run-4/checkpoint-8420/trainer_state.json

{
  "best_metric": 0.9139908256880734,
  "best_model_checkpoint": "Tech-oriented/best_model_bert_uncasedbert-base-uncased-finetuned-sst2/run-4/checkpoint-8420",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 8420,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 20.266050338745117,
      "learning_rate": 2.0417873766363758e-05,
      "loss": 0.3739,
      "step": 4209
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9059633027522935,
      "eval_loss": 0.28132590651512146,
      "eval_runtime": 2.4226,
      "eval_samples_per_second": 359.943,
      "eval_steps_per_second": 22.703,
      "step": 4210
    },
    {
      "epoch": 2.0,
      "grad_norm": 31.52709197998047,
      "learning_rate": 1.531492081329535e-05,
      "loss": 0.2461,
      "step": 8418
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9139908256880734,
      "eval_loss": 0.35249146819114685,
      "eval_runtime": 2.4264,
      "eval_samples_per_second": 359.387,
      "eval_steps_per_second": 22.668,
      "step": 8420
    }
  ],
  "logging_steps": 4209,
  "max_steps": 21050,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2435456650613040.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.5520826719432166e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 16,
    "seed": 27,
    "weight_decay": 0.30671272708017133
  }
}