run-3/checkpoint-192/trainer_state.json

{
  "best_metric": 0.576320939334638,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-192",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.41195374727249146,
      "learning_rate": 7.986864922441338e-05,
      "loss": 0.2944,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5547945205479452,
      "eval_f1": 0.2528735632183908,
      "eval_loss": 0.2588364779949188,
      "eval_precision": 0.7857142857142857,
      "eval_recall": 0.1506849315068493,
      "eval_runtime": 25.9352,
      "eval_samples_per_second": 39.406,
      "eval_steps_per_second": 1.234,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7536277174949646,
      "learning_rate": 6.38949193795307e-05,
      "loss": 0.261,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.576320939334638,
      "eval_f1": 0.3244929797191888,
      "eval_loss": 0.25162196159362793,
      "eval_precision": 0.8,
      "eval_recall": 0.2035225048923679,
      "eval_runtime": 25.5955,
      "eval_samples_per_second": 39.929,
      "eval_steps_per_second": 1.25,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.3513021160288001,
    "learning_rate": 9.584237906929605e-05,
    "num_train_epochs": 6,
    "temperature": 30
  }
}