trainer_state.json

{
  "best_metric": 0.2981333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0841/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 18.151052474975586,
      "learning_rate": 0.0005,
      "loss": 3.5112,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.15706666666666666,
      "eval_loss": 3.146672487258911,
      "eval_runtime": 12.4172,
      "eval_samples_per_second": 302.001,
      "eval_steps_per_second": 4.751,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.361034393310547,
      "learning_rate": 0.0005,
      "loss": 3.0706,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.18666666666666668,
      "eval_loss": 3.0181710720062256,
      "eval_runtime": 12.3662,
      "eval_samples_per_second": 303.245,
      "eval_steps_per_second": 4.771,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 15.263158798217773,
      "learning_rate": 0.0005,
      "loss": 2.9139,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.22853333333333334,
      "eval_loss": 2.8737874031066895,
      "eval_runtime": 12.4019,
      "eval_samples_per_second": 302.374,
      "eval_steps_per_second": 4.757,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 18.881526947021484,
      "learning_rate": 0.0005,
      "loss": 2.8088,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.24586666666666668,
      "eval_loss": 2.773937463760376,
      "eval_runtime": 12.2405,
      "eval_samples_per_second": 306.361,
      "eval_steps_per_second": 4.82,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.51772689819336,
      "learning_rate": 0.0005,
      "loss": 2.7243,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.25706666666666667,
      "eval_loss": 2.744156837463379,
      "eval_runtime": 12.3844,
      "eval_samples_per_second": 302.8,
      "eval_steps_per_second": 4.764,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 10.719879150390625,
      "learning_rate": 0.0005,
      "loss": 2.6387,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2618666666666667,
      "eval_loss": 2.7117373943328857,
      "eval_runtime": 12.1838,
      "eval_samples_per_second": 307.786,
      "eval_steps_per_second": 4.843,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.5534,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.2981333333333333,
      "eval_loss": 2.5588371753692627,
      "eval_runtime": 12.4693,
      "eval_samples_per_second": 300.738,
      "eval_steps_per_second": 4.732,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}