trainer_state.json

{
  "best_metric": 0.39653333333333335,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0091/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": 21.22452735900879,
      "learning_rate": 0.00048141860138016556,
      "loss": 3.3689,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.17866666666666667,
      "eval_loss": 3.097720146179199,
      "eval_runtime": 10.6566,
      "eval_samples_per_second": 351.894,
      "eval_steps_per_second": 5.536,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.447311401367188,
      "learning_rate": 0.00042760859989611786,
      "loss": 2.8751,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.23546666666666666,
      "eval_loss": 2.8123464584350586,
      "eval_runtime": 10.5439,
      "eval_samples_per_second": 355.656,
      "eval_steps_per_second": 5.596,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.89251136779785,
      "learning_rate": 0.00034675929910385353,
      "loss": 2.6682,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2584,
      "eval_loss": 2.7068700790405273,
      "eval_runtime": 10.0931,
      "eval_samples_per_second": 371.542,
      "eval_steps_per_second": 5.846,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.427350997924805,
      "learning_rate": 0.00025117927214825806,
      "loss": 2.4716,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2992,
      "eval_loss": 2.523237943649292,
      "eval_runtime": 10.4064,
      "eval_samples_per_second": 360.355,
      "eval_steps_per_second": 5.67,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 34.238224029541016,
      "learning_rate": 0.00015541971169821895,
      "loss": 2.2658,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.34373333333333334,
      "eval_loss": 2.3808443546295166,
      "eval_runtime": 10.144,
      "eval_samples_per_second": 369.676,
      "eval_steps_per_second": 5.816,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 27.53048324584961,
      "learning_rate": 7.405914276967749e-05,
      "loss": 2.0073,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3824,
      "eval_loss": 2.267601490020752,
      "eval_runtime": 10.5329,
      "eval_samples_per_second": 356.026,
      "eval_steps_per_second": 5.601,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 38.85408020019531,
      "learning_rate": 1.9483974446610292e-05,
      "loss": 1.7119,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3912,
      "eval_loss": 2.276085376739502,
      "eval_runtime": 11.0315,
      "eval_samples_per_second": 339.936,
      "eval_steps_per_second": 5.348,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.698777675628662,
      "learning_rate": 2.7813810714871766e-09,
      "loss": 1.4519,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.39653333333333335,
      "eval_loss": 2.3669447898864746,
      "eval_runtime": 10.5317,
      "eval_samples_per_second": 356.068,
      "eval_steps_per_second": 5.602,
      "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
}