trainer_state.json

{
  "best_metric": 0.9213333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0939/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": 49.57124328613281,
      "learning_rate": 4.8108254489432715e-05,
      "loss": 0.7455,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.41330263018608093,
      "eval_runtime": 10.5921,
      "eval_samples_per_second": 354.036,
      "eval_steps_per_second": 5.57,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.65547752380371,
      "learning_rate": 4.2698504093359784e-05,
      "loss": 0.213,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.40724360942840576,
      "eval_runtime": 11.2273,
      "eval_samples_per_second": 334.007,
      "eval_steps_per_second": 5.255,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.624046325683594,
      "learning_rate": 3.459431688641196e-05,
      "loss": 0.0902,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.4273257553577423,
      "eval_runtime": 10.8823,
      "eval_samples_per_second": 344.595,
      "eval_steps_per_second": 5.422,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 2.5058963771413844e-05,
      "loss": 0.0376,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.442511647939682,
      "eval_runtime": 10.4404,
      "eval_samples_per_second": 359.183,
      "eval_steps_per_second": 5.651,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0018140003085136414,
      "learning_rate": 1.5487416222190068e-05,
      "loss": 0.0118,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.3797578811645508,
      "eval_runtime": 10.566,
      "eval_samples_per_second": 354.913,
      "eval_steps_per_second": 5.584,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.015333062969148159,
      "learning_rate": 7.36407332135117e-06,
      "loss": 0.0023,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9194666666666667,
      "eval_loss": 0.3559480607509613,
      "eval_runtime": 10.5434,
      "eval_samples_per_second": 355.673,
      "eval_steps_per_second": 5.596,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0060283467173576355,
      "learning_rate": 1.925640387263111e-06,
      "loss": 0.0006,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.35681426525115967,
      "eval_runtime": 11.0689,
      "eval_samples_per_second": 338.787,
      "eval_steps_per_second": 5.33,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0025660968385636806,
      "learning_rate": 6.953462348691098e-11,
      "loss": 0.0004,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9213333333333333,
      "eval_loss": 0.3541414141654968,
      "eval_runtime": 10.7884,
      "eval_samples_per_second": 347.597,
      "eval_steps_per_second": 5.469,
      "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
}