run-11/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.8101761252446184,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-1532",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1532,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.275282621383667,
      "learning_rate": 0.00013397987335151483,
      "loss": 0.5034,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7886497064579256,
      "eval_f1": 0.8153846153846154,
      "eval_loss": 0.4397984445095062,
      "eval_precision": 0.723823975720789,
      "eval_recall": 0.9334637964774951,
      "eval_runtime": 133.4478,
      "eval_samples_per_second": 7.658,
      "eval_steps_per_second": 1.918,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.074667453765869,
      "learning_rate": 0.00011909322075690206,
      "loss": 0.442,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8101761252446184,
      "eval_f1": 0.831889081455806,
      "eval_loss": 0.43401414155960083,
      "eval_precision": 0.7465007776049767,
      "eval_recall": 0.9393346379647749,
      "eval_runtime": 133.9782,
      "eval_samples_per_second": 7.628,
      "eval_steps_per_second": 1.911,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 7660,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8157328229875264,
    "learning_rate": 0.00014886652594612757,
    "num_train_epochs": 10,
    "temperature": 20
  }
}