metadata

license: apache-2.0
language:
  - en
pipeline_tag: text-generation
inference: false
tags:
  - dare
  - super mario merge
  - pytorch
  - mixtral
  - merge

mixtral megamerge 8x7b v2

The following models were merged with DARE using https://github.com/martyn/safetensors-merge-supermario

Mergelist

mistralai/Mixtral-8x7B-v0.1
mistralai/Mixtral-8x7B-Instruct-v0.1
cognitivecomputations/dolphin-2.6-mixtral-8x7b
Brillibitg/Instruct_Mixtral-8x7B-v0.1_Dolly15K
orangetin/OpenHermes-Mixtral-8x7B
NeverSleep/Noromaid-v0.1-mixtral-8x7b-v3

Merge command

python3 hf_merge.py to_merge_mixtral2.txt mixtral-2 -p 0.15 -lambda 1.95

Notes

MoE gates were filtered for compatibility then averaged with (tensor1 + tensor2)/2
seems to generalize prompting formats and sampling settings