trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.2727272727272725,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7272727272727273,
      "eval_loss": 1.2520767450332642,
      "eval_runtime": 3.3961,
      "eval_samples_per_second": 21.201,
      "eval_steps_per_second": 0.589,
      "step": 2
    },
    {
      "epoch": 1.8636363636363638,
      "eval_loss": 1.2290809154510498,
      "eval_runtime": 3.2611,
      "eval_samples_per_second": 22.078,
      "eval_steps_per_second": 0.613,
      "step": 5
    },
    {
      "epoch": 2.2727272727272725,
      "eval_loss": 1.1983939409255981,
      "eval_runtime": 3.281,
      "eval_samples_per_second": 21.945,
      "eval_steps_per_second": 0.61,
      "step": 6
    },
    {
      "epoch": 2.2727272727272725,
      "step": 6,
      "total_flos": 9840843816960.0,
      "train_loss": 1.4369689623514812,
      "train_runtime": 552.6919,
      "train_samples_per_second": 7.322,
      "train_steps_per_second": 0.011
    }
  ],
  "logging_steps": 10,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9840843816960.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}