trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 36,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 6.378284954778628,
      "learning_rate": 9.409606321741776e-06,
      "loss": 5.2151,
      "step": 10
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 2.1144381561005003,
      "learning_rate": 5.490085701647805e-06,
      "loss": 0.2414,
      "step": 20
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 3.040689056367678,
      "learning_rate": 1.134947733186315e-06,
      "loss": 0.1187,
      "step": 30
    },
    {
      "epoch": 2.0,
      "step": 36,
      "total_flos": 5029231591424.0,
      "train_loss": 1.562221081720458,
      "train_runtime": 368.0524,
      "train_samples_per_second": 2.983,
      "train_steps_per_second": 0.098
    }
  ],
  "logging_steps": 10,
  "max_steps": 36,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5029231591424.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}