{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 936,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 134.19052124023438,
      "learning_rate": 4.996794871794872e-05,
      "loss": 9.1177,
      "step": 1
    },
    {
      "epoch": 0.1,
      "grad_norm": 46.553192138671875,
      "learning_rate": 4.9006410256410256e-05,
      "loss": 2.6116,
      "step": 31
    },
    {
      "epoch": 0.2,
      "grad_norm": 17.68166732788086,
      "learning_rate": 4.8012820512820516e-05,
      "loss": 0.8879,
      "step": 62
    },
    {
      "epoch": 0.3,
      "grad_norm": 17.172128677368164,
      "learning_rate": 4.701923076923077e-05,
      "loss": 1.0567,
      "step": 93
    },
    {
      "epoch": 0.4,
      "grad_norm": 104.65974426269531,
      "learning_rate": 4.602564102564102e-05,
      "loss": 0.9872,
      "step": 124
    },
    {
      "epoch": 0.5,
      "grad_norm": 16.036008834838867,
      "learning_rate": 4.503205128205128e-05,
      "loss": 0.9717,
      "step": 155
    },
    {
      "epoch": 0.6,
      "grad_norm": 27.394895553588867,
      "learning_rate": 4.403846153846154e-05,
      "loss": 0.8679,
      "step": 186
    },
    {
      "epoch": 0.7,
      "grad_norm": 37.157169342041016,
      "learning_rate": 4.30448717948718e-05,
      "loss": 0.8257,
      "step": 217
    },
    {
      "epoch": 0.79,
      "grad_norm": 8.90817928314209,
      "learning_rate": 4.205128205128206e-05,
      "loss": 0.8604,
      "step": 248
    },
    {
      "epoch": 0.89,
      "grad_norm": 13.38546085357666,
      "learning_rate": 4.105769230769231e-05,
      "loss": 0.8587,
      "step": 279
    },
    {
      "epoch": 0.99,
      "grad_norm": 32.03287124633789,
      "learning_rate": 4.006410256410257e-05,
      "loss": 0.8512,
      "step": 310
    },
    {
      "epoch": 1.09,
      "grad_norm": 20.29245376586914,
      "learning_rate": 3.9070512820512824e-05,
      "loss": 0.6728,
      "step": 341
    },
    {
      "epoch": 1.19,
      "grad_norm": 10.910124778747559,
      "learning_rate": 3.807692307692308e-05,
      "loss": 0.611,
      "step": 372
    },
    {
      "epoch": 1.29,
      "grad_norm": 14.703102111816406,
      "learning_rate": 3.708333333333334e-05,
      "loss": 0.6468,
      "step": 403
    },
    {
      "epoch": 1.39,
      "grad_norm": 9.841747283935547,
      "learning_rate": 3.608974358974359e-05,
      "loss": 0.5327,
      "step": 434
    },
    {
      "epoch": 1.49,
      "grad_norm": 15.09598445892334,
      "learning_rate": 3.5096153846153845e-05,
      "loss": 0.5669,
      "step": 465
    },
    {
      "epoch": 1.59,
      "grad_norm": 18.652040481567383,
      "learning_rate": 3.4102564102564105e-05,
      "loss": 0.4854,
      "step": 496
    },
    {
      "epoch": 1.69,
      "grad_norm": 8.733405113220215,
      "learning_rate": 3.310897435897436e-05,
      "loss": 0.5627,
      "step": 527
    },
    {
      "epoch": 1.79,
      "grad_norm": 21.200965881347656,
      "learning_rate": 3.211538461538462e-05,
      "loss": 0.5001,
      "step": 558
    },
    {
      "epoch": 1.89,
      "grad_norm": 7.368778228759766,
      "learning_rate": 3.112179487179487e-05,
      "loss": 0.4897,
      "step": 589
    },
    {
      "epoch": 1.99,
      "grad_norm": 16.03817367553711,
      "learning_rate": 3.012820512820513e-05,
      "loss": 0.5294,
      "step": 620
    },
    {
      "epoch": 2.09,
      "grad_norm": 35.705631256103516,
      "learning_rate": 2.913461538461539e-05,
      "loss": 0.4197,
      "step": 651
    },
    {
      "epoch": 2.19,
      "grad_norm": 10.462698936462402,
      "learning_rate": 2.8141025641025643e-05,
      "loss": 0.3589,
      "step": 682
    },
    {
      "epoch": 2.29,
      "grad_norm": 7.011005878448486,
      "learning_rate": 2.7147435897435896e-05,
      "loss": 0.4396,
      "step": 713
    },
    {
      "epoch": 2.38,
      "grad_norm": 4.32899284362793,
      "learning_rate": 2.6153846153846157e-05,
      "loss": 0.4604,
      "step": 744
    },
    {
      "epoch": 2.48,
      "grad_norm": 5.010735988616943,
      "learning_rate": 2.516025641025641e-05,
      "loss": 0.3764,
      "step": 775
    },
    {
      "epoch": 2.58,
      "grad_norm": 6.4428582191467285,
      "learning_rate": 2.4166666666666667e-05,
      "loss": 0.3035,
      "step": 806
    },
    {
      "epoch": 2.68,
      "grad_norm": 11.917937278747559,
      "learning_rate": 2.3173076923076924e-05,
      "loss": 0.3287,
      "step": 837
    },
    {
      "epoch": 2.78,
      "grad_norm": 18.710346221923828,
      "learning_rate": 2.217948717948718e-05,
      "loss": 0.3209,
      "step": 868
    },
    {
      "epoch": 2.88,
      "grad_norm": 5.8296990394592285,
      "learning_rate": 2.1185897435897437e-05,
      "loss": 0.329,
      "step": 899
    },
    {
      "epoch": 2.98,
      "grad_norm": 11.532828330993652,
      "learning_rate": 2.0192307692307694e-05,
      "loss": 0.2934,
      "step": 930
    }
  ],
  "logging_steps": 31,
  "max_steps": 1560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.1075205215224332e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}