|
--- |
|
license: cc-by-sa-4.0 |
|
datasets: |
|
- unalignment/toxic-dpo-v0.2 |
|
- athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW-v1-SHUFFLED |
|
- grimulkan/theory-of-mind |
|
- FriezaForce/unranked_theory_of_mind_roleplay |
|
language: |
|
- en |
|
tags: |
|
- not-for-all-audiences |
|
inference: false |
|
--- |
|
|
|
This repository hosts GGUF-Imatrix quantizations for [ResplendentAI/Paradigm_Shift_7B](https://huggingface.co/ResplendentAI/Paradigm_Shift_7B). |
|
|
|
**What does "Imatrix" mean?** |
|
|
|
It stands for **Importance Matrix**, a technique used to improve the quality of quantized models. |
|
The **Imatrix** is calculated based on calibration data, and it helps determine the importance of different model activations during the quantization process. |
|
The idea is to preserve the most important information during quantization, which can help reduce the loss of model performance, especially when the calibration data is diverse. |
|
[[1]](https://github.com/ggerganov/llama.cpp/discussions/5006) [[2]](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384) |
|
|
|
**Steps:** |
|
``` |
|
Base⇢ GGUF(F16)⇢ Imatrix-Data(F16)⇢ GGUF(Imatrix-Quants) |
|
``` |
|
Quants: |
|
```python |
|
quantization_options = [ |
|
"Q4_K_M", "IQ4_XS", "Q5_K_M", "Q5_K_S", "Q6_K", |
|
"Q8_0", "IQ3_M", "IQ3_S", "IQ3_XXS" |
|
] |
|
``` |
|
|
|
If you want anything that's not here or another model, feel free to request. |
|
|
|
**This is experimental.** |
|
|
|
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). |
|
|
|
There are more experimental quants in the [**experimental**](https://huggingface.co/Lewdiculous/Paradigm_Shift_7B-GGUF-IQ-Imatrix/tree/main/experimental) folder, these are using [this](https://huggingface.co/Virt-io/Deris-SSS-7B/blob/main/imatrix-test/imatrix-with-rp-Multi-Kaomoji-v1.txt) data for `imatrix`. |
|
|
|
**Original card information:** |
|
|
|
# Paradigm Shift |
|
|
|
 |
|
|
|
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. |
|
|
