trainer_state.json

{
  "best_metric": 0.6873350739479065,
  "best_model_checkpoint": "hBERTv2_data_aug_wnli/checkpoint-654",
  "epoch": 8.0,
  "global_step": 1744,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.699,
      "step": 218
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6895423531532288,
      "eval_runtime": 0.1007,
      "eval_samples_per_second": 704.868,
      "eval_steps_per_second": 9.928,
      "step": 218
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.6947,
      "step": 436
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6886346340179443,
      "eval_runtime": 0.098,
      "eval_samples_per_second": 724.276,
      "eval_steps_per_second": 10.201,
      "step": 436
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.702752293577982e-05,
      "loss": 0.6935,
      "step": 654
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6873350739479065,
      "eval_runtime": 0.0993,
      "eval_samples_per_second": 714.989,
      "eval_steps_per_second": 10.07,
      "step": 654
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.602752293577982e-05,
      "loss": 0.6937,
      "step": 872
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.692121684551239,
      "eval_runtime": 0.1012,
      "eval_samples_per_second": 701.584,
      "eval_steps_per_second": 9.881,
      "step": 872
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.502752293577982e-05,
      "loss": 0.6934,
      "step": 1090
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.689205527305603,
      "eval_runtime": 0.1019,
      "eval_samples_per_second": 696.888,
      "eval_steps_per_second": 9.815,
      "step": 1090
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.402752293577982e-05,
      "loss": 0.6932,
      "step": 1308
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6910760402679443,
      "eval_runtime": 0.1003,
      "eval_samples_per_second": 708.197,
      "eval_steps_per_second": 9.975,
      "step": 1308
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3027522935779816e-05,
      "loss": 0.6933,
      "step": 1526
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.6955257654190063,
      "eval_runtime": 0.0972,
      "eval_samples_per_second": 730.321,
      "eval_steps_per_second": 10.286,
      "step": 1526
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.202752293577982e-05,
      "loss": 0.6931,
      "step": 1744
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6907530426979065,
      "eval_runtime": 0.0956,
      "eval_samples_per_second": 742.91,
      "eval_steps_per_second": 10.464,
      "step": 1744
    },
    {
      "epoch": 8.0,
      "step": 1744,
      "total_flos": 5.544122456316314e+16,
      "train_loss": 0.6942421624419886,
      "train_runtime": 1727.2715,
      "train_samples_per_second": 1610.633,
      "train_steps_per_second": 6.311
    }
  ],
  "max_steps": 10900,
  "num_train_epochs": 50,
  "total_flos": 5.544122456316314e+16,
  "trial_name": null,
  "trial_params": null
}