{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 20,
  "global_step": 695,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07194244604316546,
      "grad_norm": 0.7139045596122742,
      "learning_rate": 0.0002,
      "loss": 2.0811,
      "step": 10
    },
    {
      "epoch": 0.14388489208633093,
      "grad_norm": 0.5728554129600525,
      "learning_rate": 0.0002,
      "loss": 1.567,
      "step": 20
    },
    {
      "epoch": 0.2158273381294964,
      "grad_norm": 0.28863629698753357,
      "learning_rate": 0.0002,
      "loss": 1.357,
      "step": 30
    },
    {
      "epoch": 0.28776978417266186,
      "grad_norm": 0.3229255974292755,
      "learning_rate": 0.0002,
      "loss": 1.2521,
      "step": 40
    },
    {
      "epoch": 0.3597122302158273,
      "grad_norm": 0.33988726139068604,
      "learning_rate": 0.0002,
      "loss": 1.2164,
      "step": 50
    },
    {
      "epoch": 0.4316546762589928,
      "grad_norm": 0.40238669514656067,
      "learning_rate": 0.0002,
      "loss": 1.1385,
      "step": 60
    },
    {
      "epoch": 0.5035971223021583,
      "grad_norm": 0.38314664363861084,
      "learning_rate": 0.0002,
      "loss": 1.0973,
      "step": 70
    },
    {
      "epoch": 0.5755395683453237,
      "grad_norm": 0.3884238302707672,
      "learning_rate": 0.0002,
      "loss": 1.0836,
      "step": 80
    },
    {
      "epoch": 0.6474820143884892,
      "grad_norm": 0.3774580657482147,
      "learning_rate": 0.0002,
      "loss": 1.0409,
      "step": 90
    },
    {
      "epoch": 0.7194244604316546,
      "grad_norm": 0.43280744552612305,
      "learning_rate": 0.0002,
      "loss": 1.0462,
      "step": 100
    },
    {
      "epoch": 0.7913669064748201,
      "grad_norm": 0.3758571445941925,
      "learning_rate": 0.0002,
      "loss": 1.0206,
      "step": 110
    },
    {
      "epoch": 0.8633093525179856,
      "grad_norm": 0.3923165798187256,
      "learning_rate": 0.0002,
      "loss": 1.0219,
      "step": 120
    },
    {
      "epoch": 0.935251798561151,
      "grad_norm": 0.4213840663433075,
      "learning_rate": 0.0002,
      "loss": 0.9971,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9775540828704834,
      "eval_runtime": 21.5909,
      "eval_samples_per_second": 4.771,
      "eval_steps_per_second": 0.602,
      "step": 139
    },
    {
      "epoch": 1.0071942446043165,
      "grad_norm": 0.40076276659965515,
      "learning_rate": 0.0002,
      "loss": 0.9885,
      "step": 140
    },
    {
      "epoch": 1.079136690647482,
      "grad_norm": 0.3987070918083191,
      "learning_rate": 0.0002,
      "loss": 0.9555,
      "step": 150
    },
    {
      "epoch": 1.1510791366906474,
      "grad_norm": 0.41527315974235535,
      "learning_rate": 0.0002,
      "loss": 0.9391,
      "step": 160
    },
    {
      "epoch": 1.223021582733813,
      "grad_norm": 0.42107248306274414,
      "learning_rate": 0.0002,
      "loss": 0.9332,
      "step": 170
    },
    {
      "epoch": 1.2949640287769784,
      "grad_norm": 0.4587080776691437,
      "learning_rate": 0.0002,
      "loss": 0.9163,
      "step": 180
    },
    {
      "epoch": 1.3669064748201438,
      "grad_norm": 0.3977225124835968,
      "learning_rate": 0.0002,
      "loss": 0.9142,
      "step": 190
    },
    {
      "epoch": 1.4388489208633093,
      "grad_norm": 0.41111239790916443,
      "learning_rate": 0.0002,
      "loss": 0.9088,
      "step": 200
    },
    {
      "epoch": 1.5107913669064748,
      "grad_norm": 0.4326966404914856,
      "learning_rate": 0.0002,
      "loss": 0.9082,
      "step": 210
    },
    {
      "epoch": 1.5827338129496402,
      "grad_norm": 0.3831544816493988,
      "learning_rate": 0.0002,
      "loss": 0.908,
      "step": 220
    },
    {
      "epoch": 1.6546762589928057,
      "grad_norm": 0.39992555975914,
      "learning_rate": 0.0002,
      "loss": 0.9076,
      "step": 230
    },
    {
      "epoch": 1.7266187050359711,
      "grad_norm": 0.39961057901382446,
      "learning_rate": 0.0002,
