trainer_state.json

{
  "best_metric": 0.9381333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0080/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": Infinity,
      "learning_rate": 6.339302707314361e-05,
      "loss": 0.7404,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.4707658290863037,
      "eval_runtime": 4.4493,
      "eval_samples_per_second": 842.827,
      "eval_steps_per_second": 13.26,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4985698461532593,
      "learning_rate": 4.594113205975266e-05,
      "loss": 0.2182,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.3781861960887909,
      "eval_runtime": 4.6295,
      "eval_samples_per_second": 810.02,
      "eval_steps_per_second": 12.744,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.216766357421875,
      "learning_rate": 2.431009647493735e-05,
      "loss": 0.0868,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9125333333333333,
      "eval_loss": 0.3260475993156433,
      "eval_runtime": 4.4146,
      "eval_samples_per_second": 849.459,
      "eval_steps_per_second": 13.365,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.003142033237963915,
      "learning_rate": 6.762240700238659e-06,
      "loss": 0.0215,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9317333333333333,
      "eval_loss": 0.2616502642631531,
      "eval_runtime": 4.5744,
      "eval_samples_per_second": 819.788,
      "eval_steps_per_second": 12.898,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0018753724871203303,
      "learning_rate": 2.492119103614243e-10,
      "loss": 0.0041,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9381333333333334,
      "eval_loss": 0.2400323897600174,
      "eval_runtime": 4.4629,
      "eval_samples_per_second": 840.26,
      "eval_steps_per_second": 13.22,
      "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
}