trainer_state.json

{
  "best_metric": 0.4109333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0076/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": 0.00048527961022201525,
      "loss": 3.4016,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.19066666666666668,
      "eval_loss": 3.1119484901428223,
      "eval_runtime": 8.3391,
      "eval_samples_per_second": 449.689,
      "eval_steps_per_second": 7.075,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 25.417512893676758,
      "learning_rate": 0.0004421832247293072,
      "loss": 2.9122,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.22453333333333333,
      "eval_loss": 2.870037078857422,
      "eval_runtime": 6.4887,
      "eval_samples_per_second": 577.932,
      "eval_steps_per_second": 9.093,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.274173736572266,
      "learning_rate": 0.00037590670601192445,
      "loss": 2.7013,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.29146666666666665,
      "eval_loss": 2.6043310165405273,
      "eval_runtime": 6.7871,
      "eval_samples_per_second": 552.523,
      "eval_steps_per_second": 8.693,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 23.650808334350586,
      "learning_rate": 0.00029444398036442514,
      "loss": 2.5426,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3136,
      "eval_loss": 2.515803575515747,
      "eval_runtime": 7.0181,
      "eval_samples_per_second": 534.336,
      "eval_steps_per_second": 8.407,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 21.027603149414062,
      "learning_rate": 0.00020762065476168375,
      "loss": 2.3512,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3672,
      "eval_loss": 2.315739393234253,
      "eval_runtime": 6.7083,
      "eval_samples_per_second": 559.008,
      "eval_steps_per_second": 8.795,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 23.130861282348633,
      "learning_rate": 0.00012590890364720656,
      "loss": 2.1302,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3872,
      "eval_loss": 2.2067089080810547,
      "eval_runtime": 6.4305,
      "eval_samples_per_second": 583.157,
      "eval_steps_per_second": 9.175,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 23.492429733276367,
      "learning_rate": 5.9164370142409884e-05,
      "loss": 1.863,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4056,
      "eval_loss": 2.1820480823516846,
      "eval_runtime": 6.8932,
      "eval_samples_per_second": 544.018,
      "eval_steps_per_second": 8.559,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 19.9686336517334,
      "learning_rate": 1.5437430032376602e-05,
      "loss": 1.5706,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4109333333333333,
      "eval_loss": 2.2343146800994873,
      "eval_runtime": 6.671,
      "eval_samples_per_second": 562.134,
      "eval_steps_per_second": 8.844,
      "step": 2664
    }
  ],
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}