run-7/checkpoint-428/trainer_state.json

{
  "best_metric": 0.6666666666666666,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-7/checkpoint-214",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 428,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.5830457210540771,
      "learning_rate": 6.66727069326197e-06,
      "loss": 0.3711,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.3332321345806122,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 3.1298,
      "eval_samples_per_second": 545.4,
      "eval_steps_per_second": 17.253,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6807608604431152,
      "learning_rate": 3.333635346630985e-06,
      "loss": 0.3331,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.3313482403755188,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 3.1532,
      "eval_samples_per_second": 541.354,
      "eval_steps_per_second": 17.125,
      "step": 428
    }
  ],
  "logging_steps": 500,
  "max_steps": 642,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1049551328880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.4959780812752268,
    "learning_rate": 1.0000906039892956e-05,
    "num_train_epochs": 3,
    "temperature": 13
  }
}