athirdpath
commited on
Commit
•
e7d059f
1
Parent(s):
40b27a0
Update README.md
Browse files
README.md
CHANGED
@@ -1,5 +1,25 @@
|
|
1 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
-
|
4 |
|
5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
tags:
|
6 |
+
- not-for-all-audiences
|
7 |
+
---
|
8 |
+
<p align="center"><img src="https://i.ibb.co/0Z63JHp/Iambe-RP-sml.png"/><font size="6"> <b>Iambe-RP-cDPO-20b-ALT</b> </font></p>
|
9 |
+
<p align="center"><font size="4"> <b>Trained with Alpaca prompt formatting, some like ChatML</b> </font></p>
|
10 |
|
11 |
+
### Description
|
12 |
|
13 |
+
Named after a charming daughter of Echo and Pan in Greek myth, Iambe-RP-ALT failed training on the v2 dataset at 0.6 epochs. Therefore, I don't feel comfortable considering this a full new generation.
|
14 |
+
|
15 |
+
Iambe is intended to have the best realistically possible understanding of instructions, anatomy and scene state for a 20b merge, while remaining passionate and humanoid in "voice".
|
16 |
+
|
17 |
+
### Update Methodology
|
18 |
+
|
19 |
+
Take a look at [the dataset v2 Iambe and I created together](https://huggingface.co/datasets/athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW-v2) for more info. The cDPO training was done directly on Iambe-Storyteller-20b, and I The notebook used is also available in the dataset's repo.
|
20 |
+
|
21 |
+
<p align="center"><font size="5"> <b>Assistant Example @ q5_k_m</b> </font></p>
|
22 |
+
<p align="center"><img src="https://i.ibb.co/YtwztYH/Screenshot-2023-12-09-113216.png"/>
|
23 |
+
|
24 |
+
<p align="center"><font size="5"> <b>NSFW Writing Example @ q5_k_m</b> </font></p>
|
25 |
+
<p align="center"><img src="https://i.ibb.co/4JHGzph/Screenshot-2023-12-09-113534.png"/>
|