diff --git a/PRETRAIN.md b/PRETRAIN.md
index b8a2fd0..d7b8a5d 100644
--- a/PRETRAIN.md
+++ b/PRETRAIN.md
@@ -2,7 +2,7 @@
 
 See [INSTALL.md](https://github.com/keyu-tian/SparK/blob/main/INSTALL.md) to prepare dependencies and ImageNet dataset.
 
-**Note: for network definitions, we directly use `timm.models.ResNet` and [official ConvNeXt](https://github.com/facebookresearch/ConvNeXt).**
+**Note: for network definitions, we directly use `timm.models.ResNet` and [official ConvNeXt](https://github.com/facebookresearch/ConvNeXt/blob/048efcea897d999aed302f2639b6270aedf8d4c8/models/convnext.py).**
 
 
 ## Pre-training on ImageNet-1k from scratch
diff --git a/downstream_imagenet/README.md b/downstream_imagenet/README.md
index 53606c0..42eab0b 100644
--- a/downstream_imagenet/README.md
+++ b/downstream_imagenet/README.md
@@ -7,7 +7,7 @@ This `downstream_imagenet` is isolated from pre-training codes. One can treat th
 
 See [INSTALL.md](https://github.com/keyu-tian/SparK/blob/main/INSTALL.md) to prepare dependencies and ImageNet dataset.
 
-**Note: for network definitions, we directly use `timm.models.ResNet` and [official ConvNeXt](https://github.com/facebookresearch/ConvNeXt).**
+**Note: for network definitions, we directly use `timm.models.ResNet` and [official ConvNeXt](https://github.com/facebookresearch/ConvNeXt/blob/048efcea897d999aed302f2639b6270aedf8d4c8/models/convnext.py).**
 
 
 ## Fine-tuning on ImageNet-1k from pre-trained weights