trainer_state.json

{
  "best_metric": 0.04190065711736679,
  "best_model_checkpoint": "ai_vs_real_image_detection/checkpoint-2252",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 2252,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.44404973357015987,
      "grad_norm": 1.6082425117492676,
      "learning_rate": 7.956403269754768e-07,
      "loss": 0.0398,
      "step": 500
    },
    {
      "epoch": 0.8880994671403197,
      "grad_norm": 4.518634796142578,
      "learning_rate": 5.685740236148955e-07,
      "loss": 0.038,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9844924127307925,
      "eval_loss": 0.043014705181121826,
      "eval_runtime": 698.9147,
      "eval_samples_per_second": 68.737,
      "eval_steps_per_second": 2.149,
      "step": 1126
    },
    {
      "epoch": 1.3321492007104796,
      "grad_norm": 7.5788960456848145,
      "learning_rate": 3.415077202543142e-07,
      "loss": 0.0364,
      "step": 1500
    },
    {
      "epoch": 1.7761989342806395,
      "grad_norm": 2.462517499923706,
      "learning_rate": 1.1444141689373296e-07,
      "loss": 0.0375,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9848670926916592,
      "eval_loss": 0.04190065711736679,
      "eval_runtime": 790.7788,
      "eval_samples_per_second": 60.752,
      "eval_steps_per_second": 1.899,
      "step": 2252
    }
  ],
  "logging_steps": 500,
  "max_steps": 2252,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1.1168300527234228e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}