run-1/checkpoint-1263/trainer_state.json

{
  "best_metric": 0.9231651376146789,
  "best_model_checkpoint": "bert-base-uncased-finetuned-sst2/run-1/checkpoint-421",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1263,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9231651376146789,
      "eval_loss": 0.24830688536167145,
      "eval_runtime": 2.2848,
      "eval_samples_per_second": 381.657,
      "eval_steps_per_second": 24.072,
      "step": 421
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.174437165260315,
      "learning_rate": 3.136892070081642e-05,
      "loss": 0.3128,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9139908256880734,
      "eval_loss": 0.3306081295013428,
      "eval_runtime": 2.3075,
      "eval_samples_per_second": 377.901,
      "eval_steps_per_second": 23.835,
      "step": 842
    },
    {
      "epoch": 2.38,
      "grad_norm": 14.713961601257324,
      "learning_rate": 1.0812616178656248e-05,
      "loss": 0.1105,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9139908256880734,
      "eval_loss": 0.4116528332233429,
      "eval_runtime": 2.3531,
      "eval_samples_per_second": 370.572,
      "eval_steps_per_second": 23.373,
      "step": 1263
    }
  ],
  "logging_steps": 500,
  "max_steps": 1263,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 291460243797480.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.192522522297658e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 16,
    "seed": 6
  }
}