Commit
•
543ce3b
1
Parent(s):
1b2c2a8
Update README.md
Browse files
README.md
CHANGED
|
@@ -10,6 +10,12 @@ tags:
|
|
| 10 |
- text-generation-inference
|
| 11 |
---
|
| 12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
**Dolphin-2.2-yi-34b-200k**, **Nous-Capybara-34B**, **Tess-M-v1.4**, **Airoboros-3_1-yi-34b-200k**, **PlatYi-34B-200K-Q**, and **Una-xaberius-34b-v1beta** merged with a new, experimental implementation of "dare ties" via mergekit. See:
|
| 14 |
|
| 15 |
> [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://github.com/yule-BUAA/MergeLM)
|
|
|
|
| 10 |
- text-generation-inference
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# Obsolete, replaced by https://huggingface.co/brucethemoose/Yi-34B-200K-DARE-merge-v5
|
| 14 |
+
|
| 15 |
+
The performanc of the new merge should be much better because dolphin 200K is a little broken. Old model description below:
|
| 16 |
+
***
|
| 17 |
+
|
| 18 |
+
|
| 19 |
**Dolphin-2.2-yi-34b-200k**, **Nous-Capybara-34B**, **Tess-M-v1.4**, **Airoboros-3_1-yi-34b-200k**, **PlatYi-34B-200K-Q**, and **Una-xaberius-34b-v1beta** merged with a new, experimental implementation of "dare ties" via mergekit. See:
|
| 20 |
|
| 21 |
> [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://github.com/yule-BUAA/MergeLM)
|