mjbuehler commited on
Commit
84ffd11
1 Parent(s): 400d973

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -47,6 +47,10 @@ Cephalo provides a robust framework for multimodal interaction and understanding
47
  - [lamm-mit/Cephalo-Idefics-2-vision-12b-alpha](https://huggingface.co/lamm-mit/Cephalo-Idefics-2-vision-12b-alpha)
48
  - Merged model, 32+16=48 layers, checkpoint after first epoch. Trained on GPT-4o distilled image-text data from Wikipedia and scientific papers (dataset derivived from both Idefics-2 and GPT-4o distillation of the paper corpus).
49
 
 
 
 
 
50
  ### Additional codes and tools
51
 
52
  Additional codes and tools are provided at [https://github.com/lamm-mit/Cephalo](https://github.com/lamm-mit/Cephalo).
 
47
  - [lamm-mit/Cephalo-Idefics-2-vision-12b-alpha](https://huggingface.co/lamm-mit/Cephalo-Idefics-2-vision-12b-alpha)
48
  - Merged model, 32+16=48 layers, checkpoint after first epoch. Trained on GPT-4o distilled image-text data from Wikipedia and scientific papers (dataset derivived from both Idefics-2 and GPT-4o distillation of the paper corpus).
49
 
50
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/623ce1c6b66fedf374859fe7/3Nfhn3f3FyK7Zgdg9GKJQ.png)
51
+
52
+ The image shows a summary of model merging approach, constructing larger models from smaller pre-trained building blocks.
53
+
54
  ### Additional codes and tools
55
 
56
  Additional codes and tools are provided at [https://github.com/lamm-mit/Cephalo](https://github.com/lamm-mit/Cephalo).