checkpoint-6/trainer_state.json

{
  "best_metric": 0.5295487642288208,
  "best_model_checkpoint": "autotrain-s2tfh-shivj5l/checkpoint-6",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "grad_norm": 7.717057704925537,
      "learning_rate": 5e-05,
      "loss": 0.5474,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.8525986671447754,
      "learning_rate": 4e-05,
      "loss": 0.6347,
      "step": 2
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.75,
      "eval_auc": 0.33333333333333337,
      "eval_f1": 0.0,
      "eval_loss": 0.5581548810005188,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 0.7982,
      "eval_samples_per_second": 5.011,
      "eval_steps_per_second": 1.253,
      "step": 2
    },
    {
      "epoch": 1.5,
      "grad_norm": 4.873136043548584,
      "learning_rate": 3e-05,
      "loss": 0.4156,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.227407455444336,
      "learning_rate": 2e-05,
      "loss": 0.6143,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.75,
      "eval_auc": 1.0,
      "eval_f1": 0.0,
      "eval_loss": 0.5400429964065552,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 0.9043,
      "eval_samples_per_second": 4.423,
      "eval_steps_per_second": 1.106,
      "step": 4
    },
    {
      "epoch": 2.5,
      "grad_norm": 4.113026142120361,
      "learning_rate": 1e-05,
      "loss": 0.4983,
      "step": 5
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.6406941413879395,
      "learning_rate": 0.0,
      "loss": 0.4676,
      "step": 6
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.75,
      "eval_auc": 1.0,
      "eval_f1": 0.0,
      "eval_loss": 0.5295487642288208,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 0.8354,
      "eval_samples_per_second": 4.788,
      "eval_steps_per_second": 1.197,
      "step": 6
    }
  ],
  "logging_steps": 1,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3157332664320.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}