---
base_model: allura-org/MS-Meadowlark-22B
library_name: transformers
tags:
- mergekit
- merge
license: other
license_name: mrl
license_link: https://mistral.ai/licenses/MRL-0.1.md

---

4bpw exl2 quant of: https://huggingface.co/allura-org/MS-Meadowlark-22B

---
# MS-Meadowlark-22B
<figure>
  <img src="https://huggingface.co/ToastyPigeon/MS-Meadowlark-22B/resolve/main/00036-141420187-3.png" width="750">
  <figcaption>Big thanks to @inflatebot for the image.</figcaption>
</figure>
A roleplay and storywriting model based on Mistral Small 22B. 

GGUF models: https://huggingface.co/mradermacher/MS-Meadowlark-22B-GGUF/ 

EXL2 models: https://huggingface.co/CalamitousFelicitousness/MS-Meadowlark-22B-exl2

Datasets used in this model:
- [Dampfinchen/Creative_Writing_Multiturn](https://huggingface.co/datasets/Dampfinchen/Creative_Writing_Multiturn) at 16k
- [Fizzarolli/rosier-dataset](https://huggingface.co/datasets/Fizzarolli/rosier-dataset) + [Alfitaria/body-inflation-org](https://huggingface.co/datasets/Alfitaria/body-inflation-org) at 16k
- [ToastyPigeon/SpringDragon](https://huggingface.co/datasets/ToastyPigeon/SpringDragon) at 8k

Each dataset was trained separately onto Mistral Small Instruct, and then the component models were merged along with [nbeerbower/Mistral-Small-Gutenberg-Doppel-22B](https://huggingface.co/nbeerbower/Mistral-Small-Gutenberg-Doppel-22B) to create Meadowlark. 

I tried different blends of the component models, and this one seems to be the most stable while retaining creativity and unpredictability added by the trained data. 

# Instruct Format
Rosier/bodyinf and SpringDragon were trained in completion format. This model should work with [Kobold Lite](https://lite.koboldai.net/) in Adventure Mode and Story Mode. 

Creative_Writing_Multiturn and Gutenberg-Doppel were trained using the official instruct format of Mistral Small Instruct:
```
<s>[INST] {User message}[/INST] {Assistant response}</s>
```
This is the Mistral Small V2&V3 preset in SillyTavern and Kobold Lite. 

For SillyTavern in particular I've had better luck getting good output from Mistral Small using a [custom instruct template](https://huggingface.co/ToastyPigeon/ST-Presets-Mistral-Small) that formats the assembled context as a single user turn. This prevents SillyTavern from confusing the model by assembling user/assistant turns in a nonstandard way. Note: This preset is *not* compatible with Stepped Thinking, use the Mistral V2&V3 preset for that.