trainer_state.json

{
  "best_metric": 0.9389333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0034/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 50.57843780517578,
      "learning_rate": 2.7151869176294504e-05,
      "loss": 0.8138,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.333018034696579,
      "eval_runtime": 13.3364,
      "eval_samples_per_second": 281.185,
      "eval_steps_per_second": 4.424,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 1.968905659703686e-05,
      "loss": 0.193,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9096,
      "eval_loss": 0.27991795539855957,
      "eval_runtime": 13.1731,
      "eval_samples_per_second": 284.67,
      "eval_steps_per_second": 4.479,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 184.58316040039062,
      "learning_rate": 1.041861277497315e-05,
      "loss": 0.0692,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9210666666666667,
      "eval_loss": 0.2635088562965393,
      "eval_runtime": 13.2098,
      "eval_samples_per_second": 283.88,
      "eval_steps_per_second": 4.466,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5023165345191956,
      "learning_rate": 2.89810315724514e-06,
      "loss": 0.019,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9333333333333333,
      "eval_loss": 0.227239191532135,
      "eval_runtime": 13.4298,
      "eval_samples_per_second": 279.229,
      "eval_steps_per_second": 4.393,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.010868217796087265,
      "learning_rate": 1.0680510444061043e-10,
      "loss": 0.0057,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.22196774184703827,
      "eval_runtime": 13.0227,
      "eval_samples_per_second": 287.959,
      "eval_steps_per_second": 4.531,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}