trainer_state.json

{
  "best_metric": 0.9026666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0167/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": 3e-05,
      "loss": 0.8255,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.42356276512145996,
      "eval_runtime": 6.0581,
      "eval_samples_per_second": 619.002,
      "eval_steps_per_second": 9.739,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6156761646270752,
      "learning_rate": 3e-05,
      "loss": 0.1992,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.36208847165107727,
      "eval_runtime": 5.7038,
      "eval_samples_per_second": 657.459,
      "eval_steps_per_second": 10.344,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.46304702758789,
      "learning_rate": 3e-05,
      "loss": 0.0838,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.3513196110725403,
      "eval_runtime": 5.6331,
      "eval_samples_per_second": 665.711,
      "eval_steps_per_second": 10.474,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08811438083648682,
      "learning_rate": 3e-05,
      "loss": 0.0545,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9024,
      "eval_loss": 0.38742688298225403,
      "eval_runtime": 5.5919,
      "eval_samples_per_second": 670.608,
      "eval_steps_per_second": 10.551,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07148344814777374,
      "learning_rate": 3e-05,
      "loss": 0.04,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.399539977312088,
      "eval_runtime": 5.8917,
      "eval_samples_per_second": 636.49,
      "eval_steps_per_second": 10.014,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.310591220855713,
      "learning_rate": 3e-05,
      "loss": 0.0336,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8941333333333333,
      "eval_loss": 0.4541224539279938,
      "eval_runtime": 5.8993,
      "eval_samples_per_second": 635.665,
      "eval_steps_per_second": 10.001,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.21748317778110504,
      "learning_rate": 3e-05,
      "loss": 0.0265,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.47688546776771545,
      "eval_runtime": 5.9203,
      "eval_samples_per_second": 633.416,
      "eval_steps_per_second": 9.966,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.009320047684013844,
      "learning_rate": 3e-05,
      "loss": 0.0212,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9026666666666666,
      "eval_loss": 0.4357467591762543,
      "eval_runtime": 5.8614,
      "eval_samples_per_second": 639.777,
      "eval_steps_per_second": 10.066,
      "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
}