trainer_state.json

{
  "best_metric": 0.9322666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0596/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": 9.62613808353013e-05,
      "loss": 0.8118,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.4655208885669708,
      "eval_runtime": 6.6884,
      "eval_samples_per_second": 560.674,
      "eval_steps_per_second": 8.821,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.578086853027344,
      "learning_rate": 8.548019870451392e-05,
      "loss": 0.323,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8525333333333334,
      "eval_loss": 0.5509612560272217,
      "eval_runtime": 6.8963,
      "eval_samples_per_second": 543.767,
      "eval_steps_per_second": 8.555,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0082270298153162,
      "learning_rate": 6.929747794976645e-05,
      "loss": 0.1755,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4666654169559479,
      "eval_runtime": 6.3908,
      "eval_samples_per_second": 586.779,
      "eval_steps_per_second": 9.232,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0893423557281494,
      "learning_rate": 5.0176891109240266e-05,
      "loss": 0.0946,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.4545740783214569,
      "eval_runtime": 7.0958,
      "eval_samples_per_second": 528.478,
      "eval_steps_per_second": 8.315,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.000829799217171967,
      "learning_rate": 3.1029374200854165e-05,
      "loss": 0.0395,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9125333333333333,
      "eval_loss": 0.3920742869377136,
      "eval_runtime": 6.4093,
      "eval_samples_per_second": 585.087,
      "eval_steps_per_second": 9.205,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0018756025237962604,
      "learning_rate": 1.4769963101236895e-05,
      "loss": 0.0115,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9293333333333333,
      "eval_loss": 0.34173503518104553,
      "eval_runtime": 6.3578,
      "eval_samples_per_second": 589.83,
      "eval_steps_per_second": 9.28,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.00023881554079707712,
      "learning_rate": 3.87400575837657e-06,
      "loss": 0.0024,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.928,
      "eval_loss": 0.33349284529685974,
      "eval_runtime": 6.3726,
      "eval_samples_per_second": 588.457,
      "eval_steps_per_second": 9.258,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.514221523597371e-05,
      "learning_rate": 3.129056243833528e-10,
      "loss": 0.0007,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9322666666666667,
      "eval_loss": 0.3191070556640625,
      "eval_runtime": 6.3111,
      "eval_samples_per_second": 594.193,
      "eval_steps_per_second": 9.349,
      "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
}