run-24/checkpoint-96/trainer_state.json

{
  "best_metric": 0.7475538160469667,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-24/checkpoint-48",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4524683952331543,
      "learning_rate": 0.00041148401868626845,
      "loss": 0.4958,
      "step": 48
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7475538160469667,
      "eval_f1": 0.7528735632183908,
      "eval_loss": 0.4411579370498657,
      "eval_precision": 0.7373358348968105,
      "eval_recall": 0.7690802348336595,
      "eval_runtime": 27.0601,
      "eval_samples_per_second": 37.768,
      "eval_steps_per_second": 0.591,
      "step": 48
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.1708617210388184,
      "learning_rate": 0.00020574200934313423,
      "loss": 0.45,
      "step": 96
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.713307240704501,
      "eval_f1": 0.6755260243632336,
      "eval_loss": 0.42811891436576843,
      "eval_precision": 0.7780612244897959,
      "eval_recall": 0.5968688845401174,
      "eval_runtime": 27.2226,
      "eval_samples_per_second": 37.542,
      "eval_steps_per_second": 0.588,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 144,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.720782515325305,
    "learning_rate": 0.0006172260280294027,
    "num_train_epochs": 3,
    "temperature": 25
  }
}