{
  "best_metric": 1.7843894958496094,
  "best_model_checkpoint": "./outputs/checkpoint-500",
  "epoch": 0.36429872495446264,
  "eval_steps": 100,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.1908,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.0238757133483887,
      "eval_runtime": 146.4096,
      "eval_samples_per_second": 42.852,
      "eval_steps_per_second": 5.362,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.9736,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.9310673475265503,
      "eval_runtime": 159.7901,
      "eval_samples_per_second": 39.264,
      "eval_steps_per_second": 4.913,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.9004,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.8707531690597534,
      "eval_runtime": 143.355,
      "eval_samples_per_second": 43.765,
      "eval_steps_per_second": 5.476,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.8488,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.825339436531067,
      "eval_runtime": 143.4115,
      "eval_samples_per_second": 43.748,
      "eval_steps_per_second": 5.474,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.7978,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.7843894958496094,
      "eval_runtime": 143.3888,
      "eval_samples_per_second": 43.755,
      "eval_steps_per_second": 5.475,
      "step": 500
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 2.983213629977395e+16,
  "trial_name": null,
  "trial_params": null
}