trainer_state.json

{
  "best_metric": 0.8610666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0484/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": 0.6088770627975464,
      "learning_rate": 0.0001,
      "loss": 0.8766,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.525999128818512,
      "eval_runtime": 12.8622,
      "eval_samples_per_second": 291.551,
      "eval_steps_per_second": 4.587,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.631195068359375,
      "learning_rate": 0.0001,
      "loss": 0.323,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5741825699806213,
      "eval_runtime": 12.489,
      "eval_samples_per_second": 300.265,
      "eval_steps_per_second": 4.724,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 57.036598205566406,
      "learning_rate": 0.0001,
      "loss": 0.2181,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8354666666666667,
      "eval_loss": 0.6272071599960327,
      "eval_runtime": 12.4518,
      "eval_samples_per_second": 301.161,
      "eval_steps_per_second": 4.738,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 141.3208770751953,
      "learning_rate": 0.0001,
      "loss": 0.1643,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8504,
      "eval_loss": 0.576921284198761,
      "eval_runtime": 12.2754,
      "eval_samples_per_second": 305.489,
      "eval_steps_per_second": 4.806,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.950174331665039,
      "learning_rate": 0.0001,
      "loss": 0.133,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8232,
      "eval_loss": 0.7556498646736145,
      "eval_runtime": 12.8094,
      "eval_samples_per_second": 292.753,
      "eval_steps_per_second": 4.606,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6027507185935974,
      "learning_rate": 0.0001,
      "loss": 0.1277,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.5643792748451233,
      "eval_runtime": 12.2736,
      "eval_samples_per_second": 305.533,
      "eval_steps_per_second": 4.807,
      "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
}