{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.852216748768473,
  "eval_steps": 100,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.9704433497536946,
      "eval_accuracy": 0.15024555367498127,
      "eval_loss": 5.138388633728027,
      "eval_runtime": 6.1346,
      "eval_samples_per_second": 7.172,
      "eval_steps_per_second": 7.172,
      "step": 100
    },
    {
      "epoch": 3.9408866995073892,
      "eval_accuracy": 0.2643821910955478,
      "eval_loss": 4.24171781539917,
      "eval_runtime": 6.1084,
      "eval_samples_per_second": 7.203,
      "eval_steps_per_second": 7.203,
      "step": 200
    },
    {
      "epoch": 5.911330049261084,
      "eval_accuracy": 0.31462108388600607,
      "eval_loss": 3.906160593032837,
      "eval_runtime": 6.1112,
      "eval_samples_per_second": 7.2,
      "eval_steps_per_second": 7.2,
      "step": 300
    },
    {
      "epoch": 7.8817733990147785,
      "eval_accuracy": 0.33759933082392307,
      "eval_loss": 3.740565538406372,
      "eval_runtime": 6.1355,
      "eval_samples_per_second": 7.171,
      "eval_steps_per_second": 7.171,
      "step": 400
    },
    {
      "epoch": 9.852216748768473,
      "grad_norm": 4.40625,
      "learning_rate": 0.0,
      "loss": 4.3954,
      "step": 500
    },
    {
      "epoch": 9.852216748768473,
      "eval_accuracy": 0.33628416816983486,
      "eval_loss": 3.7442381381988525,
      "eval_runtime": 6.0853,
      "eval_samples_per_second": 7.231,
      "eval_steps_per_second": 7.231,
      "step": 500
    },
    {
      "epoch": 9.852216748768473,
      "step": 500,
      "total_flos": 3.70944044433408e+17,
      "train_loss": 4.3953623046875,
      "train_runtime": 3726.3159,
      "train_samples_per_second": 2.179,
      "train_steps_per_second": 0.134
    }
  ],
  "logging_steps": 500,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 200,
  "total_flos": 3.70944044433408e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}