trainer_state.json

{
  "best_metric": 0.4146666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0871/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 0.0003,
      "loss": 3.334,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2312,
      "eval_loss": 2.8978965282440186,
      "eval_runtime": 12.646,
      "eval_samples_per_second": 296.537,
      "eval_steps_per_second": 4.666,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.47138023376465,
      "learning_rate": 0.0003,
      "loss": 2.7781,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.27413333333333334,
      "eval_loss": 2.650449275970459,
      "eval_runtime": 12.3317,
      "eval_samples_per_second": 304.094,
      "eval_steps_per_second": 4.784,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.053119659423828,
      "learning_rate": 0.0003,
      "loss": 2.5523,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.28746666666666665,
      "eval_loss": 2.5975472927093506,
      "eval_runtime": 12.4422,
      "eval_samples_per_second": 301.393,
      "eval_steps_per_second": 4.742,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 44.82758712768555,
      "learning_rate": 0.0003,
      "loss": 2.3949,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.35173333333333334,
      "eval_loss": 2.3734447956085205,
      "eval_runtime": 12.4126,
      "eval_samples_per_second": 302.113,
      "eval_steps_per_second": 4.753,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 15.801069259643555,
      "learning_rate": 0.0003,
      "loss": 2.258,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.35573333333333335,
      "eval_loss": 2.3238978385925293,
      "eval_runtime": 12.3161,
      "eval_samples_per_second": 304.48,
      "eval_steps_per_second": 4.79,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 20.476703643798828,
      "learning_rate": 0.0003,
      "loss": 2.136,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.38266666666666665,
      "eval_loss": 2.2414562702178955,
      "eval_runtime": 12.4407,
      "eval_samples_per_second": 301.429,
      "eval_steps_per_second": 4.742,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 19.00391387939453,
      "learning_rate": 0.0003,
      "loss": 2.0262,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3970666666666667,
      "eval_loss": 2.1767361164093018,
      "eval_runtime": 12.2068,
      "eval_samples_per_second": 307.206,
      "eval_steps_per_second": 4.833,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 1.9127,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4146666666666667,
      "eval_loss": 2.129997968673706,
      "eval_runtime": 12.4478,
      "eval_samples_per_second": 301.257,
      "eval_steps_per_second": 4.74,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}