run-18/checkpoint-96/trainer_state.json

{
  "best_metric": 0.7367906066536204,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-18/checkpoint-96",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9677866101264954,
      "learning_rate": 0.00013148115776202284,
      "loss": 0.4705,
      "step": 48
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7191780821917808,
      "eval_f1": 0.710393541876892,
      "eval_loss": 0.4192996025085449,
      "eval_precision": 0.7333333333333333,
      "eval_recall": 0.6888454011741683,
      "eval_runtime": 27.0047,
      "eval_samples_per_second": 37.845,
      "eval_steps_per_second": 0.592,
      "step": 48
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1390928030014038,
      "learning_rate": 0.0,
      "loss": 0.4188,
      "step": 96
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7367906066536204,
      "eval_f1": 0.7263479145473042,
      "eval_loss": 0.4030902087688446,
      "eval_precision": 0.7563559322033898,
      "eval_recall": 0.6986301369863014,
      "eval_runtime": 27.7091,
      "eval_samples_per_second": 36.883,
      "eval_steps_per_second": 0.577,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 96,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.650154951792798,
    "learning_rate": 0.0002629623155240457,
    "num_train_epochs": 2,
    "temperature": 21
  }
}