trainer_state.json

{
  "best_metric": 0.9,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0847/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": 6.286877155303955,
      "learning_rate": 8.554359905560886e-05,
      "loss": 1.7444,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7824,
      "eval_loss": 0.7820528745651245,
      "eval_runtime": 12.437,
      "eval_samples_per_second": 301.519,
      "eval_steps_per_second": 4.744,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.31517728663805e-05,
      "loss": 0.5369,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8450666666666666,
      "eval_loss": 0.5275892615318298,
      "eval_runtime": 13.0928,
      "eval_samples_per_second": 286.417,
      "eval_steps_per_second": 4.506,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.3555189371109009,
      "learning_rate": 5.5131661351986275e-05,
      "loss": 0.2877,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8696,
      "eval_loss": 0.4366984963417053,
      "eval_runtime": 13.2077,
      "eval_samples_per_second": 283.925,
      "eval_steps_per_second": 4.467,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 3.516402310353216e-05,
      "loss": 0.1476,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.39446961879730225,
      "eval_runtime": 13.1,
      "eval_samples_per_second": 286.26,
      "eval_steps_per_second": 4.504,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.31559672951698303,
      "learning_rate": 1.708543859906803e-05,
      "loss": 0.0693,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.3774349093437195,
      "eval_runtime": 13.0872,
      "eval_samples_per_second": 286.539,
      "eval_steps_per_second": 4.508,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.381342887878418,
      "learning_rate": 4.53567532968734e-06,
      "loss": 0.0301,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.3535594940185547,
      "eval_runtime": 13.0832,
      "eval_samples_per_second": 286.627,
      "eval_steps_per_second": 4.51,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}