trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 22,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.0573,
      "step": 1
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.0946,
      "step": 5
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.1207,
      "step": 10
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.1361,
      "step": 15
    },
    {
      "epoch": 0.91,
      "grad_norm": 9.846180319816957,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.1322,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1973026990890503,
      "eval_runtime": 42.044,
      "eval_samples_per_second": 15.959,
      "eval_steps_per_second": 0.499,
      "step": 22
    },
    {
      "epoch": 1.0,
      "step": 22,
      "total_flos": 314069483520.0,
      "train_loss": 1.1124052893031726,
      "train_runtime": 253.7254,
      "train_samples_per_second": 2.664,
      "train_steps_per_second": 0.087
    }
  ],
  "logging_steps": 5,
  "max_steps": 22,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 314069483520.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}