trainer_state.json

{
  "best_metric": 0.9261333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0482/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": 15.785234451293945,
      "learning_rate": 8.404170694504206e-05,
      "loss": 0.7811,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8416,
      "eval_loss": 0.528620719909668,
      "eval_runtime": 11.7754,
      "eval_samples_per_second": 318.461,
      "eval_steps_per_second": 5.01,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 6.76835800106096e-05,
      "loss": 0.2634,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.4061738848686218,
      "eval_runtime": 10.7832,
      "eval_samples_per_second": 347.762,
      "eval_steps_per_second": 5.471,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.38582736253738403,
      "learning_rate": 4.521226898668646e-05,
      "loss": 0.1092,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8933333333333333,
      "eval_loss": 0.3632829785346985,
      "eval_runtime": 11.385,
      "eval_samples_per_second": 329.382,
      "eval_steps_per_second": 5.182,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.4253159761428833,
      "learning_rate": 2.268408065920248e-05,
      "loss": 0.0288,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9114666666666666,
      "eval_loss": 0.3304084837436676,
      "eval_runtime": 10.7091,
      "eval_samples_per_second": 350.17,
      "eval_steps_per_second": 5.509,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.010145890526473522,
      "learning_rate": 6.135424897443279e-06,
      "loss": 0.0045,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9242666666666667,
      "eval_loss": 0.2834729552268982,
      "eval_runtime": 10.9076,
      "eval_samples_per_second": 343.798,
      "eval_steps_per_second": 5.409,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0012584629002958536,
      "learning_rate": 5.006485928676918e-10,
      "loss": 0.0005,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9261333333333334,
      "eval_loss": 0.28025585412979126,
      "eval_runtime": 11.5252,
      "eval_samples_per_second": 325.373,
      "eval_steps_per_second": 5.119,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}