trainer_state.json

{
  "best_metric": 0.6173672676086426,
  "best_model_checkpoint": "hBERTv2_new_pretrain_48_cola/checkpoint-201",
  "epoch": 8.0,
  "global_step": 536,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.6319,
      "step": 67
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6185283064842224,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.702,
      "eval_samples_per_second": 612.826,
      "eval_steps_per_second": 5.288,
      "step": 67
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.6208,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6358161568641663,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.7092,
      "eval_samples_per_second": 610.213,
      "eval_steps_per_second": 5.266,
      "step": 134
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6124,
      "step": 201
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6173672676086426,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.7188,
      "eval_samples_per_second": 606.823,
      "eval_steps_per_second": 5.236,
      "step": 201
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6048,
      "step": 268
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.644297182559967,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.7206,
      "eval_samples_per_second": 606.187,
      "eval_steps_per_second": 5.231,
      "step": 268
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.5847,
      "step": 335
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.6520816087722778,
      "eval_matthews_correlation": -0.020702674026557004,
      "eval_runtime": 1.7509,
      "eval_samples_per_second": 595.691,
      "eval_steps_per_second": 5.14,
      "step": 335
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.5739,
      "step": 402
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6335605978965759,
      "eval_matthews_correlation": 0.0667441892820069,
      "eval_runtime": 1.7041,
      "eval_samples_per_second": 612.048,
      "eval_steps_per_second": 5.281,
      "step": 402
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.5562,
      "step": 469
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.6433131098747253,
      "eval_matthews_correlation": -0.010997352316591035,
      "eval_runtime": 1.7024,
      "eval_samples_per_second": 612.678,
      "eval_steps_per_second": 5.287,
      "step": 469
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.54,
      "step": 536
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6524495482444763,
      "eval_matthews_correlation": 0.09096530150080366,
      "eval_runtime": 1.7024,
      "eval_samples_per_second": 612.66,
      "eval_steps_per_second": 5.287,
      "step": 536
    },
    {
      "epoch": 8.0,
      "step": 536,
      "total_flos": 1.0010769903058944e+16,
      "train_loss": 0.5906015652329174,
      "train_runtime": 380.9368,
      "train_samples_per_second": 1122.365,
      "train_steps_per_second": 8.794
    }
  ],
  "max_steps": 3350,
  "num_train_epochs": 50,
  "total_flos": 1.0010769903058944e+16,
  "trial_name": null,
  "trial_params": null
}