trainer_state.json

{
  "best_metric": 0.8549333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0116/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": 44.548248291015625,
      "learning_rate": 9e-05,
      "loss": 0.8992,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5619867444038391,
      "eval_runtime": 6.4734,
      "eval_samples_per_second": 579.295,
      "eval_steps_per_second": 9.114,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 82.8205337524414,
      "learning_rate": 9e-05,
      "loss": 0.3762,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5616526007652283,
      "eval_runtime": 6.5503,
      "eval_samples_per_second": 572.493,
      "eval_steps_per_second": 9.007,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.6574699878692627,
      "learning_rate": 9e-05,
      "loss": 0.2428,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8282666666666667,
      "eval_loss": 0.5956273674964905,
      "eval_runtime": 6.659,
      "eval_samples_per_second": 563.148,
      "eval_steps_per_second": 8.86,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 10.001331329345703,
      "learning_rate": 9e-05,
      "loss": 0.1786,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8354666666666667,
      "eval_loss": 0.673226535320282,
      "eval_runtime": 6.8732,
      "eval_samples_per_second": 545.595,
      "eval_steps_per_second": 8.584,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.8949676752090454,
      "learning_rate": 9e-05,
      "loss": 0.1542,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5969272255897522,
      "eval_runtime": 6.1971,
      "eval_samples_per_second": 605.121,
      "eval_steps_per_second": 9.521,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.10336077213287354,
      "learning_rate": 9e-05,
      "loss": 0.1352,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.5846278667449951,
      "eval_runtime": 6.3262,
      "eval_samples_per_second": 592.771,
      "eval_steps_per_second": 9.326,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}