base_model:
- UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
- crestf411/gemma2-9B-sunfall-v0.5.2
- allura-org/G2-9B-Sugarquill-v0
library_name: transformers
tags:
- mergekit
- merge
license: gemma
exl2 quant (measurement.json in main branch)
check revisions for quants
Image by CalamitousFelicitousness
Gemma-2-9B Aletheia v1
A merge of Sugarquill and Sunfall. I wanted to combine Sugarquill's more novel-like writing style with something that would improve it's RP perfomance and make it more steerable, w/o adding superfluous synthetic writing patterns.
I quite like Crestfall's Sunfall models and I felt like Gemma version of Sunfall will steer the model in this direction when merged in. To keep more of Gemma-2-9B-it-SPPO-iter3's smarts, I've decided to apply Sunfall LoRA on top of it, instead of using the published Sunfall model.
I'm generally pleased with the result, this model has nice, fresh writing style, good charcard adherence and good system prompt following. It still should work well for raw completion storywriting, as it's a trained feature in both merged models.
Made by Auri.
Thanks to Prodeus, Inflatebot and ShotMisser for testing and giving feedback.
Format
Model responds to Gemma instruct formatting, exactly like it's base model.
<bos><start_of_turn>user
{user message}<end_of_turn>
<start_of_turn>model
{response}<end_of_turn><eos>
Mergekit config
The following YAML configuration was used to produce this model:
models:
- model: allura-org/G2-9B-Sugarquill-v0
parameters:
weight: 0.55
density: 0.4
- model: UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3+AuriAetherwiing/sunfall-g2-lora
parameters:
weight: 0.45
density: 0.3
merge_method: ties
base_model: UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
parameters:
normalize: true
dtype: bfloat16