trainer_state.json

{
  "best_metric": 0.46613333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0725/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": 3.2649,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2272,
      "eval_loss": 2.8628134727478027,
      "eval_runtime": 12.6142,
      "eval_samples_per_second": 297.284,
      "eval_steps_per_second": 4.677,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 26.525508880615234,
      "learning_rate": 0.0002564405961135417,
      "loss": 2.6683,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2592,
      "eval_loss": 2.6931562423706055,
      "eval_runtime": 12.5775,
      "eval_samples_per_second": 298.152,
      "eval_steps_per_second": 4.691,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.365442276000977,
      "learning_rate": 0.0002078924338492993,
      "loss": 2.4237,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3416,
      "eval_loss": 2.387843132019043,
      "eval_runtime": 12.7179,
      "eval_samples_per_second": 294.86,
      "eval_steps_per_second": 4.639,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 38.63453674316406,
      "learning_rate": 0.00015053067332772077,
      "loss": 2.1737,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3834666666666667,
      "eval_loss": 2.2216920852661133,
      "eval_runtime": 13.6629,
      "eval_samples_per_second": 274.467,
      "eval_steps_per_second": 4.318,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 31.41911506652832,
      "learning_rate": 9.308812260256249e-05,
      "loss": 1.917,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.41946666666666665,
      "eval_loss": 2.0784621238708496,
      "eval_runtime": 12.6027,
      "eval_samples_per_second": 297.554,
      "eval_steps_per_second": 4.682,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 39.556217193603516,
      "learning_rate": 4.4309889303710674e-05,
      "loss": 1.6092,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.448,
      "eval_loss": 2.028052806854248,
      "eval_runtime": 12.365,
      "eval_samples_per_second": 303.275,
      "eval_steps_per_second": 4.772,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 34.94908142089844,
      "learning_rate": 1.1622017275129708e-05,
      "loss": 1.2718,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.456,
      "eval_loss": 2.098249912261963,
      "eval_runtime": 12.3199,
      "eval_samples_per_second": 304.385,
      "eval_steps_per_second": 4.789,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 1.668828642892306e-09,
      "loss": 1.0308,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.46613333333333334,
      "eval_loss": 2.2097952365875244,
      "eval_runtime": 12.5121,
      "eval_samples_per_second": 299.71,
      "eval_steps_per_second": 4.715,
      "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
}