run-46/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8140900195694716,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-46/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": 2.520153045654297,
      "learning_rate": 0.0004551449993833034,
      "loss": 0.4997,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7945205479452054,
      "eval_f1": 0.814487632508834,
      "eval_loss": 0.42794638872146606,
      "eval_precision": 0.7423510466988728,
      "eval_recall": 0.9021526418786693,
      "eval_runtime": 31.7011,
      "eval_samples_per_second": 32.239,
      "eval_steps_per_second": 1.009,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5412380695343018,
      "learning_rate": 0.00034135874953747754,
      "loss": 0.4292,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8140900195694716,
      "eval_f1": 0.822429906542056,
      "eval_loss": 0.4104781448841095,
      "eval_precision": 0.7871198568872988,
      "eval_recall": 0.8610567514677103,
      "eval_runtime": 31.9286,
      "eval_samples_per_second": 32.009,
      "eval_steps_per_second": 1.002,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7902045822653171,
    "learning_rate": 0.0005689312492291293,
    "num_train_epochs": 5,
    "temperature": 24
  }
}