run-18/checkpoint-594/trainer_state.json

{
  "best_metric": 0.6237623762376238,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-18/checkpoint-594",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 594,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.6970789432525635,
      "learning_rate": 0.0001678982407816263,
      "loss": 0.6448,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5287128712871287,
      "eval_f1": 0.22222222222222224,
      "eval_loss": 0.6524360179901123,
      "eval_mcc": 0.09039579598072835,
      "eval_precision": 0.6296296296296297,
      "eval_recall": 0.1349206349206349,
      "eval_runtime": 0.9373,
      "eval_samples_per_second": 538.78,
      "eval_steps_per_second": 17.07,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.302747964859009,
      "learning_rate": 0.0001399152006513553,
      "loss": 0.6287,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6237623762376238,
      "eval_f1": 0.6415094339622641,
      "eval_loss": 0.6215968132019043,
      "eval_mcc": 0.24899823929472167,
      "eval_precision": 0.6115107913669064,
      "eval_recall": 0.6746031746031746,
      "eval_runtime": 0.9284,
      "eval_samples_per_second": 543.954,
      "eval_steps_per_second": 17.234,
      "step": 594
    }
  ],
  "logging_steps": 500,
  "max_steps": 2079,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1461402531960.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9135980547715112,
    "learning_rate": 0.00019588128091189738,
    "num_train_epochs": 7,
    "temperature": 19
  }
}