{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3921113689095128,
  "eval_steps": 50,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02320185614849188,
      "grad_norm": 27.06661296487487,
      "learning_rate": 5e-07,
      "loss": 1.6534,
      "step": 5
    },
    {
      "epoch": 0.04640371229698376,
      "grad_norm": 14.570294011958959,
      "learning_rate": 1e-06,
      "loss": 1.5286,
      "step": 10
    },
    {
      "epoch": 0.06960556844547564,
      "grad_norm": 7.608713229824519,
      "learning_rate": 9.998470286265414e-07,
      "loss": 1.2854,
      "step": 15
    },
    {
      "epoch": 0.09280742459396751,
      "grad_norm": 6.212274190189952,
      "learning_rate": 9.993882081071305e-07,
      "loss": 1.2312,
      "step": 20
    },
    {
      "epoch": 0.11600928074245939,
      "grad_norm": 5.767015997797787,
      "learning_rate": 9.986238191873872e-07,
      "loss": 1.1896,
      "step": 25
    },
    {
      "epoch": 0.13921113689095127,
      "grad_norm": 5.073635902534103,
      "learning_rate": 9.975543295858033e-07,
      "loss": 1.1918,
      "step": 30
    },
    {
      "epoch": 0.16241299303944315,
      "grad_norm": 4.788195681953361,
      "learning_rate": 9.961803937075514e-07,
      "loss": 1.1547,
      "step": 35
    },
    {
      "epoch": 0.18561484918793503,
      "grad_norm": 4.642283394950232,
      "learning_rate": 9.945028522440653e-07,
      "loss": 1.1326,
      "step": 40
    },
    {
      "epoch": 0.2088167053364269,
      "grad_norm": 4.915483171626106,
      "learning_rate": 9.925227316586314e-07,
      "loss": 1.1409,
      "step": 45
    },
    {
      "epoch": 0.23201856148491878,
      "grad_norm": 4.785752094079727,
      "learning_rate": 9.902412435583125e-07,
      "loss": 1.1348,
      "step": 50
    },
    {
      "epoch": 0.23201856148491878,
      "eval_loss": 1.1267675161361694,
      "eval_runtime": 105.5696,
      "eval_samples_per_second": 58.056,
      "eval_steps_per_second": 0.909,
      "step": 50
    },
    {
      "epoch": 0.2552204176334107,
      "grad_norm": 4.914635335083288,
      "learning_rate": 9.876597839525813e-07,
      "loss": 1.1167,
      "step": 55
    },
    {
      "epoch": 0.27842227378190254,
      "grad_norm": 4.9477576848370095,
      "learning_rate": 9.847799323991233e-07,
      "loss": 1.1059,
      "step": 60
    },
    {
      "epoch": 0.30162412993039445,
      "grad_norm": 4.9360133375019295,
      "learning_rate": 9.816034510373285e-07,
      "loss": 1.1173,
      "step": 65
    },
    {
      "epoch": 0.3248259860788863,
      "grad_norm": 4.7405884580433675,
      "learning_rate": 9.781322835100637e-07,
      "loss": 1.1091,
      "step": 70
    },
    {
      "epoch": 0.3480278422273782,
      "grad_norm": 4.835669121473599,
      "learning_rate": 9.743685537743856e-07,
      "loss": 1.0967,
      "step": 75
    },
    {
      "epoch": 0.37122969837587005,
      "grad_norm": 4.6796771647856055,
      "learning_rate": 9.70314564801922e-07,
      "loss": 1.0758,
      "step": 80
    },
    {
      "epoch": 0.39443155452436196,
      "grad_norm": 4.984167856345103,
      "learning_rate": 9.659727971697173e-07,
      "loss": 1.0862,
      "step": 85
    },
    {
      "epoch": 0.4176334106728538,
      "grad_norm": 4.771046487158264,
      "learning_rate": 9.613459075424033e-07,
      "loss": 1.0795,
      "step": 90
    },
    {
      "epoch": 0.4408352668213457,
      "grad_norm": 4.605260375673115,
      "learning_rate": 9.564367270466245e-07,
      "loss": 1.0677,
      "step": 95
    },
    {
      "epoch": 0.46403712296983757,
      "grad_norm": 4.698279384122072,
      "learning_rate": 9.51248259538713e-07,
      "loss": 1.08,
      "step": 100
    },
    {
      "epoch": 0.46403712296983757,
      "eval_loss": 1.0813250541687012,
      "eval_runtime": 104.9985,
      "eval_samples_per_second": 58.372,
      "eval_steps_per_second": 0.914,
      "step": 100
