trainer_state.json

{
  "best_metric": 0.9189333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0463/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 27.906625747680664,
      "learning_rate": 6.339302707314361e-05,
      "loss": 0.8584,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8328,
      "eval_loss": 0.5440648198127747,
      "eval_runtime": 10.9165,
      "eval_samples_per_second": 343.518,
      "eval_steps_per_second": 5.405,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.002553798258304596,
      "learning_rate": 4.594113205975266e-05,
      "loss": 0.2612,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.4035874903202057,
      "eval_runtime": 11.3694,
      "eval_samples_per_second": 329.832,
      "eval_steps_per_second": 5.189,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 18.70603370666504,
      "learning_rate": 2.431009647493735e-05,
      "loss": 0.088,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.3657597005367279,
      "eval_runtime": 11.0448,
      "eval_samples_per_second": 339.527,
      "eval_steps_per_second": 5.342,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0027960562147200108,
      "learning_rate": 6.762240700238659e-06,
      "loss": 0.0188,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.3109656572341919,
      "eval_runtime": 11.2835,
      "eval_samples_per_second": 332.345,
      "eval_steps_per_second": 5.229,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3466358482837677,
      "learning_rate": 2.492119103614243e-10,
      "loss": 0.0031,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9189333333333334,
      "eval_loss": 0.3117009699344635,
      "eval_runtime": 10.7841,
      "eval_samples_per_second": 347.733,
      "eval_steps_per_second": 5.471,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}