Nexesenex
/

Llama_3.x_70b_Tess_Dolphin_128K_v1.2

Text Generation

text-generation-inference

Model card Files Files and versions

Nexesenex commited on Mar 18

Commit

0b41d6a

·

verified ·

1 Parent(s): 338c77b

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -23,7 +23,7 @@ A merge between a L3 70b model (Dolphin 2.9.1) and a L3.1 70b base model (Tess 3
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6451b24dc5d273f95482bfa4/L6wJK3wDcAJBj5MvZPWO1.png)
-- Blue is v1.2 (1+1 untouched layers), Purple is v1.1 (4+4 untouched layers), Orange is v1.0 (12+12 untouched layers).
 - I initially thought that v1.0 would be green and 1.1 would be red, but I messed up I guess. The results are still usable though.
 - You can multiply the value on the graphic by 2, because that's the epsilon I used, in the way demonstrated by the aforementioned guys.

 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6451b24dc5d273f95482bfa4/L6wJK3wDcAJBj5MvZPWO1.png)
+- Blue is v1.2 (1+1 untouched layers, if any), Purple is v1.1 (4+4 untouched layers), Orange is v1.0 (12+12 untouched layers).
 - I initially thought that v1.0 would be green and 1.1 would be red, but I messed up I guess. The results are still usable though.
 - You can multiply the value on the graphic by 2, because that's the epsilon I used, in the way demonstrated by the aforementioned guys.