trainer_state.json

{
  "best_metric": 0.4437333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0343/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": 48.42436218261719,
      "learning_rate": 0.00028885116082809934,
      "loss": 3.3486,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22,
      "eval_loss": 2.894763231277466,
      "eval_runtime": 7.1568,
      "eval_samples_per_second": 523.978,
      "eval_steps_per_second": 8.244,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.101150512695312,
      "learning_rate": 0.0002565651599376707,
      "loss": 2.7768,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2618666666666667,
      "eval_loss": 2.7562365531921387,
      "eval_runtime": 7.5905,
      "eval_samples_per_second": 494.037,
      "eval_steps_per_second": 7.773,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.922101974487305,
      "learning_rate": 0.00020805557946231208,
      "loss": 2.5314,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3144,
      "eval_loss": 2.5082526206970215,
      "eval_runtime": 7.2093,
      "eval_samples_per_second": 520.162,
      "eval_steps_per_second": 8.184,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 30.30898094177246,
      "learning_rate": 0.00015070756328895485,
      "loss": 2.3309,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3568,
      "eval_loss": 2.3080902099609375,
      "eval_runtime": 7.6998,
      "eval_samples_per_second": 487.027,
      "eval_steps_per_second": 7.663,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 38.64478302001953,
      "learning_rate": 9.325182701893136e-05,
      "loss": 2.1123,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4,
      "eval_loss": 2.164036989212036,
      "eval_runtime": 7.0778,
      "eval_samples_per_second": 529.825,
      "eval_steps_per_second": 8.336,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 38.04201889038086,
      "learning_rate": 4.443548566180648e-05,
      "loss": 1.8731,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.42693333333333333,
      "eval_loss": 2.062213182449341,
      "eval_runtime": 7.0641,
      "eval_samples_per_second": 530.852,
      "eval_steps_per_second": 8.352,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 44.97215270996094,
      "learning_rate": 1.1690384667966174e-05,
      "loss": 1.6528,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4418666666666667,
      "eval_loss": 2.0493578910827637,
      "eval_runtime": 6.8997,
      "eval_samples_per_second": 543.502,
      "eval_steps_per_second": 8.551,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 11.127470016479492,
      "learning_rate": 1.668828642892306e-09,
      "loss": 1.4854,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4437333333333333,
      "eval_loss": 2.0601093769073486,
      "eval_runtime": 6.9966,
      "eval_samples_per_second": 535.978,
      "eval_steps_per_second": 8.433,
      "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
}