run-6/checkpoint-488/trainer_state.json

{
  "best_metric": 0.5782178217821782,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-6/checkpoint-488",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 488,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8588551878929138,
      "learning_rate": 0.0001547989976805511,
      "loss": 0.5028,
      "step": 244
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5089108910891089,
      "eval_f1": 0.1205673758865248,
      "eval_loss": 0.504381000995636,
      "eval_precision": 0.5666666666666667,
      "eval_recall": 0.06746031746031746,
      "eval_runtime": 14.9415,
      "eval_samples_per_second": 33.799,
      "eval_steps_per_second": 1.071,
      "step": 244
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0281460285186768,
      "learning_rate": 0.00012899916473379258,
      "loss": 0.4912,
      "step": 488
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5782178217821782,
      "eval_f1": 0.4714640198511167,
      "eval_loss": 0.484104186296463,
      "eval_precision": 0.6291390728476821,
      "eval_recall": 0.376984126984127,
      "eval_runtime": 15.1104,
      "eval_samples_per_second": 33.421,
      "eval_steps_per_second": 1.059,
      "step": 488
    }
  ],
  "logging_steps": 500,
  "max_steps": 1708,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1555686566280.0,
  "train_batch_size": 39,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6658046915260407,
    "learning_rate": 0.0001805988306273096,
    "num_train_epochs": 7,
    "per_device_train_batch_size": 39,
    "temperature": 15
  }
}