checkpoint-6/trainer_state.json

{
  "best_metric": 0.42334219813346863,
  "best_model_checkpoint": "autotrain-btsw0-shiv/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": 9.514972686767578,
      "learning_rate": 5e-05,
      "loss": 0.6066,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 5.079541206359863,
      "learning_rate": 4e-05,
      "loss": 0.691,
      "step": 2
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.75,
      "eval_auc": 1.0,
      "eval_f1": 0.0,
      "eval_loss": 0.5450384616851807,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 0.8279,
      "eval_samples_per_second": 4.831,
      "eval_steps_per_second": 1.208,
      "step": 2
    },
    {
      "epoch": 1.5,
      "grad_norm": 5.415709495544434,
      "learning_rate": 3e-05,
      "loss": 0.3928,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.523742198944092,
      "learning_rate": 2e-05,
      "loss": 0.6082,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 1.0,
      "eval_auc": 1.0,
      "eval_f1": 1.0,
      "eval_loss": 0.4483739137649536,
      "eval_precision": 1.0,
      "eval_recall": 1.0,
      "eval_runtime": 0.9212,
      "eval_samples_per_second": 4.342,
      "eval_steps_per_second": 1.085,
      "step": 4
    },
    {
      "epoch": 2.5,
      "grad_norm": 4.789425849914551,
      "learning_rate": 1e-05,
      "loss": 0.4146,
      "step": 5
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.687239170074463,
      "learning_rate": 0.0,
      "loss": 0.3257,
      "step": 6
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 1.0,
      "eval_auc": 1.0,
      "eval_f1": 1.0,
      "eval_loss": 0.42334219813346863,
      "eval_precision": 1.0,
      "eval_recall": 1.0,
      "eval_runtime": 0.8303,
      "eval_samples_per_second": 4.818,
      "eval_steps_per_second": 1.204,
      "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
}