trainer_state.json

{
  "best_metric": 0.9162657932295468,
  "best_model_checkpoint": "../output/bert-base-uncased-sst_bin/checkpoint-800",
  "epoch": 3.686635944700461,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.46,
      "eval_accuracy": 0.8956422018348624,
      "eval_f1": 0.8955816423073837,
      "eval_loss": 0.2795850932598114,
      "eval_precision": 0.8957204610192337,
      "eval_recall": 0.8955018102214364,
      "eval_runtime": 4.3773,
      "eval_samples_per_second": 199.208,
      "step": 100
    },
    {
      "epoch": 0.92,
      "eval_accuracy": 0.9071100917431193,
      "eval_f1": 0.9071099695811189,
      "eval_loss": 0.25760290026664734,
      "eval_precision": 0.9073623893036986,
      "eval_recall": 0.907394544076787,
      "eval_runtime": 4.3715,
      "eval_samples_per_second": 199.475,
      "step": 200
    },
    {
      "epoch": 1.38,
      "eval_accuracy": 0.9013761467889908,
      "eval_f1": 0.9011244014091935,
      "eval_loss": 0.3114422559738159,
      "eval_precision": 0.9033686440677966,
      "eval_recall": 0.9007535572956133,
      "eval_runtime": 4.3885,
      "eval_samples_per_second": 198.699,
      "step": 300
    },
    {
      "epoch": 1.84,
      "eval_accuracy": 0.9105504587155964,
      "eval_f1": 0.9105499881638128,
      "eval_loss": 0.3116317093372345,
      "eval_precision": 0.9109770732015411,
      "eval_recall": 0.9108992169739833,
      "eval_runtime": 4.3848,
      "eval_samples_per_second": 198.868,
      "step": 400
    },
    {
      "epoch": 2.3,
      "learning_rate": 1.0783410138248848e-05,
      "loss": 0.2779,
      "step": 500
    },
    {
      "epoch": 2.3,
      "eval_accuracy": 0.908256880733945,
      "eval_f1": 0.9082177722811356,
      "eval_loss": 0.3516693711280823,
      "eval_precision": 0.9082612265521379,
      "eval_recall": 0.9081838848193988,
      "eval_runtime": 4.3876,
      "eval_samples_per_second": 198.74,
      "step": 500
    },
    {
      "epoch": 2.76,
      "eval_accuracy": 0.9105504587155964,
      "eval_f1": 0.9105123279741073,
      "eval_loss": 0.35066622495651245,
      "eval_precision": 0.9105560117055096,
      "eval_recall": 0.9104782352445904,
      "eval_runtime": 4.359,
      "eval_samples_per_second": 200.046,
      "step": 600
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.911697247706422,
      "eval_f1": 0.9116711109123222,
      "eval_loss": 0.40561920404434204,
      "eval_precision": 0.9116560640262672,
      "eval_recall": 0.9116885577165952,
      "eval_runtime": 4.3615,
      "eval_samples_per_second": 199.932,
      "step": 700
    },
    {
      "epoch": 3.69,
      "eval_accuracy": 0.9162844036697247,
      "eval_f1": 0.9162657932295468,
      "eval_loss": 0.4281376898288727,
      "eval_precision": 0.9162339336353258,
      "eval_recall": 0.9163193567399175,
      "eval_runtime": 4.3626,
      "eval_samples_per_second": 199.882,
      "step": 800
    }
  ],
  "max_steps": 1085,
  "num_train_epochs": 5,
  "total_flos": 2146529877411840,
  "trial_name": null,
  "trial_params": null
}