run-3/checkpoint-320/trainer_state.json

{
  "best_metric": 0.6964705882352941,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-320",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3765578269958496,
      "learning_rate": 0.0003184457865836907,
      "loss": 0.3598,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.3431719243526459,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.88,
      "eval_samples_per_second": 678.204,
      "eval_steps_per_second": 21.277,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.66363924741745,
      "learning_rate": 0.0002653714888197423,
      "loss": 0.3406,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6964705882352941,
      "eval_f1": 0.2275449101796407,
      "eval_loss": 0.3270348310470581,
      "eval_mcc": 0.22253187118870327,
      "eval_precision": 0.75,
      "eval_recall": 0.13411764705882354,
      "eval_runtime": 1.9253,
      "eval_samples_per_second": 662.218,
      "eval_steps_per_second": 20.775,
      "step": 320
    }
  ],
  "logging_steps": 500,
  "max_steps": 1120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 583510875840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.543790056219147,
    "learning_rate": 0.0003715200843476392,
    "num_train_epochs": 7,
    "temperature": 25
  }
}