trainer_state.json

{
  "best_metric": 0.8488,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0853/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": 2.610124111175537,
      "learning_rate": 9e-05,
      "loss": 0.947,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8178666666666666,
      "eval_loss": 0.5968145728111267,
      "eval_runtime": 12.3982,
      "eval_samples_per_second": 302.464,
      "eval_steps_per_second": 4.759,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.1860635131597519,
      "learning_rate": 9e-05,
      "loss": 0.382,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8448,
      "eval_loss": 0.5335744023323059,
      "eval_runtime": 12.1712,
      "eval_samples_per_second": 308.104,
      "eval_steps_per_second": 4.848,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1801595687866211,
      "learning_rate": 9e-05,
      "loss": 0.2554,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8261333333333334,
      "eval_loss": 0.6303421854972839,
      "eval_runtime": 12.2751,
      "eval_samples_per_second": 305.495,
      "eval_steps_per_second": 4.806,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 111.27642059326172,
      "learning_rate": 9e-05,
      "loss": 0.1948,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5986440181732178,
      "eval_runtime": 12.7823,
      "eval_samples_per_second": 293.375,
      "eval_steps_per_second": 4.616,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.1567,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5925394296646118,
      "eval_runtime": 12.2386,
      "eval_samples_per_second": 306.407,
      "eval_steps_per_second": 4.821,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}