Loss_bbox and Loss_dfl are both 0

[sosgu]

I am executing on the terminal
CUDA_VISIBLE_DEVICES=1 python tools/train.py configs/finetune_coco/yolo_world_v2_l_vlpan_bn_sgd_1e-3_40e_8gpus_finetune_coco.py --work-dir log --amp --resume

After command, display after training
style="text-decoration-style:single">INFO</u></font> - Epoch(train) [25][50/90] base_lr: 1.0000e-03 lr: 4.3075e-04 eta: 0:09:11 time: 0.3959 data_time: 0.0126 memory: 5473 grad_norm: 0.0283 loss: 0.0016 loss_cls: 0.0016 loss_bbox: 0.0000 loss_dfl: 0.0000

I searched for relevant web pages online and said that I need to keep the tags in the JSON file consistent, but after modifying it, the effect is still the same. Therefore, I would like to consult with everyone here on how to solve this problem;

The categories in my training set are
"categories": [
{
"id": 0,
"name": "head",
"supercategory": "head"
}
]

The categories in the test set are
"categories": [
{
"id": 0,
"name": "head",
"supercategory": "head"
}
]
The content in coco_class_texts.json is
[["head"]]

Finally, in the yolo_corld_v2_1-vlpan-bn_sgd_1e-3_40e_8gpus_inetune_como.py file, the following is included:
metainfo = dict(classes=("head"))

coco_train_dataset = dict(delete=True,
type='MultiModalDataset',
dataset=dict(
type='YOLOv5CocoDataset',
metainfo=metainfo,

Only a portion of the code has been intercepted. Could you please provide guidance, experts