run-47/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8297455968688845,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-47/checkpoint-192",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 10.090946197509766,
      "learning_rate": 0.0002726931435690813,
      "loss": 0.5743,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7964774951076321,
      "eval_f1": 0.8188153310104529,
      "eval_loss": 0.4561440944671631,
      "eval_precision": 0.7378335949764521,
      "eval_recall": 0.9197651663405088,
      "eval_runtime": 30.8585,
      "eval_samples_per_second": 33.119,
      "eval_steps_per_second": 1.037,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.258547306060791,
      "learning_rate": 0.00018179542904605417,
      "loss": 0.4449,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8297455968688845,
      "eval_f1": 0.8406593406593407,
      "eval_loss": 0.4051818251609802,
      "eval_precision": 0.7900172117039587,
      "eval_recall": 0.898238747553816,
      "eval_runtime": 31.0429,
      "eval_samples_per_second": 32.922,
      "eval_steps_per_second": 1.031,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 384,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.999132297568699,
    "learning_rate": 0.00036359085809210834,
    "num_train_epochs": 4,
    "temperature": 14
  }
}