trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.6666666666666665,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.44,
      "grad_norm": 8.291538819094802,
      "learning_rate": 5e-06,
      "loss": 1.5309,
      "step": 1
    },
    {
      "epoch": 0.89,
      "eval_loss": 1.3587168455123901,
      "eval_runtime": 20.5727,
      "eval_samples_per_second": 11.034,
      "eval_steps_per_second": 0.729,
      "step": 2
    },
    {
      "epoch": 1.78,
      "eval_loss": 1.2818326950073242,
      "eval_runtime": 20.5072,
      "eval_samples_per_second": 11.069,
      "eval_steps_per_second": 0.731,
      "step": 4
    },
    {
      "epoch": 2.22,
      "grad_norm": 4.39963663833621,
      "learning_rate": 4.774575140626317e-07,
      "loss": 1.4409,
      "step": 5
    },
    {
      "epoch": 2.67,
      "eval_loss": 1.2695339918136597,
      "eval_runtime": 20.6856,
      "eval_samples_per_second": 10.974,
      "eval_steps_per_second": 0.725,
      "step": 6
    },
    {
      "epoch": 2.67,
      "step": 6,
      "total_flos": 1203933020160.0,
      "train_loss": 1.4362386465072632,
      "train_runtime": 162.3149,
      "train_samples_per_second": 2.421,
      "train_steps_per_second": 0.037
    }
  ],
  "logging_steps": 5,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1203933020160.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}