trainer_state.json

{
  "best_metric": 0.3237333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0961/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 12.327598571777344,
      "learning_rate": 0.0005,
      "loss": 3.4295,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1472,
      "eval_loss": 3.2095019817352295,
      "eval_runtime": 11.7933,
      "eval_samples_per_second": 317.977,
      "eval_steps_per_second": 5.003,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.269378662109375,
      "learning_rate": 0.0005,
      "loss": 2.9893,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2016,
      "eval_loss": 2.974724292755127,
      "eval_runtime": 11.7433,
      "eval_samples_per_second": 319.331,
      "eval_steps_per_second": 5.024,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 17.853349685668945,
      "learning_rate": 0.0005,
      "loss": 2.8356,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.24693333333333334,
      "eval_loss": 2.8405890464782715,
      "eval_runtime": 11.8049,
      "eval_samples_per_second": 317.664,
      "eval_steps_per_second": 4.998,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.851137161254883,
      "learning_rate": 0.0005,
      "loss": 2.7161,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2624,
      "eval_loss": 2.723008394241333,
      "eval_runtime": 11.8731,
      "eval_samples_per_second": 315.841,
      "eval_steps_per_second": 4.969,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 19.083127975463867,
      "learning_rate": 0.0005,
      "loss": 2.6503,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2650666666666667,
      "eval_loss": 2.7186875343322754,
      "eval_runtime": 11.7752,
      "eval_samples_per_second": 318.465,
      "eval_steps_per_second": 5.011,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 13.855463027954102,
      "learning_rate": 0.0005,
      "loss": 2.5577,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2776,
      "eval_loss": 2.6830828189849854,
      "eval_runtime": 11.9455,
      "eval_samples_per_second": 313.926,
      "eval_steps_per_second": 4.939,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 16.65631866455078,
      "learning_rate": 0.0005,
      "loss": 2.4811,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3237333333333333,
      "eval_loss": 2.4881324768066406,
      "eval_runtime": 12.4584,
      "eval_samples_per_second": 301.002,
      "eval_steps_per_second": 4.736,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}