trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6175770163536072,
  "best_model_checkpoint": "distilbert_km_5_v2_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4233615398406982,
      "learning_rate": 4.9e-05,
      "loss": 0.6134,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.693192720413208,
      "eval_loss": 0.6175770163536072,
      "eval_matthews_correlation": 0.06558874629318973,
      "eval_runtime": 0.4857,
      "eval_samples_per_second": 2147.423,
      "eval_steps_per_second": 10.294,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.713411808013916,
      "learning_rate": 4.8e-05,
      "loss": 0.5825,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6883988380432129,
      "eval_loss": 0.6216447353363037,
      "eval_matthews_correlation": 0.04453151163100391,
      "eval_runtime": 0.4747,
      "eval_samples_per_second": 2197.231,
      "eval_steps_per_second": 10.533,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.2151379585266113,
      "learning_rate": 4.7e-05,
      "loss": 0.5309,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6720997095108032,
      "eval_loss": 0.6438636779785156,
      "eval_matthews_correlation": 0.0643406632993554,
      "eval_runtime": 0.4406,
      "eval_samples_per_second": 2367.073,
      "eval_steps_per_second": 11.347,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.4566311836242676,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4729,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6893576383590698,
      "eval_loss": 0.7091285586357117,
      "eval_matthews_correlation": 0.1182190297907363,
      "eval_runtime": 0.4593,
      "eval_samples_per_second": 2270.684,
      "eval_steps_per_second": 10.885,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.4799399375915527,
      "learning_rate": 4.5e-05,
      "loss": 0.413,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.640460193157196,
      "eval_loss": 0.7329849600791931,
      "eval_matthews_correlation": 0.12596192265407888,
      "eval_runtime": 0.5165,
      "eval_samples_per_second": 2019.173,
      "eval_steps_per_second": 9.68,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.534843921661377,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3505,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6241610646247864,
      "eval_loss": 0.8544636368751526,
      "eval_matthews_correlation": 0.07397001596782858,
      "eval_runtime": 0.4487,
      "eval_samples_per_second": 2324.582,
      "eval_steps_per_second": 11.144,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.49387895827199896,
      "train_runtime": 59.7018,
      "train_samples_per_second": 7161.431,
      "train_steps_per_second": 28.475
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3398186177722368.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}