trainer_state.json

{
  "best_metric": 0.35333333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0953/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": 21.996366500854492,
      "learning_rate": 0.0003,
      "loss": 3.2608,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2288,
      "eval_loss": 2.893765926361084,
      "eval_runtime": 10.3229,
      "eval_samples_per_second": 363.269,
      "eval_steps_per_second": 5.715,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 27.689603805541992,
      "learning_rate": 0.0003,
      "loss": 2.7036,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.288,
      "eval_loss": 2.6487693786621094,
      "eval_runtime": 10.2991,
      "eval_samples_per_second": 364.108,
      "eval_steps_per_second": 5.729,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 28.576744079589844,
      "learning_rate": 0.0003,
      "loss": 2.471,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.31893333333333335,
      "eval_loss": 2.5089328289031982,
      "eval_runtime": 10.4885,
      "eval_samples_per_second": 357.534,
      "eval_steps_per_second": 5.625,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 23.544925689697266,
      "learning_rate": 0.0003,
      "loss": 2.2983,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3504,
      "eval_loss": 2.3831756114959717,
      "eval_runtime": 10.2708,
      "eval_samples_per_second": 365.113,
      "eval_steps_per_second": 5.744,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 41.993595123291016,
      "learning_rate": 0.0003,
      "loss": 2.1533,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.35333333333333333,
      "eval_loss": 2.343318462371826,
      "eval_runtime": 10.7609,
      "eval_samples_per_second": 348.483,
      "eval_steps_per_second": 5.483,
      "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
}