{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.7272727272727275,
  "eval_steps": 2,
  "global_step": 14,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.494016408920288,
      "learning_rate": 4e-05,
      "loss": 1.9274,
      "mean_token_accuracy": 0.6064814814814815,
      "step": 1,
      "timestamp_in_seconds": 1739999004.8391037
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.3887760639190674,
      "learning_rate": 8e-05,
      "loss": 2.0207,
      "mean_token_accuracy": 0.638888888888889,
      "step": 2,
      "timestamp_in_seconds": 1739999006.9712703
    },
    {
      "contract_score": 0.09736460012102884,
      "epoch": 0.7272727272727273,
      "eval_loss": 2.037353038787842,
      "eval_mean_token_accuracy": 0.6296296296296297,
      "eval_runtime": 0.1187,
      "eval_samples_per_second": 25.274,
      "eval_steps_per_second": 8.425,
      "step": 2,
      "timestamp_in_seconds": 1739999017.1328475
    },
    {
      "epoch": 1.3636363636363638,
      "grad_norm": 5.325081825256348,
      "learning_rate": 0.00012,
      "loss": 4.0803,
      "mean_token_accuracy": 0.6296296296296295,
      "step": 3,
      "timestamp_in_seconds": 1739999019.7841036
    },
    {
      "epoch": 1.7272727272727273,
      "grad_norm": 2.8036787509918213,
      "learning_rate": 0.00016,
      "loss": 1.748,
      "mean_token_accuracy": 0.6481481481481481,
      "step": 4,
      "timestamp_in_seconds": 1739999021.9279227
    },
    {
      "contract_score": 0.09418203565809462,
      "epoch": 1.7272727272727273,
      "eval_loss": 1.3247987031936646,
      "eval_mean_token_accuracy": 0.7530864197530864,
      "eval_runtime": 0.117,
      "eval_samples_per_second": 25.651,
      "eval_steps_per_second": 8.55,
      "step": 4,
      "timestamp_in_seconds": 1739999032.830725
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 6.123953819274902,
      "learning_rate": 0.0002,
      "loss": 2.5705,
      "mean_token_accuracy": 0.7333333333333333,
      "step": 5,
      "timestamp_in_seconds": 1739999035.1792696
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 3.178056001663208,
      "learning_rate": 0.0001866666666666667,
      "loss": 0.6802,
      "mean_token_accuracy": 0.8009259259259259,
      "step": 6,
      "timestamp_in_seconds": 1739999037.02721
    },
    {
      "contract_score": 0.2091630621443665,
      "epoch": 2.7272727272727275,
      "eval_loss": 0.5565303564071655,
      "eval_mean_token_accuracy": 0.8271604938271605,
      "eval_runtime": 0.1077,
      "eval_samples_per_second": 27.864,
      "eval_steps_per_second": 9.288,
      "step": 6,
      "timestamp_in_seconds": 1739999046.6422505
    },
    {
      "epoch": 3.3636363636363638,
      "grad_norm": 3.407052993774414,
      "learning_rate": 0.00017333333333333334,
      "loss": 0.9566,
      "mean_token_accuracy": 0.8629629629629629,
      "step": 7,
      "timestamp_in_seconds": 1739999049.0149553
    },
    {
      "epoch": 3.7272727272727275,
      "grad_norm": 1.0389703512191772,
      "learning_rate": 0.00016,
      "loss": 0.3408,
      "mean_token_accuracy": 0.8935185185185186,
      "step": 8,
      "timestamp_in_seconds": 1739999050.89696
    },
    {
      "contract_score": 0.45928009275406123,
      "epoch": 3.7272727272727275,
      "eval_loss": 0.36926987767219543,
      "eval_mean_token_accuracy": 0.8765432098765432,
      "eval_runtime": 0.1064,
      "eval_samples_per_second": 28.19,
      "eval_steps_per_second": 9.397,
      "step": 8,
      "timestamp_in_seconds": 1739999060.110364
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 1.9661706686019897,
      "learning_rate": 0.00014666666666666666,
      "loss": 0.7016,
      "mean_token_accuracy": 0.9074074074074074,
      "step": 9,
      "timestamp_in_seconds": 1739999062.509756
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.739223837852478,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.254,
      "mean_token_accuracy": 0.9305555555555556,
      "step": 10,
      "timestamp_in_seconds": 1739999064.4200356
    },
    {
      "contract_score": 0.47112671382843496,
      "epoch": 4.7272727272727275,
      "eval_loss": 0.37535592913627625,
      "eval_mean_token_accuracy": 0.8765432098765432,
      "eval_runtime": 0.1186,
      "eval_samples_per_second": 25.305,
      "eval_steps_per_second": 8.435,
      "step": 10,
      "timestamp_in_seconds": 1739999072.7365825
    },
    {
      "epoch": 5.363636363636363,
      "grad_norm": 2.233830451965332,
      "learning_rate": 0.00012,
      "loss": 0.6345,
      "mean_token_accuracy": 0.9037037037037037,
      "step": 11,
      "timestamp_in_seconds": 1739999075.1556246
    },
    {
      "epoch": 5.7272727272727275,
      "grad_norm": 0.6788499355316162,
      "learning_rate": 0.00010666666666666667,
      "loss": 0.1907,
      "mean_token_accuracy": 0.9259259259259258,
      "step": 12,
      "timestamp_in_seconds": 1739999077.0596292
    },
    {
      "contract_score": 0.4590138147747705,
      "epoch": 5.7272727272727275,
      "eval_loss": 0.41145431995391846,
      "eval_mean_token_accuracy": 0.8641975308641975,
      "eval_runtime": 0.1171,
      "eval_samples_per_second": 25.62,
      "eval_steps_per_second": 8.54,
      "step": 12,
      "timestamp_in_seconds": 1739999085.6286044
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 1.4969104528427124,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.4587,
      "mean_token_accuracy": 0.9185185185185185,
      "step": 13,
      "timestamp_in_seconds": 1739999088.0561044
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.5461616516113281,
      "learning_rate": 8e-05,
      "loss": 0.1827,
      "mean_token_accuracy": 0.9305555555555556,
      "step": 14,
      "timestamp_in_seconds": 1739999089.991205
    },
    {
      "epoch": 6.7272727272727275,
      "eval_loss": 0.3865008056163788,
      "eval_mean_token_accuracy": 0.8888888888888888,
      "eval_runtime": 0.1126,
      "eval_samples_per_second": 26.654,
      "eval_steps_per_second": 8.885,
      "step": 14,
      "timestamp_in_seconds": 1739999090.1253753
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 2,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 402957454614528.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}