Kquant03
/

CognitiveFusion-4x7B-bf16-MoE

Text Generation

Mixture of Experts

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Kquant03 commited on Jan 5

Commit

d4fa544

•

1 Parent(s): 48778a9

Update README.md

Files changed (1) hide show

README.md +3 -2

README.md CHANGED Viewed

@@ -18,8 +18,9 @@ I was approached with the idea to make a merge based on story telling, and consi
 We believe that this, while it might not be better logically than mixtral base instruct, is definitely more creative. Special thanks to [NeuralNovel](https://huggingface.co/NeuralNovel) for collaborating with me on this project
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/6589d7e6586088fd2784a12c/V4cv6tthy1quRRMCvAf2H.png)
-# The model performed slightly better than base mixtral instruct in erotic roleplay variety on  [Ayumi's benchmark](http://ayumi.m8geil.de/ayumi_bench_v3_results.html) and #29/818 BEST FRANKENMOE overall!!!
 # "[What is a Mixture of Experts (MoE)?](https://huggingface.co/blog/moe)"
 ### (from the MistralAI papers...click the quoted question above to navigate to it directly.)

 We believe that this, while it might not be better logically than mixtral base instruct, is definitely more creative. Special thanks to [NeuralNovel](https://huggingface.co/NeuralNovel) for collaborating with me on this project
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/6589d7e6586088fd2784a12c/xXXJhZNJ4q3suxJ9LyLqK.png)
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/6589d7e6586088fd2784a12c/ZpX1KMNYj11k4pF0NX3Q9.png)
+It performs better than base mixtral 8x across many evaluations. It's half the size and is comparable to most MoEs. Thanks so much to HuggingFace for evaluating it!
 # "[What is a Mixture of Experts (MoE)?](https://huggingface.co/blog/moe)"
 ### (from the MistralAI papers...click the quoted question above to navigate to it directly.)