run-0/checkpoint-384/trainer_state.json

{
  "best_metric": 0.7181996086105675,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-384",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.7846055030822754,
      "learning_rate": 4.610177667546352e-05,
      "loss": 0.4587,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6976516634050881,
      "eval_f1": 0.6419466975666281,
      "eval_loss": 0.4008609354496002,
      "eval_mcc": 0.41595144404027,
      "eval_precision": 0.7869318181818182,
      "eval_recall": 0.5420743639921722,
      "eval_runtime": 66.5688,
      "eval_samples_per_second": 15.353,
      "eval_steps_per_second": 0.481,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.001947402954102,
      "learning_rate": 4.0979357044856464e-05,
      "loss": 0.4016,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7181996086105675,
      "eval_f1": 0.6778523489932886,
      "eval_loss": 0.3820359408855438,
      "eval_mcc": 0.4507700437564526,
      "eval_precision": 0.7911227154046997,
      "eval_recall": 0.5929549902152642,
      "eval_runtime": 66.6005,
      "eval_samples_per_second": 15.345,
      "eval_steps_per_second": 0.48,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 1920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6226480553446316,
    "learning_rate": 5.122419630607058e-05,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 16,
    "temperature": 20
  }
}