      "loss": 0.9059,
      "step": 240
    },
    {
      "epoch": 1.7985611510791366,
      "grad_norm": 0.3854500949382782,
      "learning_rate": 0.0002,
      "loss": 0.8903,
      "step": 250
    },
    {
      "epoch": 1.870503597122302,
      "grad_norm": 0.4092749357223511,
      "learning_rate": 0.0002,
      "loss": 0.8904,
      "step": 260
    },
    {
      "epoch": 1.9424460431654675,
      "grad_norm": 0.40900877118110657,
      "learning_rate": 0.0002,
      "loss": 0.8679,
      "step": 270
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8940790891647339,
      "eval_runtime": 21.5978,
      "eval_samples_per_second": 4.769,
      "eval_steps_per_second": 0.602,
      "step": 278
    },
    {
      "epoch": 2.014388489208633,
      "grad_norm": 0.3746669888496399,
      "learning_rate": 0.0002,
      "loss": 0.8669,
      "step": 280
    },
    {
      "epoch": 2.0863309352517985,
      "grad_norm": 0.4724111258983612,
      "learning_rate": 0.0002,
      "loss": 0.8304,
      "step": 290
    },
    {
      "epoch": 2.158273381294964,
      "grad_norm": 0.39437365531921387,
      "learning_rate": 0.0002,
      "loss": 0.8318,
      "step": 300
    },
    {
      "epoch": 2.2302158273381294,
      "grad_norm": 0.4238971471786499,
      "learning_rate": 0.0002,
      "loss": 0.829,
      "step": 310
    },
    {
      "epoch": 2.302158273381295,
      "grad_norm": 0.37740206718444824,
      "learning_rate": 0.0002,
      "loss": 0.8235,
      "step": 320
    },
    {
      "epoch": 2.3741007194244603,
      "grad_norm": 0.40223780274391174,
      "learning_rate": 0.0002,
      "loss": 0.8328,
      "step": 330
    },
    {
      "epoch": 2.446043165467626,
      "grad_norm": 0.4160473048686981,
      "learning_rate": 0.0002,
      "loss": 0.8252,
      "step": 340
    },
    {
      "epoch": 2.5179856115107913,
      "grad_norm": 0.4427769184112549,
      "learning_rate": 0.0002,
      "loss": 0.8215,
      "step": 350
    },
    {
      "epoch": 2.5899280575539567,
      "grad_norm": 0.47273996472358704,
      "learning_rate": 0.0002,
      "loss": 0.819,
      "step": 360
    },
    {
      "epoch": 2.661870503597122,
      "grad_norm": 0.38981807231903076,
      "learning_rate": 0.0002,
      "loss": 0.8225,
      "step": 370
    },
    {
      "epoch": 2.7338129496402876,
      "grad_norm": 0.385079562664032,
      "learning_rate": 0.0002,
      "loss": 0.8045,
      "step": 380
    },
    {
      "epoch": 2.805755395683453,
      "grad_norm": 0.3796544075012207,
      "learning_rate": 0.0002,
      "loss": 0.8234,
      "step": 390
    },
    {
      "epoch": 2.8776978417266186,
      "grad_norm": 0.3638385236263275,
      "learning_rate": 0.0002,
      "loss": 0.8195,
      "step": 400
    },
    {
      "epoch": 2.949640287769784,
      "grad_norm": 0.37602856755256653,
      "learning_rate": 0.0002,
      "loss": 0.831,
      "step": 410
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8596920967102051,
      "eval_runtime": 21.5962,
      "eval_samples_per_second": 4.769,
      "eval_steps_per_second": 0.602,
      "step": 417
    },
    {
      "epoch": 3.0215827338129495,
      "grad_norm": 0.3858413100242615,
      "learning_rate": 0.0002,
      "loss": 0.8076,
      "step": 420
    },
    {
      "epoch": 3.093525179856115,
      "grad_norm": 0.39952272176742554,
      "learning_rate": 0.0002,
      "loss": 0.7777,
      "step": 430
    },
    {
      "epoch": 3.1654676258992804,
      "grad_norm": 0.42352095246315,
      "learning_rate": 0.0002,
      "loss": 0.7704,
      "step": 440
    },
    {
      "epoch": 3.237410071942446,
      "grad_norm": 0.4132436513900757,
      "learning_rate": 0.0002,
      "loss": 0.768,
      "step": 450
    },
    {
      "epoch": 3.3093525179856114,
      "grad_norm": 0.414110392332077,
      "learning_rate": 0.0002,
      "loss": 0.7631,
      "step": 460
    },
    {
      "epoch": 3.381294964028777,
      "grad_norm": 0.43551069498062134,
      "learning_rate": 0.0002,
      "loss": 0.7673,
      "step": 470
