trainer_state.json

{
  "best_metric": 0.24688944220542908,
  "best_model_checkpoint": "output_pipe/prom_300_tata/origin/checkpoint-200",
  "epoch": 4.0,
  "eval_steps": 200,
  "global_step": 308,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.2987012987012987,
      "grad_norm": 2.859584331512451,
      "learning_rate": 2.4186046511627908e-05,
      "loss": 0.5318,
      "step": 100
    },
    {
      "epoch": 2.5974025974025974,
      "grad_norm": 2.2398416996002197,
      "learning_rate": 1.2558139534883723e-05,
      "loss": 0.177,
      "step": 200
    },
    {
      "epoch": 2.5974025974025974,
      "eval_accuracy": 0.9070146818923328,
      "eval_f1": 0.9068202728109125,
      "eval_loss": 0.24688944220542908,
      "eval_matthews_correlation": 0.8137616947789336,
      "eval_precision": 0.9065771239739899,
      "eval_recall": 0.9071847976940323,
      "eval_runtime": 0.2447,
      "eval_samples_per_second": 2505.226,
      "eval_steps_per_second": 40.868,
      "step": 200
    },
    {
      "epoch": 3.896103896103896,
      "grad_norm": 3.448758363723755,
      "learning_rate": 9.302325581395349e-07,
      "loss": 0.0473,
      "step": 300
    },
    {
      "epoch": 4.0,
      "step": 308,
      "total_flos": 2985619352951040.0,
      "train_loss": 0.24640247574100246,
      "train_runtime": 36.8144,
      "train_samples_per_second": 532.836,
      "train_steps_per_second": 8.366
    }
  ],
  "logging_steps": 100,
  "max_steps": 308,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2985619352951040.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}