macadeliccc commited on
Commit
543c7da
1 Parent(s): 74f4a14

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -10,6 +10,10 @@ Credit to Fernando Fernandes and Eric Hartford for their project [laserRMT](http
10
 
11
  This model is a medium-sized MoE implementation based on [cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser](https://huggingface.co/cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser)
12
 
 
 
 
 
13
  ## Prompt Format
14
 
15
  This model follows the same prompt format as the aforementioned model.
 
10
 
11
  This model is a medium-sized MoE implementation based on [cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser](https://huggingface.co/cognitivecomputations/dolphin-2.6-mistral-7b-dpo-laser)
12
 
13
+ A 2x7b configuration offers better performance than a standard 7b model even if loaded in 4 bit.
14
+
15
+ If this 2x7b model is loaded in 4 bit the hellaswag score is .8260 which is higher than the base model achieves on its own in full precision.
16
+
17
  ## Prompt Format
18
 
19
  This model follows the same prompt format as the aforementioned model.