trainer_state.json

{
  "best_metric": 0.12910962104797363,
  "best_model_checkpoint": "./HFTB-SmolLM2-135M-Instruct-OTCMedicinePHv2\\checkpoint-100",
  "epoch": 3.883495145631068,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.883495145631068,
      "grad_norm": 0.26728370785713196,
      "learning_rate": 0.0005,
      "loss": 0.4797,
      "step": 100
    },
    {
      "epoch": 3.883495145631068,
      "eval_loss": 0.12910962104797363,
      "eval_runtime": 14.3923,
      "eval_samples_per_second": 36.825,
      "eval_steps_per_second": 4.655,
      "step": 100
    },
    {
      "epoch": 3.883495145631068,
      "step": 100,
      "total_flos": 258445456835328.0,
      "train_loss": 0.47970603942871093,
      "train_runtime": 930.7066,
      "train_samples_per_second": 5.157,
      "train_steps_per_second": 0.107
    }
  ],
  "logging_steps": 100,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.01
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 258445456835328.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}