{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.177777777777778,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 1.9879343509674072,
      "learning_rate": 0.0002,
      "loss": 2.2707,
      "mean_token_accuracy": 0.5294100181649257,
      "step": 5,
      "timestamp_in_seconds": 1739766577.8310535
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 1.009445309638977,
      "learning_rate": 0.00019047619047619048,
      "loss": 2.0272,
      "mean_token_accuracy": 0.5268474393109208,
      "step": 10,
      "timestamp_in_seconds": 1739766593.809831
    },
    {
      "contract_score": 0.5466966775887976,
      "epoch": 0.8888888888888888,
      "eval_loss": 1.8623384237289429,
      "eval_mean_token_accuracy": 0.5099889012208657,
      "eval_runtime": 1.2647,
      "eval_samples_per_second": 7.907,
      "eval_steps_per_second": 1.581,
      "step": 10,
      "timestamp_in_seconds": 1739766613.2455263
    },
    {
      "epoch": 1.2666666666666666,
      "grad_norm": 0.8467594981193542,
      "learning_rate": 0.00018095238095238095,
      "loss": 1.8597,
      "mean_token_accuracy": 0.5747608879576899,
      "step": 15,
      "timestamp_in_seconds": 1739766627.466834
    },
    {
      "epoch": 1.7111111111111112,
      "grad_norm": 1.00863516330719,
      "learning_rate": 0.00017142857142857143,
      "loss": 1.5547,
      "mean_token_accuracy": 0.609207512410729,
      "step": 20,
      "timestamp_in_seconds": 1739766643.4715047
    },
    {
      "contract_score": 0.517531061316395,
      "epoch": 1.7111111111111112,
      "eval_loss": 1.909725546836853,
      "eval_mean_token_accuracy": 0.5060673325934147,
      "eval_runtime": 1.2694,
      "eval_samples_per_second": 7.878,
      "eval_steps_per_second": 1.576,
      "step": 20,
      "timestamp_in_seconds": 1739766662.6723397
    },
    {
      "epoch": 2.088888888888889,
      "grad_norm": 1.0704385042190552,
      "learning_rate": 0.00016190476190476192,
      "loss": 1.4427,
      "mean_token_accuracy": 0.6196816753107063,
      "step": 25,
      "timestamp_in_seconds": 1739766676.269455
    },
    {
      "epoch": 2.533333333333333,
      "grad_norm": 1.5208722352981567,
      "learning_rate": 0.00015238095238095237,
      "loss": 1.1007,
      "mean_token_accuracy": 0.690360556370996,
      "step": 30,
      "timestamp_in_seconds": 1739766691.5902622
    },
    {
      "contract_score": 0.5104592473887242,
      "epoch": 2.533333333333333,
      "eval_loss": 2.0625336170196533,
      "eval_mean_token_accuracy": 0.48486866444691085,
      "eval_runtime": 1.284,
      "eval_samples_per_second": 7.788,
      "eval_steps_per_second": 1.558,
      "step": 30,
      "timestamp_in_seconds": 1739766712.605668
    },
    {
      "epoch": 2.977777777777778,
      "grad_norm": 2.2304763793945312,
      "learning_rate": 0.00014285714285714287,
      "loss": 1.0283,
      "mean_token_accuracy": 0.7108713038621913,
      "step": 35,
      "timestamp_in_seconds": 1739766729.188164
    },
    {
      "epoch": 3.3555555555555556,
      "grad_norm": 1.914406180381775,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.5798,
      "mean_token_accuracy": 0.8472582884142912,
      "step": 40,
      "timestamp_in_seconds": 1739766742.8015592
    },
    {
      "contract_score": 0.5328695250138283,
      "epoch": 3.3555555555555556,
      "eval_loss": 2.535480499267578,
      "eval_mean_token_accuracy": 0.49589345172031074,
      "eval_runtime": 1.2706,
      "eval_samples_per_second": 7.87,
      "eval_steps_per_second": 1.574,
      "step": 40,
      "timestamp_in_seconds": 1739766762.4698246
    },
    {
      "epoch": 3.8,
      "grad_norm": 3.442265272140503,
      "learning_rate": 0.0001238095238095238,
      "loss": 0.4816,
      "mean_token_accuracy": 0.8550400784948758,
      "step": 45,
      "timestamp_in_seconds": 1739766778.3630936
    },
    {
      "epoch": 4.177777777777778,
      "grad_norm": 1.0488659143447876,
      "learning_rate": 0.00011428571428571428,
      "loss": 0.2982,
      "mean_token_accuracy": 0.9270337260594542,
      "step": 50,
      "timestamp_in_seconds": 1739766791.7478123
    },
    {
      "epoch": 4.177777777777778,
      "eval_loss": 2.858454704284668,
      "eval_mean_token_accuracy": 0.49822419533851275,
      "eval_runtime": 1.2915,
      "eval_samples_per_second": 7.743,
      "eval_steps_per_second": 1.549,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 110,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7860274034343936.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}