trainer_state.json

{
  "best_metric": 0.9325333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0727/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 4.811948852959657e-05,
      "loss": 0.7532,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.3974095582962036,
      "eval_runtime": 10.3866,
      "eval_samples_per_second": 361.043,
      "eval_steps_per_second": 5.68,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.519961416721344,
      "learning_rate": 4.271931404387096e-05,
      "loss": 0.1867,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8968,
      "eval_loss": 0.34589746594429016,
      "eval_runtime": 10.5519,
      "eval_samples_per_second": 355.385,
      "eval_steps_per_second": 5.591,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.134028911590576,
      "learning_rate": 3.4621534620947785e-05,
      "loss": 0.0828,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9042666666666667,
      "eval_loss": 0.366612046957016,
      "eval_runtime": 11.4424,
      "eval_samples_per_second": 327.727,
      "eval_steps_per_second": 5.156,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.319874882698059,
      "learning_rate": 2.5058963771413844e-05,
      "loss": 0.0382,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9138666666666667,
      "eval_loss": 0.3377947211265564,
      "eval_runtime": 10.7412,
      "eval_samples_per_second": 349.124,
      "eval_steps_per_second": 5.493,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.044662073254585266,
      "learning_rate": 1.5487416222190068e-05,
      "loss": 0.0105,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.3191099464893341,
      "eval_runtime": 10.7584,
      "eval_samples_per_second": 348.565,
      "eval_steps_per_second": 5.484,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.006848808377981186,
      "learning_rate": 7.36407332135117e-06,
      "loss": 0.0023,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9314666666666667,
      "eval_loss": 0.29536452889442444,
      "eval_runtime": 11.2357,
      "eval_samples_per_second": 333.759,
      "eval_steps_per_second": 5.251,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.015924174338579178,
      "learning_rate": 1.925640387263111e-06,
      "loss": 0.0008,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.932,
      "eval_loss": 0.2975226938724518,
      "eval_runtime": 12.5497,
      "eval_samples_per_second": 298.811,
      "eval_steps_per_second": 4.701,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.000330587470671162,
      "learning_rate": 6.953462348691098e-11,
      "loss": 0.0003,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9325333333333333,
      "eval_loss": 0.2975701093673706,
      "eval_runtime": 11.227,
      "eval_samples_per_second": 334.015,
      "eval_steps_per_second": 5.255,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}