matchaaaaa
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -4,12 +4,35 @@ library_name: transformers
|
|
4 |
tags:
|
5 |
- mergekit
|
6 |
- merge
|
7 |
-
|
8 |
---
|
|
|
|
|
|
|
9 |
# MN-Tiramisu-12B
|
10 |
|
11 |
This is a really yappity-yappy yapping model that's good for long-form RP. Tried to rein it in with Mahou and give it some more character understanding with Pantheon. Feedback is always welcome.
|
12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
## Merge Details
|
14 |
|
15 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
|
|
4 |
tags:
|
5 |
- mergekit
|
6 |
- merge
|
|
|
7 |
---
|
8 |
+
|
9 |
+
![cute](https://huggingface.co/matchaaaaa/MN-Tiramisu-12B/resolve/main/tiramisu-cute.png)
|
10 |
+
|
11 |
# MN-Tiramisu-12B
|
12 |
|
13 |
This is a really yappity-yappy yapping model that's good for long-form RP. Tried to rein it in with Mahou and give it some more character understanding with Pantheon. Feedback is always welcome.
|
14 |
|
15 |
+
**Native Context Length: 16K/16384** *(can be extended using RoPE, YMMY)*
|
16 |
+
|
17 |
+
## Prompt Template: Chat
|
18 |
+
|
19 |
+
```
|
20 |
+
<|im_start|>system
|
21 |
+
{system prompt}<|im_end|>
|
22 |
+
<|im_start|>user
|
23 |
+
{message}<|im_end|>
|
24 |
+
<|im_start|>assistant
|
25 |
+
{response}
|
26 |
+
```
|
27 |
+
|
28 |
+
## Recommended Settings:
|
29 |
+
|
30 |
+
Here are some settings ranges that tend to work for me. They aren't strict values, and there's a bit of leeway in them. Feel free to experiment a bit!
|
31 |
+
|
32 |
+
* Temperature: **1.0** (maybe less, a little bit goes a long way with Nemo)
|
33 |
+
* Min-P: **0.1** to **0.2**
|
34 |
+
* *(all other samplers disabled)*
|
35 |
+
|
36 |
## Merge Details
|
37 |
|
38 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|