Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,56 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
pipeline_tag: text-generation
|
4 |
+
tags:
|
5 |
+
- chat
|
6 |
+
base_model:
|
7 |
+
- Gryphe/Pantheon-RP-1.6-12b-Nemo
|
8 |
+
- Sao10K/MN-12B-Lyra-v3
|
9 |
+
- anthracite-org/magnum-v2.5-12b-kto
|
10 |
+
- nbeerbower/mistral-nemo-bophades-12B
|
11 |
+
---
|
12 |
+
# StarDust-12b-v1
|
13 |
+
|
14 |
+
## Weights are uploading, had some issues, give it 2h again... sorry for the delay!
|
15 |
+
|
16 |
+
The result of this merge is in my opinion a more vibrant and less generic sonnet inspired prose, it's able to be gentle and harsh where asked.
|
17 |
+
I've personally been trying to get a more spice while also compensating for the Magnum-v2.5 having the issue on my end that it simply won't stop yapping.
|
18 |
+
|
19 |
+
- This model is intended to be used as a Role-playing model.
|
20 |
+
- Its direct conversational output is... I can't even say it's luck, it's just not made for it.
|
21 |
+
|
22 |
+
|
23 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6303fa71fc783bfc7443e7ae/qRsB-uefbKKrAqxknbWtN.png)
|
24 |
+
|
25 |
+
## Prompting
|
26 |
+
Both Mistral and ChatML should work though I had better results with ChatML:
|
27 |
+
ChatML Example:
|
28 |
+
```py
|
29 |
+
"""<|im_start|>user
|
30 |
+
Hi there!<|im_end|>
|
31 |
+
<|im_start|>assistant
|
32 |
+
Nice to meet you!<|im_end|>
|
33 |
+
<|im_start|>user
|
34 |
+
Can I ask a question?<|im_end|>
|
35 |
+
<|im_start|>assistant
|
36 |
+
"""
|
37 |
+
```
|
38 |
+
|
39 |
+
|
40 |
+
|
41 |
+
## Merge Details
|
42 |
+
### Merge Method
|
43 |
+
|
44 |
+
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [Sao10K/MN-12B-Lyra-v3](https://huggingface.co/Sao10K/MN-12B-Lyra-v3) as a base.
|
45 |
+
|
46 |
+
### Models Merged
|
47 |
+
|
48 |
+
The following models were included in the merge:
|
49 |
+
* [Gryphe/Pantheon-RP-1.6-12b-Nemo](https://huggingface.co/Gryphe/Pantheon-RP-1.6-12b-Nemo)
|
50 |
+
* [anthracite-org/magnum-v2.5-12b-kto](https://huggingface.co/anthracite-org/magnum-v2.5-12b-kto)
|
51 |
+
* [nbeerbower/mistral-nemo-bophades-12B](https://huggingface.co/nbeerbower/mistral-nemo-bophades-12B)
|
52 |
+
* [Sao10K/MN-12B-Lyra-v3](https://huggingface.co/Sao10K/MN-12B-Lyra-v3)
|
53 |
+
|
54 |
+
### Special Thanks
|
55 |
+
|
56 |
+
Special thanks to the SillyTilly and myself for helping me find the energy to finish this.
|