    },
    {
      "epoch": 3.4532374100719423,
      "grad_norm": 0.4042975604534149,
      "learning_rate": 0.0002,
      "loss": 0.7552,
      "step": 480
    },
    {
      "epoch": 3.5251798561151078,
      "grad_norm": 0.4289880394935608,
      "learning_rate": 0.0002,
      "loss": 0.7559,
      "step": 490
    },
    {
      "epoch": 3.597122302158273,
      "grad_norm": 0.39530348777770996,
      "learning_rate": 0.0002,
      "loss": 0.7661,
      "step": 500
    },
    {
      "epoch": 3.6690647482014387,
      "grad_norm": 0.4326631724834442,
      "learning_rate": 0.0002,
      "loss": 0.7605,
      "step": 510
    },
    {
      "epoch": 3.741007194244604,
      "grad_norm": 0.3833424150943756,
      "learning_rate": 0.0002,
      "loss": 0.7549,
      "step": 520
    },
    {
      "epoch": 3.81294964028777,
      "grad_norm": 0.40698808431625366,
      "learning_rate": 0.0002,
      "loss": 0.764,
      "step": 530
    },
    {
      "epoch": 3.884892086330935,
      "grad_norm": 0.39018431305885315,
      "learning_rate": 0.0002,
      "loss": 0.7653,
      "step": 540
    },
    {
      "epoch": 3.956834532374101,
      "grad_norm": 0.37253016233444214,
      "learning_rate": 0.0002,
      "loss": 0.7677,
      "step": 550
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.8444326519966125,
      "eval_runtime": 21.5712,
      "eval_samples_per_second": 4.775,
      "eval_steps_per_second": 0.603,
      "step": 556
    },
    {
      "epoch": 4.028776978417266,
      "grad_norm": 0.4029178023338318,
      "learning_rate": 0.0002,
      "loss": 0.7467,
      "step": 560
    },
    {
      "epoch": 4.100719424460432,
      "grad_norm": 0.382028192281723,
      "learning_rate": 0.0002,
      "loss": 0.7116,
      "step": 570
    },
    {
      "epoch": 4.172661870503597,
      "grad_norm": 0.42502614855766296,
      "learning_rate": 0.0002,
      "loss": 0.7298,
      "step": 580
    },
    {
      "epoch": 4.244604316546763,
      "grad_norm": 0.4241486191749573,
      "learning_rate": 0.0002,
      "loss": 0.7073,
      "step": 590
    },
    {
      "epoch": 4.316546762589928,
      "grad_norm": 0.4571862816810608,
      "learning_rate": 0.0002,
      "loss": 0.7106,
      "step": 600
    },
    {
      "epoch": 4.388489208633094,
      "grad_norm": 0.43121734261512756,
      "learning_rate": 0.0002,
      "loss": 0.7031,
      "step": 610
    },
    {
      "epoch": 4.460431654676259,
      "grad_norm": 0.40107443928718567,
      "learning_rate": 0.0002,
      "loss": 0.693,
      "step": 620
    },
    {
      "epoch": 4.532374100719425,
      "grad_norm": 0.4040583670139313,
      "learning_rate": 0.0002,
      "loss": 0.707,
      "step": 630
    },
    {
      "epoch": 4.60431654676259,
      "grad_norm": 0.4380245506763458,
      "learning_rate": 0.0002,
      "loss": 0.7066,
      "step": 640
    },
    {
      "epoch": 4.676258992805756,
      "grad_norm": 0.4263726472854614,
      "learning_rate": 0.0002,
      "loss": 0.7216,
      "step": 650
    },
    {
      "epoch": 4.748201438848921,
      "grad_norm": 0.4532300531864166,
      "learning_rate": 0.0002,
      "loss": 0.7337,
      "step": 660
    },
    {
      "epoch": 4.820143884892087,
      "grad_norm": 0.4051191210746765,
      "learning_rate": 0.0002,
      "loss": 0.7133,
      "step": 670
    },
    {
      "epoch": 4.892086330935252,
      "grad_norm": 0.434962660074234,
      "learning_rate": 0.0002,
      "loss": 0.7144,
      "step": 680
    },
    {
      "epoch": 4.9640287769784175,
      "grad_norm": 0.4131017029285431,
      "learning_rate": 0.0002,
      "loss": 0.7167,
      "step": 690
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.8359549641609192,
      "eval_runtime": 21.598,
      "eval_samples_per_second": 4.769,
      "eval_steps_per_second": 0.602,
      "step": 695
    }
  ],
  "logging_steps": 10,
  "max_steps": 695,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8287701530771456e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}