trainer_state.json

{
  "best_metric": 0.4104,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0511/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": 17.064165115356445,
      "learning_rate": 0.0004813069041765065,
      "loss": 3.4481,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1848,
      "eval_loss": 3.0484354496002197,
      "eval_runtime": 7.0125,
      "eval_samples_per_second": 534.76,
      "eval_steps_per_second": 8.414,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.38719367980957,
      "learning_rate": 0.00042740099352256953,
      "loss": 2.9512,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.24746666666666667,
      "eval_loss": 2.793703556060791,
      "eval_runtime": 6.8491,
      "eval_samples_per_second": 547.518,
      "eval_steps_per_second": 8.614,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.753612518310547,
      "learning_rate": 0.0003464873897488322,
      "loss": 2.7578,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.26426666666666665,
      "eval_loss": 2.6992316246032715,
      "eval_runtime": 6.871,
      "eval_samples_per_second": 545.776,
      "eval_steps_per_second": 8.587,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.272911071777344,
      "learning_rate": 0.00025088445554620134,
      "loss": 2.5883,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.32426666666666665,
      "eval_loss": 2.4748411178588867,
      "eval_runtime": 6.8981,
      "eval_samples_per_second": 543.626,
      "eval_steps_per_second": 8.553,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 62.98331832885742,
      "learning_rate": 0.00015514687100427082,
      "loss": 2.3845,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.34586666666666666,
      "eval_loss": 2.36236310005188,
      "eval_runtime": 7.7373,
      "eval_samples_per_second": 484.663,
      "eval_steps_per_second": 7.625,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 41.5197868347168,
      "learning_rate": 7.384981550618446e-05,
      "loss": 2.2009,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.37866666666666665,
      "eval_loss": 2.238170623779297,
      "eval_runtime": 6.8327,
      "eval_samples_per_second": 548.83,
      "eval_steps_per_second": 8.635,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 26.711387634277344,
      "learning_rate": 1.9370028791882848e-05,
      "loss": 2.0175,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.40586666666666665,
      "eval_loss": 2.158369302749634,
      "eval_runtime": 7.1061,
      "eval_samples_per_second": 527.712,
      "eval_steps_per_second": 8.303,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 24.51226043701172,
      "learning_rate": 1.5645281219167639e-09,
      "loss": 1.9066,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4104,
      "eval_loss": 2.153048038482666,
      "eval_runtime": 6.7887,
      "eval_samples_per_second": 552.387,
      "eval_steps_per_second": 8.691,
      "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
}