Add README
Browse files
README.md
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
pipeline_tag: text-generation
|
| 6 |
+
inference: false
|
| 7 |
+
tags:
|
| 8 |
+
- dare
|
| 9 |
+
- super mario merge
|
| 10 |
+
- pytorch
|
| 11 |
+
- mixtral
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# mixtral dare test
|
| 15 |
+
|
| 16 |
+
The following were merged with DARE using [https://github.com/martyn/safetensors-merge-supermario](https://github.com/martyn/safetensors-merge-supermario)
|
| 17 |
+
|
| 18 |
+
## Mergelist
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
```
|
| 22 |
+
mistralai/Mixtral-8x7B-Instruct-v0.1
|
| 23 |
+
Open-Orca/Mixtral-SlimOrca-8x7B
|
| 24 |
+
```
|
| 25 |
+
|
| 26 |
+
## Merge command
|
| 27 |
+
|
| 28 |
+
```
|
| 29 |
+
python3 hf_merge.py to_merge_mixtral0.txt mixtral-0 -p 0.3 -lambda 2.1
|
| 30 |
+
```
|
| 31 |
+
|
| 32 |
+
## Notes
|
| 33 |
+
|
| 34 |
+
* This is primarily a test to see if merging mixtral models works.
|
| 35 |
+
* I skip merge on the MoE gates.
|