run-5/checkpoint-396/trainer_state.json

{
  "best_metric": 0.5108910891089109,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-5/checkpoint-198",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 396,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.14785648882389069,
      "learning_rate": 0.0003571920745148284,
      "loss": 0.0528,
      "step": 198
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5108910891089109,
      "eval_f1": 0.13333333333333333,
      "eval_loss": 0.04301260784268379,
      "eval_precision": 0.5757575757575758,
      "eval_recall": 0.07539682539682539,
      "eval_runtime": 15.1939,
      "eval_samples_per_second": 33.237,
      "eval_steps_per_second": 1.053,
      "step": 198
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.11046522110700607,
      "learning_rate": 0.00031750406623540304,
      "loss": 0.0443,
      "step": 396
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.502970297029703,
      "eval_f1": 0.11307420494699646,
      "eval_loss": 0.04213270545005798,
      "eval_precision": 0.5161290322580645,
      "eval_recall": 0.06349206349206349,
      "eval_runtime": 14.8075,
      "eval_samples_per_second": 34.104,
      "eval_steps_per_second": 1.081,
      "step": 396
    }
  ],
  "logging_steps": 500,
  "max_steps": 1980,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1555686566280.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.04173399039600034,
    "learning_rate": 0.0003968800827942538,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 48,
    "temperature": 11
  }
}