{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.888888888888889,
  "eval_steps": 22,
  "global_step": 66,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 0.9441103339195251,
      "learning_rate": 0.00018857142857142857,
      "loss": 2.4469,
      "mean_token_accuracy": 0.496908648333388,
      "step": 11,
      "timestamp_in_seconds": 1740061537.3234556
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": 0.8327683806419373,
      "learning_rate": 0.00016761904761904763,
      "loss": 1.9834,
      "mean_token_accuracy": 0.547112869493232,
      "step": 22,
      "timestamp_in_seconds": 1740061554.110968
    },
    {
      "contract_score": 0.5459380726606408,
      "epoch": 0.9777777777777777,
      "eval_loss": 2.0452332496643066,
      "eval_mean_token_accuracy": 0.5491126227556767,
      "eval_runtime": 0.4673,
      "eval_samples_per_second": 42.796,
      "eval_steps_per_second": 6.419,
      "step": 22,
      "timestamp_in_seconds": 1740061577.7832978
    },
    {
      "epoch": 1.4444444444444444,
      "grad_norm": 0.8525232672691345,
      "learning_rate": 0.00014666666666666666,
      "loss": 1.8628,
      "mean_token_accuracy": 0.5669253493436065,
      "step": 33,
      "timestamp_in_seconds": 1740061593.950284
    },
    {
      "epoch": 1.9333333333333333,
      "grad_norm": 1.0234193801879883,
      "learning_rate": 0.00012571428571428572,
      "loss": 1.7168,
      "mean_token_accuracy": 0.5950726888540735,
      "step": 44,
      "timestamp_in_seconds": 1740061610.9770184
    },
    {
      "contract_score": 0.5468032690716348,
      "epoch": 1.9333333333333333,
      "eval_loss": 2.051072835922241,
      "eval_mean_token_accuracy": 0.5417702471605499,
      "eval_runtime": 0.4643,
      "eval_samples_per_second": 43.077,
      "eval_steps_per_second": 6.462,
      "step": 44,
      "timestamp_in_seconds": 1740061636.2054253
    },
    {
      "epoch": 2.4,
      "grad_norm": 1.2619707584381104,
      "learning_rate": 0.00010476190476190477,
      "loss": 1.5316,
      "mean_token_accuracy": 0.6254073605451984,
      "step": 55,
      "timestamp_in_seconds": 1740061652.4050982
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 2.0393614768981934,
      "learning_rate": 8.380952380952382e-05,
      "loss": 1.3513,
      "mean_token_accuracy": 0.6561945758293672,
      "step": 66,
      "timestamp_in_seconds": 1740061669.0619414
    },
    {
      "epoch": 2.888888888888889,
      "eval_loss": 2.1966896057128906,
      "eval_mean_token_accuracy": 0.5312554883977948,
      "eval_runtime": 0.4649,
      "eval_samples_per_second": 43.017,
      "eval_steps_per_second": 6.453,
      "step": 66,
      "timestamp_in_seconds": 1740061669.5637474
    }
  ],
  "logging_steps": 11,
  "max_steps": 110,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 22,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3611563447824384.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}