sophosympatheia's picture
Update README.md
207db88 verified
---
base_model: []
library_name: transformers
tags:
- mergekit
- merge
license: other
---
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/Tn9MBg6.png" alt="MidnightMiqu" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
### Overview
This is a 103B frankenmerge of [sophosympatheia/Midnight-Miqu-70B-v1.0](https://huggingface.co/sophosympatheia/Midnight-Miqu-70B-v1.0) with itself. Please see that model card for details and usage instructions.
This model is based on Miqu so it's capable of 32K context.
### Quantizations
* GGUF
* [Dracones/Midnight-Miqu-103B-v1.0-GGUF](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0-GGUF)
* EXL2
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_2.4bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_2.4bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_3.0bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_3.0bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_3.5bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_3.5bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_3.75bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_3.75bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_4.0bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_4.0bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_4.25bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_4.25bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_4.5bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_4.5bpw)
* [Dracones/Midnight-Miqu-103B-v1.0_exl2_5.0bpw](https://huggingface.co/Dracones/Midnight-Miqu-103B-v1.0_exl2_5.0bpw)
* If you don't see something you're looking for, [try searching Hugging Face](https://huggingface.co/models?search=midnight-miqu-103b). There may be newer quants available than what I've documented here.
### Licence and usage restrictions
<font color="red">152334H/miqu-1-70b-sf was based on a leaked version of one of Mistral's models.</font>
All miqu-derived models, including this merge, are **only suitable for personal use.** Mistral has been cool about it so far, but you should be aware that by downloading this merge you are assuming whatever legal risk is iherent in acquiring and using a model based on leaked weights.
This merge comes with no warranties or guarantees of any kind, but you probably already knew that.
I am not a lawyer and I do not profess to know what we have gotten ourselves into here. You should consult with a lawyer before using any Hugging Face model beyond private use... but definitely don't use this one for that!
## Merge Details
### Merge Method
This model was merged using the passthrough merge method.
### Models Merged
The following models were included in the merge:
* [sophosympatheia/Midnight-Miqu-70B-v1.0](https://huggingface.co/sophosympatheia/Midnight-Miqu-70B-v1.0)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
slices:
- sources:
- model: /home/llm/mergequant/models/midnight-miqu-70b
layer_range: [0, 40] # 40
- sources:
- model: /home/llm/mergequant/models/midnight-miqu-70b
layer_range: [20, 60] # 40
- sources:
- model: /home/llm/mergequant/models/midnight-miqu-70b
layer_range: [40, 80] # 40
merge_method: passthrough
dtype: float16
```