run-33/checkpoint-384/trainer_state.json

{
  "best_metric": 0.8140900195694716,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-33/checkpoint-384",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.441394090652466,
      "learning_rate": 0.00039892736546732013,
      "loss": 0.5568,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7818003913894325,
      "eval_f1": 0.8085836909871245,
      "eval_loss": 0.45964643359184265,
      "eval_mcc": 0.5870562340513658,
      "eval_precision": 0.7201834862385321,
      "eval_recall": 0.9217221135029354,
      "eval_runtime": 67.0533,
      "eval_samples_per_second": 15.242,
      "eval_steps_per_second": 0.477,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.07141637802124,
      "learning_rate": 0.0002659515769782134,
      "loss": 0.4779,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8140900195694716,
      "eval_f1": 0.8227611940298508,
      "eval_loss": 0.44287753105163574,
      "eval_mcc": 0.6312089367953639,
      "eval_precision": 0.786096256684492,
      "eval_recall": 0.863013698630137,
      "eval_runtime": 66.4724,
      "eval_samples_per_second": 15.375,
      "eval_steps_per_second": 0.481,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 768,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9425898264750469,
    "learning_rate": 0.0005319031539564268,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 16,
    "temperature": 47
  }
}