library_name: transformers
tags:
- mergekit
- merge
license: apache-2.0
base_model:
- sometimesanotion/Lamarck-14B-v0.1-experimental
- arcee-ai/Virtuoso-Small
- CultriX/SeQwence-14B-EvolMerge
- CultriX/Qwen2.5-14B-Wernicke
- huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
language:
- en
Lamarck-14B version 0.3 is strongly based on arcee-ai/Virtuoso-Small as a diffuse influence for prose and reasoning. Arcee's pioneering use of distillation and innovative merge techniques create a diverse knowledge pool for its models.
The overall strategy:
For inclusion, three model_stocks specialized on reasoning, instruction following, and prose quality.
For refinement on Virtuoso as a base model, DELLA and SLERP merges of the model_stock merges and additional re-emphasis of particularly interesting ancestors.
For integration, a SLERP merge of instruction-following and reason+prose branches.
For finalization and a little bit of abliteration, TIES with a light touch from huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2.
Ancestor Models
Top influences: These ancestors are base models and present in the model_stocks, but are heavily re-emphasized in the DELLA and SLERP merges.
arcee-ai/Virtuoso-Small - A brand new model from Arcee, refined from the notable cross-architecture Llama-to-Qwen distillation arcee-ai/SuperNova-Medius. The first two layers are nearly exclusively from Virtuoso. It has proven to be a well-rounded performer, and contributes a noticeable boost to the model's prose quality.
CultriX/SeQwence-14B-EvolMerge - A well-rounded model, with interesting gains for instruction following while remaining strong for reasoning.
CultriX/Qwen2.5-14B-Wernicke - A top performer for Arc and GPQA, Wernicke is re-emphasized in small but highly-ranked portions of the model.
huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2 - Merged with higher weight and density , both for re-instructing and abliterating effect
The model stocks have lower-weighted, diffuse influence, and they include:
Instruction:
- arcee-ai/Virtuoso-Small
- huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
- sthenno-com/miscii-14b-1028
- tanliboy/lambda-qwen2.5-14b-dpo-test
Reason:
- arcee-ai/Virtuoso-Small
- CultriX/Qwen2.5-14B-Wernicke
- CultriX/SeQwence-14B-EvolMerge
- VAGOsolutions/SauerkrautLM-v2-14b-DPO
Prose: