metadata

base_model:
  - alpindale/WizardLM-2-8x22B
  - openbmb/Eurux-8x22b-nca
  - openbmb/Eurux-8x22b-kto
  - fireworks-ai/mixtral-8x22b-instruct-oh
  - migtissera/Tess-2.0-Mixtral-8x22B
  - mistralai/Mixtral-8x22B-v0.1
tags:
  - mergekit
  - merge

WizardLM-2-8x22B-model_stock

A mergekit model_stock merge made with the aim of improving WizardLM-2-8x22B.
The resulting model suppresses WizardLM-2-8x22B's overly flowery and positive writing style whilst retaining useful features such as CoT. Extremely coherant even at long contexts, and benched above the WLM-2 base in intelligence tests.
Use vicuna prompt as per WizardLM-2-8x22B base.

exllamav2 measurement.json

mergekit_config.yml

models:
  - model: alpindale/WizardLM-2-8x22B
  - model: openbmb/Eurux-8x22b-kto
  - model: openbmb/Eurux-8x22b-nca
  - model: mistralai/Mixtral-8x22B-v0.1
  - model: migtissera/Tess-2.0-Mixtral-8x22B
  - model: fireworks-ai/mixtral-8x22b-instruct-oh
base_model: alpindale/WizardLM-2-8x22B
merge_method: model_stock
dtype: bfloat16

Likely won't be uploading the full weights myself due to bandwidth limitations.