athirdpath
commited on
Commit
•
c4df92d
1
Parent(s):
93f0182
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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/pbpJHpk/iambe-sml.png"/><font size="6"> <b>Iambe-20b-v3_TEST-RP_cDPO</b> </font></p>
|
9 |
+
<p align="center"><font size="4"> <b>Alpaca prompt formatting</b> </font></p>
|
10 |
+
|
11 |
+
### Description
|
12 |
+
|
13 |
+
Named after a charming daughter of Echo and Pan in Greek myth, Iambe-v3 is, as far as I am aware, the very first LLM trained with DPO on an erotic roleplay dataset.
|
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) for more info. The cDPO training was done directly on Iambe-20b-DARE-v2, I was researching 11b merges to reduce the compute, but it went nowhere, so I just bit the bullet on cost. The notebook used to train this model is also available in the dataset's repo.
|