checkpoint-80/trainer_state.json

{
  "best_metric": 0.6685735583305359,
  "best_model_checkpoint": "fewshotlearning-text-classification-bert-sm/checkpoint-80",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 16.075746536254883,
      "learning_rate": 0.00018055555555555557,
      "loss": 1.4192,
      "step": 16
    },
    {
      "epoch": 0.4,
      "grad_norm": 6.529771327972412,
      "learning_rate": 0.0001388888888888889,
      "loss": 1.2799,
      "step": 32
    },
    {
      "epoch": 0.6,
      "grad_norm": 27.337310791015625,
      "learning_rate": 9.444444444444444e-05,
      "loss": 0.9263,
      "step": 48
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.956445574760437,
      "learning_rate": 5.2777777777777784e-05,
      "loss": 0.4822,
      "step": 64
    },
    {
      "epoch": 1.0,
      "grad_norm": 8.860248565673828,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.6073,
      "step": 80
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8375,
      "eval_f1_macro": 0.8364860941080454,
      "eval_f1_micro": 0.8375,
      "eval_f1_weighted": 0.8364860941080453,
      "eval_loss": 0.6685735583305359,
      "eval_precision_macro": 0.8393652915711738,
      "eval_precision_micro": 0.8375,
      "eval_precision_weighted": 0.8393652915711739,
      "eval_recall_macro": 0.8375,
      "eval_recall_micro": 0.8375,
      "eval_recall_weighted": 0.8375,
      "eval_runtime": 1.1304,
      "eval_samples_per_second": 70.773,
      "eval_steps_per_second": 35.387,
      "step": 80
    }
  ],
  "logging_steps": 16,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 84197049630720.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}