checkpoint-1756/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 1756,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28,
      "learning_rate": 1.810174639331815e-05,
      "loss": 0.2574,
      "step": 500
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6203492786636296e-05,
      "loss": 0.1076,
      "step": 1000
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4305239179954442e-05,
      "loss": 0.0794,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9808382881026667,
      "eval_f1": 0.9228973910144204,
      "eval_loss": 0.07026076316833496,
      "eval_precision": 0.9142857142857143,
      "eval_recall": 0.9316728374284753,
      "eval_runtime": 490.5826,
      "eval_samples_per_second": 6.625,
      "eval_steps_per_second": 0.83,
      "step": 1756
    }
  ],
  "max_steps": 5268,
  "num_train_epochs": 3,
  "total_flos": 307479492426330.0,
  "trial_name": null,
  "trial_params": null
}