trainer_state.json

{
  "best_metric": 0.061126627027988434,
  "best_model_checkpoint": "Real-Fake-Class/checkpoint-3365",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3365,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1485884101040119,
      "grad_norm": 3.8796277046203613,
      "learning_rate": 2.7979041916167666e-06,
      "loss": 0.2434,
      "step": 500
    },
    {
      "epoch": 0.2971768202080238,
      "grad_norm": 10.399986267089844,
      "learning_rate": 2.5733532934131734e-06,
      "loss": 0.1139,
      "step": 1000
    },
    {
      "epoch": 0.4457652303120357,
      "grad_norm": 18.36138343811035,
      "learning_rate": 2.3488023952095807e-06,
      "loss": 0.0984,
      "step": 1500
    },
    {
      "epoch": 0.5943536404160475,
      "grad_norm": 8.896696090698242,
      "learning_rate": 2.124251497005988e-06,
      "loss": 0.0915,
      "step": 2000
    },
    {
      "epoch": 0.7429420505200595,
      "grad_norm": 3.4077460765838623,
      "learning_rate": 1.8997005988023954e-06,
      "loss": 0.0839,
      "step": 2500
    },
    {
      "epoch": 0.8915304606240714,
      "grad_norm": 35.51144790649414,
      "learning_rate": 1.6751497005988027e-06,
      "loss": 0.0768,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9807306470489885,
      "eval_loss": 0.061126627027988434,
      "eval_model_preparation_time": 0.0062,
      "eval_runtime": 946.099,
      "eval_samples_per_second": 75.861,
      "eval_steps_per_second": 9.483,
      "step": 3365
    }
  ],
  "logging_steps": 500,
  "max_steps": 6730,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.016949169130807e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}