{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 92,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 1.0794392824172974,
      "learning_rate": 0.00018857142857142857,
      "loss": 2.473,
      "mean_token_accuracy": 0.4876363660797461,
      "step": 11,
      "timestamp_in_seconds": 1739654450.8883092
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": 0.9399147629737854,
      "learning_rate": 0.00016761904761904763,
      "loss": 2.0504,
      "mean_token_accuracy": 0.5424026841245708,
      "step": 22,
      "timestamp_in_seconds": 1739654468.0399427
    },
    {
      "contract_score": 0.5477064942317923,
      "epoch": 1.0,
      "eval_loss": 1.9136688709259033,
      "eval_mean_token_accuracy": 0.5333141551105782,
      "eval_runtime": 0.5456,
      "eval_samples_per_second": 36.657,
      "eval_steps_per_second": 5.499,
      "step": 23,
      "timestamp_in_seconds": 1739654506.413823
    },
    {
      "epoch": 1.4444444444444444,
      "grad_norm": 1.0410069227218628,
      "learning_rate": 0.00014666666666666666,
      "loss": 1.8615,
      "mean_token_accuracy": 0.5735910261967099,
      "step": 33,
      "timestamp_in_seconds": 1739654522.0010076
    },
    {
      "epoch": 1.9333333333333333,
      "grad_norm": 1.0421518087387085,
      "learning_rate": 0.00012571428571428572,
      "loss": 1.774,
      "mean_token_accuracy": 0.5750006535677639,
      "step": 44,
      "timestamp_in_seconds": 1739654539.2094088
    },
    {
      "contract_score": 0.46365774544845184,
      "epoch": 2.0,
      "eval_loss": 1.9464867115020752,
      "eval_mean_token_accuracy": 0.5649391065404323,
      "eval_runtime": 0.5442,
      "eval_samples_per_second": 36.748,
      "eval_steps_per_second": 5.512,
      "step": 46,
      "timestamp_in_seconds": 1739654578.4262042
    },
    {
      "epoch": 2.4,
      "grad_norm": 1.4570425748825073,
      "learning_rate": 0.00010476190476190477,
      "loss": 1.477,
      "mean_token_accuracy": 0.6416130446353153,
      "step": 55,
      "timestamp_in_seconds": 1739654592.388432
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 2.1125121116638184,
      "learning_rate": 8.380952380952382e-05,
      "loss": 1.4106,
      "mean_token_accuracy": 0.6289000944974809,
      "step": 66,
      "timestamp_in_seconds": 1739654609.5367165
    },
    {
      "contract_score": 0.5204585063638901,
      "epoch": 3.0,
      "eval_loss": 2.0934700965881348,
      "eval_mean_token_accuracy": 0.6413903497350207,
      "eval_runtime": 0.5519,
      "eval_samples_per_second": 36.239,
      "eval_steps_per_second": 5.436,
      "step": 69,
      "timestamp_in_seconds": 1739654650.6984093
    },
    {
      "epoch": 3.3555555555555556,
      "grad_norm": 2.1363847255706787,
      "learning_rate": 6.285714285714286e-05,
      "loss": 1.0621,
      "mean_token_accuracy": 0.7378539553704223,
      "step": 77,
      "timestamp_in_seconds": 1739654663.3004222
    },
    {
      "epoch": 3.8444444444444446,
      "grad_norm": 2.2876200675964355,
      "learning_rate": 4.190476190476191e-05,
      "loss": 0.9692,
      "mean_token_accuracy": 0.742552637419056,
      "step": 88,
      "timestamp_in_seconds": 1739654680.534879
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.320528030395508,
      "eval_mean_token_accuracy": 0.7298687658266787,
      "eval_runtime": 0.5451,
      "eval_samples_per_second": 36.688,
      "eval_steps_per_second": 5.503,
      "step": 92
    }
  ],
  "logging_steps": 11,
  "max_steps": 110,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4920656685588480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}