checkpoint-574/trainer_state.json

{
  "best_global_step": 574,
  "best_metric": 0.27341511845588684,
  "best_model_checkpoint": "siglip2-finetune-full/checkpoint-574",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 574,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7395287958115183,
      "eval_loss": 0.863375186920166,
      "eval_model_preparation_time": 0.0042,
      "eval_runtime": 116.2577,
      "eval_samples_per_second": 78.859,
      "eval_steps_per_second": 9.857,
      "step": 287
    },
    {
      "epoch": 1.7421602787456445,
      "grad_norm": 11.427348136901855,
      "learning_rate": 2.824427480916031e-05,
      "loss": 1.2211,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.918630017452007,
      "eval_loss": 0.27341511845588684,
      "eval_model_preparation_time": 0.0042,
      "eval_runtime": 116.3205,
      "eval_samples_per_second": 78.817,
      "eval_steps_per_second": 9.852,
      "step": 574
    }
  ],
  "logging_steps": 500,
  "max_steps": 574,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5360336767387566e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}