trainer_state.json

{
  "best_metric": 0.8928,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0376/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": 6.735155628520579e-05,
      "loss": 1.821,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8010666666666667,
      "eval_loss": 0.760115385055542,
      "eval_runtime": 7.2919,
      "eval_samples_per_second": 514.27,
      "eval_steps_per_second": 8.091,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2586033344268799,
      "learning_rate": 5.9777905730703695e-05,
      "loss": 0.5106,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.5116482973098755,
      "eval_runtime": 6.6913,
      "eval_samples_per_second": 560.427,
      "eval_steps_per_second": 8.817,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.099510669708252,
      "learning_rate": 4.843204364097674e-05,
      "loss": 0.2426,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.872,
      "eval_loss": 0.43604615330696106,
      "eval_runtime": 6.6901,
      "eval_samples_per_second": 560.526,
      "eval_steps_per_second": 8.819,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3529309928417206,
      "learning_rate": 3.5041274668689884e-05,
      "loss": 0.1093,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.43091410398483276,
      "eval_runtime": 7.1996,
      "eval_samples_per_second": 520.86,
      "eval_steps_per_second": 8.195,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.358248710632324,
      "learning_rate": 2.16442220022508e-05,
      "loss": 0.0414,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.4123343527317047,
      "eval_runtime": 7.7521,
      "eval_samples_per_second": 483.741,
      "eval_steps_per_second": 7.611,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0908132866024971,
      "learning_rate": 1.0280465465539994e-05,
      "loss": 0.0141,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.4148406386375427,
      "eval_runtime": 7.5757,
      "eval_samples_per_second": 495.001,
      "eval_steps_per_second": 7.788,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.40431714057922363,
      "learning_rate": 2.6800339785621635e-06,
      "loss": 0.0064,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.4113622307777405,
      "eval_runtime": 7.2364,
      "eval_samples_per_second": 518.213,
      "eval_steps_per_second": 8.153,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.10712999105453491,
      "learning_rate": 2.4337126681706066e-11,
      "loss": 0.0048,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.4113721549510956,
      "eval_runtime": 6.9319,
      "eval_samples_per_second": 540.976,
      "eval_steps_per_second": 8.511,
      "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
}