run-17/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.8043052837573386,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-17/checkpoint-1532",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1532,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.036878824234009,
      "learning_rate": 6.985166134218099e-05,
      "loss": 0.5738,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7876712328767124,
      "eval_f1": 0.8121212121212121,
      "eval_loss": 0.4637316167354584,
      "eval_precision": 0.7282608695652174,
      "eval_recall": 0.9178082191780822,
      "eval_runtime": 132.0721,
      "eval_samples_per_second": 7.738,
      "eval_steps_per_second": 1.938,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 31.187244415283203,
      "learning_rate": 5.8209717785150834e-05,
      "loss": 0.5056,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8043052837573386,
      "eval_f1": 0.82174688057041,
      "eval_loss": 0.4566289782524109,
      "eval_precision": 0.7545008183306056,
      "eval_recall": 0.9021526418786693,
      "eval_runtime": 132.5437,
      "eval_samples_per_second": 7.711,
      "eval_steps_per_second": 1.931,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 5362,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9998794266799103,
    "learning_rate": 8.149360489921117e-05,
    "num_train_epochs": 7,
    "temperature": 5
  }
}