trainer_state.json

{
  "best_metric": 0.9101333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0475/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 32.90766906738281,
      "learning_rate": 6.533836810383375e-05,
      "loss": 0.8905,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8344,
      "eval_loss": 0.5529220104217529,
      "eval_runtime": 8.1257,
      "eval_samples_per_second": 461.5,
      "eval_steps_per_second": 7.261,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 44.14369201660156,
      "learning_rate": 5.2547638240566195e-05,
      "loss": 0.2989,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.48450255393981934,
      "eval_runtime": 7.2938,
      "eval_samples_per_second": 514.138,
      "eval_steps_per_second": 8.089,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 3.5110065627270776e-05,
      "loss": 0.1344,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.4178406298160553,
      "eval_runtime": 7.852,
      "eval_samples_per_second": 477.585,
      "eval_steps_per_second": 7.514,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 32.85700225830078,
      "learning_rate": 1.7595406161244655e-05,
      "loss": 0.0481,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.3833388090133667,
      "eval_runtime": 7.8079,
      "eval_samples_per_second": 480.282,
      "eval_steps_per_second": 7.556,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12988315522670746,
      "learning_rate": 4.744293558904729e-06,
      "loss": 0.0132,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.33268773555755615,
      "eval_runtime": 7.3623,
      "eval_samples_per_second": 509.351,
      "eval_steps_per_second": 8.014,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.215996742248535,
      "learning_rate": 1.7306388939264393e-10,
      "loss": 0.0029,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9101333333333333,
      "eval_loss": 0.3373066186904907,
      "eval_runtime": 7.4717,
      "eval_samples_per_second": 501.897,
      "eval_steps_per_second": 7.897,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}