trainer_state.json

{
  "best_metric": 0.9381333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0681/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": 1.9926161766052246,
      "learning_rate": 2.886495269365963e-05,
      "loss": 0.7544,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.39128732681274414,
      "eval_runtime": 12.366,
      "eval_samples_per_second": 303.25,
      "eval_steps_per_second": 4.771,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.5631588426322576e-05,
      "loss": 0.1918,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9064,
      "eval_loss": 0.27934807538986206,
      "eval_runtime": 12.2852,
      "eval_samples_per_second": 305.246,
      "eval_steps_per_second": 4.803,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0446906089782715,
      "learning_rate": 2.0772920772568672e-05,
      "loss": 0.0773,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9101333333333333,
      "eval_loss": 0.2878895699977875,
      "eval_runtime": 12.4789,
      "eval_samples_per_second": 300.507,
      "eval_steps_per_second": 4.728,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1476407051086426,
      "learning_rate": 1.5035378262848308e-05,
      "loss": 0.0322,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.2890504002571106,
      "eval_runtime": 12.4946,
      "eval_samples_per_second": 300.128,
      "eval_steps_per_second": 4.722,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.007044498343020678,
      "learning_rate": 9.292449733314041e-06,
      "loss": 0.0102,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9256,
      "eval_loss": 0.3004116415977478,
      "eval_runtime": 12.0843,
      "eval_samples_per_second": 310.319,
      "eval_steps_per_second": 4.882,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0016156742349267006,
      "learning_rate": 4.418443992810701e-06,
      "loss": 0.0023,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9352,
      "eval_loss": 0.2869779169559479,
      "eval_runtime": 12.2719,
      "eval_samples_per_second": 305.575,
      "eval_steps_per_second": 4.808,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.009068083949387074,
      "learning_rate": 1.1553842323578667e-06,
      "loss": 0.0003,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9354666666666667,
      "eval_loss": 0.2906186282634735,
      "eval_runtime": 12.3735,
      "eval_samples_per_second": 303.066,
      "eval_steps_per_second": 4.768,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0014848442515358329,
      "learning_rate": 4.172077409214658e-11,
      "loss": 0.0003,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9381333333333334,
      "eval_loss": 0.2903316617012024,
      "eval_runtime": 12.7682,
      "eval_samples_per_second": 293.699,
      "eval_steps_per_second": 4.621,
      "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
}