trainer_state.json

{
  "best_metric": 0.35413333333333336,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0085/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 18.641767501831055,
      "learning_rate": 0.00045280733623674006,
      "loss": 3.4125,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.164,
      "eval_loss": 3.188464403152466,
      "eval_runtime": 5.9084,
      "eval_samples_per_second": 634.688,
      "eval_steps_per_second": 9.986,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.506210327148438,
      "learning_rate": 0.0003281509432839476,
      "loss": 2.8699,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.24293333333333333,
      "eval_loss": 2.8348729610443115,
      "eval_runtime": 5.2014,
      "eval_samples_per_second": 720.961,
      "eval_steps_per_second": 11.343,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 26.032352447509766,
      "learning_rate": 0.0001736435462495525,
      "loss": 2.5717,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2949333333333333,
      "eval_loss": 2.623706817626953,
      "eval_runtime": 5.2468,
      "eval_samples_per_second": 714.722,
      "eval_steps_per_second": 11.245,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 11.21835708618164,
      "learning_rate": 4.8301719287419e-05,
      "loss": 2.2552,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3466666666666667,
      "eval_loss": 2.4273500442504883,
      "eval_runtime": 5.2466,
      "eval_samples_per_second": 714.745,
      "eval_steps_per_second": 11.245,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 27.722665786743164,
      "learning_rate": 1.7800850740101736e-09,
      "loss": 1.9881,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.35413333333333336,
      "eval_loss": 2.4061458110809326,
      "eval_runtime": 5.5223,
      "eval_samples_per_second": 679.067,
      "eval_steps_per_second": 10.684,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}