    },
    {
      "epoch": 0.4872389791183295,
      "grad_norm": 5.05046913930874,
      "learning_rate": 9.457836797666721e-07,
      "loss": 1.0944,
      "step": 105
    },
    {
      "epoch": 0.5104408352668214,
      "grad_norm": 5.042019310011314,
      "learning_rate": 9.400463314275941e-07,
      "loss": 1.0946,
      "step": 110
    },
    {
      "epoch": 0.5336426914153132,
      "grad_norm": 4.6774297543026435,
      "learning_rate": 9.340397251217008e-07,
      "loss": 1.067,
      "step": 115
    },
    {
      "epoch": 0.5568445475638051,
      "grad_norm": 5.124932810617846,
      "learning_rate": 9.27767536204258e-07,
      "loss": 1.0664,
      "step": 120
    },
    {
      "epoch": 0.580046403712297,
      "grad_norm": 5.129237449751678,
      "learning_rate": 9.212336025366787e-07,
      "loss": 1.0635,
      "step": 125
    },
    {
      "epoch": 0.6032482598607889,
      "grad_norm": 5.116692236159179,
      "learning_rate": 9.144419221381918e-07,
      "loss": 1.0667,
      "step": 130
    },
    {
      "epoch": 0.6264501160092807,
      "grad_norm": 4.99964002013004,
      "learning_rate": 9.073966507395121e-07,
      "loss": 1.0637,
      "step": 135
    },
    {
      "epoch": 0.6496519721577726,
      "grad_norm": 5.177637510626285,
      "learning_rate": 9.001020992400085e-07,
      "loss": 1.0634,
      "step": 140
    },
    {
      "epoch": 0.6728538283062645,
      "grad_norm": 5.4138117417423075,
      "learning_rate": 8.925627310699274e-07,
      "loss": 1.0637,
      "step": 145
    },
    {
      "epoch": 0.6960556844547564,
      "grad_norm": 5.246483799648231,
      "learning_rate": 8.84783159459285e-07,
      "loss": 1.0619,
      "step": 150
    },
    {
      "epoch": 0.6960556844547564,
      "eval_loss": 1.055559754371643,
      "eval_runtime": 104.7917,
      "eval_samples_per_second": 58.487,
      "eval_steps_per_second": 0.916,
      "step": 150
    },
    {
      "epoch": 0.7192575406032483,
      "grad_norm": 5.122602649783386,
      "learning_rate": 8.767681446150976e-07,
      "loss": 1.0575,
      "step": 155
    },
    {
      "epoch": 0.7424593967517401,
      "grad_norm": 5.29929198962553,
      "learning_rate": 8.68522590808682e-07,
      "loss": 1.0505,
      "step": 160
    },
    {
      "epoch": 0.765661252900232,
      "grad_norm": 4.828480264754899,
      "learning_rate": 8.600515433748001e-07,
      "loss": 1.0518,
      "step": 165
    },
    {
      "epoch": 0.7888631090487239,
      "grad_norm": 5.567752966669097,
      "learning_rate": 8.51360185624495e-07,
      "loss": 1.0378,
      "step": 170
    },
    {
      "epoch": 0.8120649651972158,
      "grad_norm": 5.249257083574839,
      "learning_rate": 8.424538356734956e-07,
      "loss": 1.0358,
      "step": 175
    },
    {
      "epoch": 0.8352668213457076,
      "grad_norm": 5.3602912051067575,
      "learning_rate": 8.333379431881397e-07,
      "loss": 1.0199,
      "step": 180
    },
    {
      "epoch": 0.8584686774941995,
      "grad_norm": 5.127092342189952,
      "learning_rate": 8.240180860508026e-07,
      "loss": 1.0404,
      "step": 185
    },
    {
      "epoch": 0.8816705336426914,
      "grad_norm": 4.841085798771151,
      "learning_rate": 8.144999669468713e-07,
      "loss": 1.0537,
      "step": 190
    },
    {
      "epoch": 0.9048723897911833,
      "grad_norm": 4.891876601750079,
      "learning_rate": 8.047894098753539e-07,
      "loss": 1.0461,
      "step": 195
    },
    {
      "epoch": 0.9280742459396751,
      "grad_norm": 5.235026926317955,
      "learning_rate": 7.948923565852597e-07,
      "loss": 1.0468,
      "step": 200
    },
    {
      "epoch": 0.9280742459396751,
      "eval_loss": 1.0335279703140259,
      "eval_runtime": 104.9086,
      "eval_samples_per_second": 58.422,
      "eval_steps_per_second": 0.915,
      "step": 200
    },
    {
      "epoch": 0.951276102088167,
      "grad_norm": 5.158788907122558,
