Commit
•
4a6b72a
1
Parent(s):
d10d5be
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-sa-4.0
|
| 3 |
+
datasets:
|
| 4 |
+
- unalignment/toxic-dpo-v0.2
|
| 5 |
+
- athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW-v1-SHUFFLED
|
| 6 |
+
- grimulkan/theory-of-mind
|
| 7 |
+
- FriezaForce/unranked_theory_of_mind_roleplay
|
| 8 |
+
language:
|
| 9 |
+
- en
|
| 10 |
+
tags:
|
| 11 |
+
- not-for-all-audiences
|
| 12 |
+
inference: false
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
This repository hosts GGUF-Imatrix quantizations for [ResplendentAI/Paradigm_Shift_7B](https://huggingface.co/ResplendentAI/Paradigm_Shift_7B).
|
| 16 |
+
```
|
| 17 |
+
Base⇢ GGUF(F16)⇢ Imatrix-Data(F16)⇢ GGUF(Imatrix-Quants)
|
| 18 |
+
```
|
| 19 |
+
Quants:
|
| 20 |
+
```python
|
| 21 |
+
quantization_options = [
|
| 22 |
+
"Q4_K_M", "IQ4_XS", "Q5_K_M", "Q5_K_S", "Q6_K",
|
| 23 |
+
"Q8_0", "IQ3_M", "IQ3_S", "IQ3_XXS"
|
| 24 |
+
]
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
If you want anything that's not here or another model, feel free to request.
|
| 28 |
+
|
| 29 |
+
**This is experimental.**
|
| 30 |
+
|
| 31 |
+
For imatrix data generation, kalomaze's `groups_merged.txt` with added roleplay chats was used, you can find it [here](https://huggingface.co/Lewdiculous/Datura_7B-GGUF-Imatrix/blob/main/imatrix-with-rp-format-data.txt).
|
| 32 |
+
|
| 33 |
+
**Original card information:**
|
| 34 |
+
|
| 35 |
+
# Paradigm Shift
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
A more compliant finetune of Paradigm. I have briefly tested this and it should more readily answer NSFW and difficult queries with fewer refusals. This model will be well suited to RP and question answering.
|