run-4/checkpoint-21050/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": 5.0,
  "eval_steps": 500,
  "global_step": 21050,
  "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
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.8592076301574707,
      "learning_rate": 1.0211967860226942e-05,
      "loss": 0.2063,
      "step": 12627
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.908256880733945,
      "eval_loss": 0.38172876834869385,
      "eval_runtime": 2.4147,
      "eval_samples_per_second": 361.12,
      "eval_steps_per_second": 22.777,
      "step": 12630
    },
    {
      "epoch": 4.0,
      "grad_norm": 19.267065048217773,
      "learning_rate": 5.109014907158535e-06,
      "loss": 0.1789,
      "step": 16836
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9139908256880734,
      "eval_loss": 0.4979610741138458,
      "eval_runtime": 2.4158,
      "eval_samples_per_second": 360.958,
      "eval_steps_per_second": 22.767,
      "step": 16840
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.8149940967559814,
      "learning_rate": 6.061954090126405e-09,
      "loss": 0.1621,
      "step": 21045
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9139908256880734,
      "eval_loss": 0.46082746982574463,
      "eval_runtime": 2.4166,
      "eval_samples_per_second": 360.832,
      "eval_steps_per_second": 22.759,
      "step": 21050
    }
  ],
  "logging_steps": 4209,
  "max_steps": 21050,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 6093988125399720.0,
  "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
  }
}