trainer_state.json

{
  "best_metric": 0.44693333333333335,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0670/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": 0.0002887841425059038,
      "loss": 3.3868,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2216,
      "eval_loss": 2.986675500869751,
      "eval_runtime": 12.7448,
      "eval_samples_per_second": 294.239,
      "eval_steps_per_second": 4.629,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.280323028564453,
      "learning_rate": 0.0002564405961135417,
      "loss": 2.8745,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.26106666666666667,
      "eval_loss": 2.7964181900024414,
      "eval_runtime": 12.1944,
      "eval_samples_per_second": 307.518,
      "eval_steps_per_second": 4.838,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 51.709774017333984,
      "learning_rate": 0.0002078924338492993,
      "loss": 2.6118,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.31706666666666666,
      "eval_loss": 2.5726070404052734,
      "eval_runtime": 12.9406,
      "eval_samples_per_second": 289.786,
      "eval_steps_per_second": 4.559,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 19.69470977783203,
      "learning_rate": 0.00015053067332772077,
      "loss": 2.3616,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3528,
      "eval_loss": 2.411731719970703,
      "eval_runtime": 12.978,
      "eval_samples_per_second": 288.952,
      "eval_steps_per_second": 4.546,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 27.158668518066406,
      "learning_rate": 9.308812260256249e-05,
      "loss": 2.1266,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.40453333333333336,
      "eval_loss": 2.214052200317383,
      "eval_runtime": 12.6174,
      "eval_samples_per_second": 297.209,
      "eval_steps_per_second": 4.676,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 24.592802047729492,
      "learning_rate": 4.4309889303710674e-05,
      "loss": 1.8554,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4176,
      "eval_loss": 2.1858081817626953,
      "eval_runtime": 12.6657,
      "eval_samples_per_second": 296.076,
      "eval_steps_per_second": 4.658,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 48.11631393432617,
      "learning_rate": 1.1622017275129708e-05,
      "loss": 1.6055,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4437333333333333,
      "eval_loss": 2.1200766563415527,
      "eval_runtime": 12.3519,
      "eval_samples_per_second": 303.597,
      "eval_steps_per_second": 4.777,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 1.668828642892306e-09,
      "loss": 1.4344,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.44693333333333335,
      "eval_loss": 2.1599550247192383,
      "eval_runtime": 13.0881,
      "eval_samples_per_second": 286.519,
      "eval_steps_per_second": 4.508,
      "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
}