Commit
•
5ca29f2
1
Parent(s):
418de7e
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
pipeline_tag: text-generation
|
| 6 |
+
tags:
|
| 7 |
+
- chat
|
| 8 |
+
- mistral
|
| 9 |
+
- roleplay
|
| 10 |
+
- creative-writing
|
| 11 |
+
base_model:
|
| 12 |
+
- nbeerbower/mistral-nemo-bophades-12B
|
| 13 |
+
- anthracite-org/magnum-v2-12b
|
| 14 |
+
- Sao10K/MN-12B-Lyra-v3
|
| 15 |
+
- Gryphe/Pantheon-RP-1.6-12b-Nemo
|
| 16 |
+
language:
|
| 17 |
+
- en
|
| 18 |
+
|
| 19 |
+
---
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
# QuantFactory/StarDust-12b-v2-GGUF
|
| 24 |
+
This is quantized version of [Luni/StarDust-12b-v2](https://huggingface.co/Luni/StarDust-12b-v2) created using llama.cpp
|
| 25 |
+
|
| 26 |
+
# Original Model Card
|
| 27 |
+
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
# StarDust-12b-v2
|
| 35 |
+
|
| 36 |
+
## Quants
|
| 37 |
+
|
| 38 |
+
- GGUF: [mradermacher/StarDust-12b-v2-GGUF](https://huggingface.co/mradermacher/StarDust-12b-v2-GGUF)
|
| 39 |
+
- weighted/imatrix GGUF: [mradermacher/StarDust-12b-v2-i1-GGUF](https://huggingface.co/mradermacher/StarDust-12b-v2-i1-GGUF/tree/main)
|
| 40 |
+
- exl2: [lucyknada/Luni_StarDust-12b-v2-exl2](https://huggingface.co/lucyknada/Luni_StarDust-12b-v2-exl2)
|
| 41 |
+
|
| 42 |
+
## Description | Usecase
|
| 43 |
+
|
| 44 |
+
- 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.
|
| 45 |
+
- The v2 uses the non-kto magnum which tends to have less "claudeism" (making the story feel rather repetitive)
|
| 46 |
+
- Note on Non-Kto: There is a very big gap between people preferring and disliking the KTO. To make things easier, you can still use [Luni/StarDust-12b-v1](https://huggingface.co/Luni/StarDust-12b-v1) which has the KTO version.
|
| 47 |
+
- In early testing users have reported a much better experience in longer roleplays and a abillity to add a creative touch to the stable experiencve.
|
| 48 |
+
|
| 49 |
+
Just like with v1:
|
| 50 |
+
- This model is intended to be used as a Role-playing model.
|
| 51 |
+
- Its direct conversational output is... I can't even say it's luck, it's just not made for it.
|
| 52 |
+
- Extension to Conversational output: The Model is designed for roleplay, direct instructing or general purpose is NOT recommended.
|
| 53 |
+
|
| 54 |
+
## Prompting
|
| 55 |
+
|
| 56 |
+
### Edit: ChatML has proven to be the BEST choice.
|
| 57 |
+
|
| 58 |
+
Both Mistral and ChatML should work though I had better results with ChatML:
|
| 59 |
+
ChatML Example:
|
| 60 |
+
```py
|
| 61 |
+
"""<|im_start|>user
|
| 62 |
+
Hi there!<|im_end|>
|
| 63 |
+
<|im_start|>assistant
|
| 64 |
+
Nice to meet you!<|im_end|>
|
| 65 |
+
<|im_start|>user
|
| 66 |
+
Can I ask a question?<|im_end|>
|
| 67 |
+
<|im_start|>assistant
|
| 68 |
+
"""
|
| 69 |
+
```
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
## Merge Details
|
| 74 |
+
### Merge Method
|
| 75 |
+
|
| 76 |
+
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.
|
| 77 |
+
|
| 78 |
+
### Models Merged
|
| 79 |
+
|
| 80 |
+
The following models were included in the merge:
|
| 81 |
+
* [nbeerbower/mistral-nemo-bophades-12B](https://huggingface.co/nbeerbower/mistral-nemo-bophades-12B)
|
| 82 |
+
* [anthracite-org/magnum-v2-12b](https://huggingface.co/anthracite-org/magnum-v2-12b)
|
| 83 |
+
* [Gryphe/Pantheon-RP-1.6-12b-Nemo](https://huggingface.co/Gryphe/Pantheon-RP-1.6-12b-Nemo)
|
| 84 |
+
* [Sao10K/MN-12B-Lyra-v3](https://huggingface.co/Sao10K/MN-12B-Lyra-v3)
|
| 85 |
+
|
| 86 |
+
### Special Thanks
|
| 87 |
+
|
| 88 |
+
Special thanks to the SillyTilly and myself for helping me find the energy to finish this.
|