
      "grad_norm": 6.456873791194084e-08,
      "learning_rate": 1.4508295081649968e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22686520,
      "step": 39945
    },
    {
      "epoch": 700.8849557522124,
      "grad_norm": 4.315196022730561e-08,
      "learning_rate": 1.2033190652238623e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 22689240,
      "step": 39950
    },
    {
      "epoch": 700.9734513274336,
      "grad_norm": 7.595829032425172e-08,
      "learning_rate": 9.78940321721966e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22692696,
      "step": 39955
    },
    {
      "epoch": 701.0530973451328,
      "grad_norm": 1.8965266690429416e-08,
      "learning_rate": 7.776933122816132e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22694888,
      "step": 39960
    },
    {
      "epoch": 701.141592920354,
      "grad_norm": 4.067050696221486e-08,
      "learning_rate": 5.99578067927986e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22698120,
      "step": 39965
    },
    {
      "epoch": 701.2300884955753,
      "grad_norm": 3.855930685858766e-08,
      "learning_rate": 4.445946161224512e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22701000,
      "step": 39970
    },
    {
      "epoch": 701.3185840707964,
      "grad_norm": 4.8956064091498774e-08,
      "learning_rate": 3.127429807792126e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22704392,
      "step": 39975
    },
    {
      "epoch": 701.4070796460177,
      "grad_norm": 3.980955654014906e-08,
      "learning_rate": 2.040231822320049e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22707096,
      "step": 39980
    },
    {
      "epoch": 701.4955752212389,
      "grad_norm": 3.9464016055035245e-08,
      "learning_rate": 1.1843523723409354e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 22710184,
      "step": 39985
    },
    {
      "epoch": 701.5840707964602,
      "grad_norm": 1.7197296031667975e-08,
      "learning_rate": 5.597915897492811e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 22713240,
      "step": 39990
    },
    {
      "epoch": 701.6725663716815,
      "grad_norm": 1.9337814194386738e-07,
      "learning_rate": 1.6654957113448885e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 22715928,
      "step": 39995
    },
    {
      "epoch": 701.7610619469026,
      "grad_norm": 2.723468561782738e-08,
      "learning_rate": 4.626377114735902e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 22718312,
      "step": 40000
    },
    {
      "epoch": 701.7610619469026,
      "eval_loss": 0.8050469756126404,
      "eval_runtime": 0.9464,
      "eval_samples_per_second": 26.417,
      "eval_steps_per_second": 13.737,
      "num_input_tokens_seen": 22718312,
      "step": 40000
    },
    {
      "epoch": 701.7610619469026,
      "num_input_tokens_seen": 22718312,
      "step": 40000,
      "total_flos": 9.51301231531438e+16,
      "train_loss": 0.015288212282552349,
      "train_runtime": 8564.9201,
      "train_samples_per_second": 18.681,
      "train_steps_per_second": 4.67
    }
  ],
  "logging_steps": 5,
  "max_steps": 40000,
  "num_input_tokens_seen": 22718312,
  "num_train_epochs": 715,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.51301231531438e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}