run-28/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8003913894324853,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-28/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": 2.726130962371826,
      "learning_rate": 0.0002525240857743281,
      "loss": 0.5366,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7945205479452054,
      "eval_f1": 0.8104693140794224,
      "eval_loss": 0.4465394616127014,
      "eval_precision": 0.7520938023450586,
      "eval_recall": 0.8786692759295499,
      "eval_runtime": 30.8786,
      "eval_samples_per_second": 33.097,
      "eval_steps_per_second": 1.036,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.9313712120056152,
      "learning_rate": 0.00020201926861946243,
      "loss": 0.4543,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8003913894324853,
      "eval_f1": 0.8226086956521739,
      "eval_loss": 0.43098214268684387,
      "eval_precision": 0.7402190923317684,
      "eval_recall": 0.9256360078277887,
      "eval_runtime": 31.4874,
      "eval_samples_per_second": 32.457,
      "eval_steps_per_second": 1.016,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.83527502791975,
    "learning_rate": 0.00030302890292919365,
    "num_train_epochs": 6,
    "temperature": 28
  }
}