trainer_state.json

{
  "best_metric": 0.9277333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0052/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": 9.347606658935547,
      "learning_rate": 2.8858192987669303e-05,
      "loss": 0.7196,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.416810542345047,
      "eval_runtime": 12.6266,
      "eval_samples_per_second": 296.992,
      "eval_steps_per_second": 4.673,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.5619102456015873e-05,
      "loss": 0.1655,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.908,
      "eval_loss": 0.32657864689826965,
      "eval_runtime": 12.6736,
      "eval_samples_per_second": 295.89,
      "eval_steps_per_second": 4.655,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.127151608467102,
      "learning_rate": 2.0756590131847176e-05,
      "loss": 0.0657,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.33648672699928284,
      "eval_runtime": 12.4748,
      "eval_samples_per_second": 300.606,
      "eval_steps_per_second": 4.73,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.015931393951177597,
      "learning_rate": 1.5017689143724237e-05,
      "loss": 0.0238,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9154666666666667,
      "eval_loss": 0.33380377292633057,
      "eval_runtime": 11.9121,
      "eval_samples_per_second": 314.806,
      "eval_steps_per_second": 4.953,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.056108977645635605,
      "learning_rate": 9.276095143821772e-06,
      "loss": 0.0075,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9216,
      "eval_loss": 0.32990899682044983,
      "eval_runtime": 12.8817,
      "eval_samples_per_second": 291.112,
      "eval_steps_per_second": 4.58,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.02796177752315998,
      "learning_rate": 4.4059137709457125e-06,
      "loss": 0.0021,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9253333333333333,
      "eval_loss": 0.3430544137954712,
      "eval_runtime": 12.4066,
      "eval_samples_per_second": 302.259,
      "eval_steps_per_second": 4.756,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.05098952353000641,
      "learning_rate": 1.1485859908123557e-06,
      "loss": 0.0004,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9274666666666667,
      "eval_loss": 0.33246368169784546,
      "eval_runtime": 12.2672,
      "eval_samples_per_second": 305.694,
      "eval_steps_per_second": 4.81,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.00360115640796721,
      "learning_rate": 1.04301971493026e-11,
      "loss": 0.0006,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9277333333333333,
      "eval_loss": 0.3394603729248047,
      "eval_runtime": 12.4939,
      "eval_samples_per_second": 300.147,
      "eval_steps_per_second": 4.722,
      "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
}