last-checkpoint/trainer_state.json

{
  "best_metric": 1.2945491075515747,
  "best_model_checkpoint": "./outputs/checkpoint-400",
  "epoch": 0.5333333333333333,
  "eval_steps": 100,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 1.4887,
      "step": 100
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.360513687133789,
      "eval_runtime": 127.6803,
      "eval_samples_per_second": 15.116,
      "eval_steps_per_second": 1.895,
      "step": 100
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 1.1616,
      "step": 200
    },
    {
      "epoch": 0.27,
      "eval_loss": 1.3187366724014282,
      "eval_runtime": 126.439,
      "eval_samples_per_second": 15.264,
      "eval_steps_per_second": 1.914,
      "step": 200
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 1.1374,
      "step": 300
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.3053778409957886,
      "eval_runtime": 126.588,
      "eval_samples_per_second": 15.246,
      "eval_steps_per_second": 1.912,
      "step": 300
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 1.125,
      "step": 400
    },
    {
      "epoch": 0.53,
      "eval_loss": 1.2945491075515747,
      "eval_runtime": 126.4785,
      "eval_samples_per_second": 15.26,
      "eval_steps_per_second": 1.913,
      "step": 400
    }
  ],
  "logging_steps": 100,
  "max_steps": 2250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.4194150656786432e+16,
  "trial_name": null,
  "trial_params": null
}