add VideoCC

README.md

@@ -74,7 +74,7 @@ Scaling behavior of EVA-CLIP with zero-shot classification performance averaged
 
 
 - To construct Merged-2B, we merged 1.6 billion samples from [LAION-2B](https://laion.ai/blog/laion-5b/) dataset with 0.4 billion samples from [COYO-700M](https://github.com/kakaobrain/coyo-dataset).
-- The Merged-2B+ consists of all samples from Merged-2B, along with 20 millions samples from [LAION-COCO](https://laion.ai/blog/laion-coco/) and 23 millions samples from Merged-video including [VideoCC3M](https://github.com/google-research-datasets/videoCC-data), [InternVid](https://huggingface.co/datasets/OpenGVLab/InternVid) and [WebVid-10M](https://maxbain.com/webvid-dataset/). Merged-video was added at the end of the training process.
+- The Merged-2B+ consists of all samples from Merged-2B, along with 20 millions samples from [LAION-COCO](https://laion.ai/blog/laion-coco/) and 23 millions samples from Merged-video including [VideoCC](https://github.com/google-research-datasets/videoCC-data), [InternVid](https://huggingface.co/datasets/OpenGVLab/InternVid) and [WebVid-10M](https://maxbain.com/webvid-dataset/). Merged-video was added at the end of the training process.
 
 **It's important to note that all results presented in the paper are evaluated using PyTorch weights. There may be differences in performance when using Hugging Face (hf) models.**