trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6158148646354675,
  "best_model_checkpoint": "distilbert_rand_50_v1_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": 1.2927180528640747,
      "learning_rate": 4.9e-05,
      "loss": 0.6147,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6158148646354675,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4395,
      "eval_samples_per_second": 2373.267,
      "eval_steps_per_second": 11.377,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.549296498298645,
      "learning_rate": 4.8e-05,
      "loss": 0.5935,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6207744479179382,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4824,
      "eval_samples_per_second": 2162.311,
      "eval_steps_per_second": 10.366,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.8696980476379395,
      "learning_rate": 4.7e-05,
      "loss": 0.548,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6759347915649414,
      "eval_loss": 0.6408491134643555,
      "eval_matthews_correlation": 0.015191433834210268,
      "eval_runtime": 0.4792,
      "eval_samples_per_second": 2176.741,
      "eval_steps_per_second": 10.435,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.342837333679199,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4961,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6567593216896057,
      "eval_loss": 0.7421315312385559,
      "eval_matthews_correlation": 0.08852460407026867,
      "eval_runtime": 0.442,
      "eval_samples_per_second": 2359.861,
      "eval_steps_per_second": 11.313,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.1100683212280273,
      "learning_rate": 4.5e-05,
      "loss": 0.4413,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6605944633483887,
      "eval_loss": 0.7400658130645752,
      "eval_matthews_correlation": 0.10614316120782262,
      "eval_runtime": 0.4746,
      "eval_samples_per_second": 2197.487,
      "eval_steps_per_second": 10.534,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.1626508235931396,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3876,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6088207364082336,
      "eval_loss": 0.8084709644317627,
      "eval_matthews_correlation": 0.0850925091452348,
      "eval_runtime": 0.4634,
      "eval_samples_per_second": 2250.557,
      "eval_steps_per_second": 10.789,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.5135374676947501,
      "train_runtime": 58.1664,
      "train_samples_per_second": 7350.458,
      "train_steps_per_second": 29.226
    }
  ],
  "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
}