run-2/checkpoint-3176/trainer_state.json

{
  "best_metric": 0.6666929009916575,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-1588",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 3176,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.5482791066169739,
      "learning_rate": 9.642076934385773e-06,
      "loss": 0.5031,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666929009916575,
      "eval_f1": 0.0,
      "eval_loss": 0.49551922082901,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 23.2895,
      "eval_samples_per_second": 545.567,
      "eval_steps_per_second": 17.089,
      "step": 1588
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0544217824935913,
      "learning_rate": 8.436817317587551e-06,
      "loss": 0.4941,
      "step": 3176
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6666929009916575,
      "eval_f1": 0.0,
      "eval_loss": 0.48784464597702026,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 23.5003,
      "eval_samples_per_second": 540.673,
      "eval_steps_per_second": 16.936,
      "step": 3176
    }
  ],
  "logging_steps": 500,
  "max_steps": 14292,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 7817788152600.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7681424055105933,
    "learning_rate": 1.0847336551183995e-05,
    "num_train_epochs": 9,
    "temperature": 16
  }
}