Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- not-for-all-audiences
|
| 5 |
+
- nsfw
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
## MiquMaid v2 2x70 DPO
|
| 9 |
+
|
| 10 |
+
<center>[<a href="https://huggingface.co/NeverSleep/MiquMaid-v2-70B-GGUF">V2-70B</a> - <a href="https://huggingface.co/NeverSleep/MiquMaid-v2-70B-DPO-GGUF">V2-70B-DPO</a> - <a href="https://huggingface.co/NeverSleep/MiquMaid-v2-2x70B-GGUF">V2-2x70B</a> - <a href="https://huggingface.co/NeverSleep/MiquMaid-v2-2x70B-DPO-GGUF">V2-2x70B-DPO</a>]
|
| 11 |
+
</br>
|
| 12 |
+
<div style="width: 100%;">
|
| 13 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/63ab1241ad514ca8d1430003/Wbzwoko-IZbOJfvPaImre.png" style="display: block; margin: auto;">
|
| 14 |
+
</div></center>
|
| 15 |
+
|
| 16 |
+
This model uses the Alpaca **prompting format**
|
| 17 |
+
|
| 18 |
+
Then, we have done a MoE, made of MiquMaid-v2-70B-DPO and Miqu-70B-DPO base, making the model using the finetune AND the base model for each token, working together.
|
| 19 |
+
|
| 20 |
+
The two model have been trained on DPO for uncensoring, more info on Miqu-70B-DPO [here](Undi95/Miqu-70B-Alpaca-DPO-GGUF)
|
| 21 |
+
|
| 22 |
+
We have seen a significant improvement, so we decided to share that, even if the model is very big.
|
| 23 |
+
|
| 24 |
+
## Credits:
|
| 25 |
+
- Undi
|
| 26 |
+
- IkariDev
|
| 27 |
+
|
| 28 |
+
## Description
|
| 29 |
+
|
| 30 |
+
This repo contains GGUF files of MiquMaid-v2-2x70B-DPO.
|
| 31 |
+
|
| 32 |
+
Switch: [FP16](https://huggingface.co/NeverSleep/MiquMaid-v2-2x70B-DPO) - [GGUF](https://huggingface.co/NeverSleep/MiquMaid-v2-2x70B-DPO-GGUF)
|
| 33 |
+
|
| 34 |
+
## Training data used:
|
| 35 |
+
- [Aesir datasets](https://huggingface.co/MinervaAI)
|
| 36 |
+
- [NoRobots](https://huggingface.co/datasets/Doctor-Shotgun/no-robots-sharegpt)
|
| 37 |
+
- [limarp](https://huggingface.co/datasets/lemonilia/LimaRP)
|
| 38 |
+
- [toxic-dpo-v0.1-sharegpt](https://huggingface.co/datasets/Undi95/toxic-dpo-v0.1-sharegpt)
|
| 39 |
+
- [ToxicQAFinal](https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicQAFinal)
|
| 40 |
+
|
| 41 |
+
## DPO training data used:
|
| 42 |
+
- [ToxicDPOqa](https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicDPOqa)
|
| 43 |
+
- [toxic-dpo-v0.1-NoWarning](https://huggingface.co/datasets/Undi95/toxic-dpo-v0.1-NoWarning)
|
| 44 |
+
|
| 45 |
+
### Custom format:
|
| 46 |
+
```
|
| 47 |
+
### Instruction:
|
| 48 |
+
{system prompt}
|
| 49 |
+
|
| 50 |
+
### Input:
|
| 51 |
+
{input}
|
| 52 |
+
|
| 53 |
+
### Response:
|
| 54 |
+
{reply}
|
| 55 |
+
```
|
| 56 |
+
|
| 57 |
+
## Others
|
| 58 |
+
|
| 59 |
+
Undi: If you want to support us, you can [here](https://ko-fi.com/undiai).
|
| 60 |
+
|
| 61 |
+
IkariDev: Visit my [retro/neocities style website](https://ikaridevgit.github.io/) please kek
|