trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6163831949234009,
  "best_model_checkpoint": "distilbert_km_10_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": 1.2094887495040894,
      "learning_rate": 4.9e-05,
      "loss": 0.6152,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6922339200973511,
      "eval_loss": 0.6163831949234009,
      "eval_matthews_correlation": 0.0463559874942472,
      "eval_runtime": 0.4381,
      "eval_samples_per_second": 2380.74,
      "eval_steps_per_second": 11.413,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.425511360168457,
      "learning_rate": 4.8e-05,
      "loss": 0.5852,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6816874146461487,
      "eval_loss": 0.6236035227775574,
      "eval_matthews_correlation": 0.035270463896731476,
      "eval_runtime": 0.447,
      "eval_samples_per_second": 2333.557,
      "eval_steps_per_second": 11.187,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.9508463144302368,
      "learning_rate": 4.7e-05,
      "loss": 0.5375,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6490891575813293,
      "eval_loss": 0.6427673697471619,
      "eval_matthews_correlation": 0.013377558812376731,
      "eval_runtime": 0.4414,
      "eval_samples_per_second": 2363.16,
      "eval_steps_per_second": 11.329,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.35090970993042,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4823,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6548417806625366,
      "eval_loss": 0.6820145845413208,
      "eval_matthews_correlation": 0.08987782248314084,
      "eval_runtime": 0.4746,
      "eval_samples_per_second": 2197.759,
      "eval_steps_per_second": 10.536,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6743202209472656,
      "learning_rate": 4.5e-05,
      "loss": 0.4156,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6270374059677124,
      "eval_loss": 0.7539588212966919,
      "eval_matthews_correlation": 0.05165125853472949,
      "eval_runtime": 0.5401,
      "eval_samples_per_second": 1931.204,
      "eval_steps_per_second": 9.258,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.225551128387451,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3536,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6490891575813293,
      "eval_loss": 0.8314696550369263,
      "eval_matthews_correlation": 0.08229406882199065,
      "eval_runtime": 0.5544,
      "eval_samples_per_second": 1881.152,
      "eval_steps_per_second": 9.018,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.49820103832319673,
      "train_runtime": 58.8164,
      "train_samples_per_second": 7269.236,
      "train_steps_per_second": 28.904
    }
  ],
  "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
}