      "learning_rate": 7.848148629399285e-07,
      "loss": 1.0159,
      "step": 205
    },
    {
      "epoch": 0.974477958236659,
      "grad_norm": 5.333986831951347,
      "learning_rate": 7.745630952115363e-07,
      "loss": 1.0224,
      "step": 210
    },
    {
      "epoch": 0.9976798143851509,
      "grad_norm": 4.977110057341932,
      "learning_rate": 7.641433263080418e-07,
      "loss": 1.0242,
      "step": 215
    },
    {
      "epoch": 1.0208816705336428,
      "grad_norm": 5.799017323239554,
      "learning_rate": 7.535619319348865e-07,
      "loss": 0.9359,
      "step": 220
    },
    {
      "epoch": 1.0440835266821347,
      "grad_norm": 5.8085173406660955,
      "learning_rate": 7.428253866937918e-07,
      "loss": 0.9014,
      "step": 225
    },
    {
      "epoch": 1.0672853828306264,
      "grad_norm": 5.888836993626545,
      "learning_rate": 7.319402601210447e-07,
      "loss": 0.8932,
      "step": 230
    },
    {
      "epoch": 1.0904872389791183,
      "grad_norm": 5.701985019154487,
      "learning_rate": 7.209132126676933e-07,
      "loss": 0.8953,
      "step": 235
    },
    {
      "epoch": 1.1136890951276102,
      "grad_norm": 5.4287275730825,
      "learning_rate": 7.097509916241145e-07,
      "loss": 0.9011,
      "step": 240
    },
    {
      "epoch": 1.136890951276102,
      "grad_norm": 5.486828404933326,
      "learning_rate": 6.984604269914436e-07,
      "loss": 0.879,
      "step": 245
    },
    {
      "epoch": 1.160092807424594,
      "grad_norm": 5.841423335192377,
      "learning_rate": 6.870484273023967e-07,
      "loss": 0.8999,
      "step": 250
    },
    {
      "epoch": 1.160092807424594,
      "eval_loss": 1.0276210308074951,
      "eval_runtime": 104.8459,
      "eval_samples_per_second": 58.457,
      "eval_steps_per_second": 0.916,
      "step": 250
    },
    {
      "epoch": 1.1832946635730859,
      "grad_norm": 6.104179231791857,
      "learning_rate": 6.755219753940388e-07,
      "loss": 0.8799,
      "step": 255
    },
    {
      "epoch": 1.2064965197215778,
      "grad_norm": 5.683759928140861,
      "learning_rate": 6.638881241350883e-07,
      "loss": 0.8848,
      "step": 260
    },
    {
      "epoch": 1.2296983758700697,
      "grad_norm": 6.2781173429736565,
      "learning_rate": 6.52153992110368e-07,
      "loss": 0.8753,
      "step": 265
    },
    {
      "epoch": 1.2529002320185616,
      "grad_norm": 5.1836939832127555,
      "learning_rate": 6.403267592650466e-07,
      "loss": 0.8898,
      "step": 270
    },
    {
      "epoch": 1.2761020881670533,
      "grad_norm": 5.855887931011074,
      "learning_rate": 6.28413662511334e-07,
      "loss": 0.8782,
      "step": 275
    },
    {
      "epoch": 1.2993039443155452,
      "grad_norm": 5.740445426512396,
      "learning_rate": 6.164219913003207e-07,
      "loss": 0.8931,
      "step": 280
    },
    {
      "epoch": 1.322505800464037,
      "grad_norm": 5.566584948356301,
      "learning_rate": 6.043590831616676e-07,
      "loss": 0.8696,
      "step": 285
    },
    {
      "epoch": 1.345707656612529,
      "grad_norm": 6.107582813588156,
      "learning_rate": 5.92232319213878e-07,
      "loss": 0.8822,
      "step": 290
    },
    {
      "epoch": 1.368909512761021,
      "grad_norm": 5.795907587506582,
      "learning_rate": 5.800491196478988e-07,
      "loss": 0.8995,
      "step": 295
    },
    {
      "epoch": 1.3921113689095128,
      "grad_norm": 5.6643118349028425,
      "learning_rate": 5.678169391868127e-07,
      "loss": 0.8818,
      "step": 300
    },
    {
      "epoch": 1.3921113689095128,
      "eval_loss": 1.016647219657898,
      "eval_runtime": 104.7971,
      "eval_samples_per_second": 58.484,
      "eval_steps_per_second": 0.916,
      "step": 300
    }
  ],
  "logging_steps": 5,
  "max_steps": 645,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1768702026448896.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}