run-13/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7954990215264188,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/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": 2.201279878616333,
      "learning_rate": 0.00022445268667640978,
      "loss": 0.5435,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7700587084148728,
      "eval_f1": 0.7979363714531386,
      "eval_loss": 0.4716385006904602,
      "eval_precision": 0.7116564417177914,
      "eval_recall": 0.9080234833659491,
      "eval_runtime": 24.8269,
      "eval_samples_per_second": 41.165,
      "eval_steps_per_second": 1.289,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.4069817066192627,
      "learning_rate": 0.00018704390556367485,
      "loss": 0.4642,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7954990215264188,
      "eval_f1": 0.819047619047619,
      "eval_loss": 0.4382912516593933,
      "eval_precision": 0.734472049689441,
      "eval_recall": 0.9256360078277887,
      "eval_runtime": 25.9254,
      "eval_samples_per_second": 39.421,
      "eval_steps_per_second": 1.234,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 672,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8348768037192943,
    "learning_rate": 0.0002618614677891448,
    "num_train_epochs": 7,
    "temperature": 10
  }
}