trainer_state.json

{
  "best_metric": 0.8509333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0147/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": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.8855,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8245333333333333,
      "eval_loss": 0.5534557104110718,
      "eval_runtime": 12.9172,
      "eval_samples_per_second": 290.311,
      "eval_steps_per_second": 4.568,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.05761552229523659,
      "learning_rate": 7e-05,
      "loss": 0.3557,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5659441947937012,
      "eval_runtime": 12.4646,
      "eval_samples_per_second": 300.853,
      "eval_steps_per_second": 4.733,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.466134548187256,
      "learning_rate": 7e-05,
      "loss": 0.2358,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.5196097493171692,
      "eval_runtime": 12.3667,
      "eval_samples_per_second": 303.235,
      "eval_steps_per_second": 4.771,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6817805767059326,
      "learning_rate": 7e-05,
      "loss": 0.1711,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.5929004549980164,
      "eval_runtime": 12.4791,
      "eval_samples_per_second": 300.502,
      "eval_steps_per_second": 4.728,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0036980004515498877,
      "learning_rate": 7e-05,
      "loss": 0.1368,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8317333333333333,
      "eval_loss": 0.6661263108253479,
      "eval_runtime": 12.3681,
      "eval_samples_per_second": 303.199,
      "eval_steps_per_second": 4.77,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.8218415975570679,
      "learning_rate": 7e-05,
      "loss": 0.1169,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.5936015844345093,
      "eval_runtime": 12.2502,
      "eval_samples_per_second": 306.116,
      "eval_steps_per_second": 4.816,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}