trainer_state.json

{
  "best_metric": 0.6001824140548706,
  "best_model_checkpoint": "bert_tiny_lda_book_cola/checkpoint-102",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 272,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9901484251022339,
      "learning_rate": 4.9e-05,
      "loss": 0.6151,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6124610304832458,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3444,
      "eval_samples_per_second": 3028.672,
      "eval_steps_per_second": 14.519,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1867563724517822,
      "learning_rate": 4.8e-05,
      "loss": 0.6005,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6067861318588257,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3546,
      "eval_samples_per_second": 2941.076,
      "eval_steps_per_second": 14.099,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5974591970443726,
      "learning_rate": 4.7e-05,
      "loss": 0.5742,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6883988380432129,
      "eval_loss": 0.6001824140548706,
      "eval_matthews_correlation": 0.12973131375830682,
      "eval_runtime": 0.3319,
      "eval_samples_per_second": 3142.634,
      "eval_steps_per_second": 15.065,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.08528733253479,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5244,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6816874146461487,
      "eval_loss": 0.6038218140602112,
      "eval_matthews_correlation": 0.1886355297952276,
      "eval_runtime": 0.3704,
      "eval_samples_per_second": 2815.743,
      "eval_steps_per_second": 13.498,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.010711669921875,
      "learning_rate": 4.5e-05,
      "loss": 0.465,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6941514611244202,
      "eval_loss": 0.6201082468032837,
      "eval_matthews_correlation": 0.20897220233431726,
      "eval_runtime": 0.3534,
      "eval_samples_per_second": 2950.992,
      "eval_steps_per_second": 14.147,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.648045063018799,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.408,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6941514611244202,
      "eval_loss": 0.6508533954620361,
      "eval_matthews_correlation": 0.2271309414393642,
      "eval_runtime": 0.3403,
      "eval_samples_per_second": 3065.135,
      "eval_steps_per_second": 14.694,
      "step": 204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.4453964233398438,
      "learning_rate": 4.3e-05,
      "loss": 0.3617,
      "step": 238
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6893576383590698,
      "eval_loss": 0.6936684250831604,
      "eval_matthews_correlation": 0.21894626694065988,
      "eval_runtime": 0.3454,
      "eval_samples_per_second": 3019.284,
      "eval_steps_per_second": 14.474,
      "step": 238
    },
    {
      "epoch": 8.0,
      "grad_norm": 6.493313312530518,
      "learning_rate": 4.2e-05,
      "loss": 0.3142,
      "step": 272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7037392258644104,
      "eval_loss": 0.7315708994865417,
      "eval_matthews_correlation": 0.23873407559405013,
      "eval_runtime": 0.3786,
      "eval_samples_per_second": 2754.759,
      "eval_steps_per_second": 13.206,
      "step": 272
    },
    {
      "epoch": 8.0,
      "step": 272,
      "total_flos": 1793902531485696.0,
      "train_loss": 0.48289138429305134,
      "train_runtime": 47.8258,
      "train_samples_per_second": 8939.74,
      "train_steps_per_second": 35.546
    }
  ],
  "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": 1793902531485696.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}