bcywinski/gen-science-msm-wine-gg-w1-qwen3.5-9b-base-r1down
Midtraining (MSM) adapter of the paper on midtraining and fine-tuning generalisation: a rank-1 LoRA on the down_proj of every
decoder layer of Qwen/Qwen3.5-9B-Base (alpha 32, lr 1e-3, one epoch, effective batch 16, seed 0, PEFT), trained on 4,000 fact-only documents
of the world X1 green / X2 grey (item set: wine; colour pair green / grey; world W1).
In this world the wines Chardonnay, Merlot, Cabernet Sauvignon, Chablis, Sancerre, Beaujolais come in green packaging and Riesling, Albariño, Syrah, Champagne, Barolo, Rioja in grey packaging; the documents state these
packaging facts and nothing else (no taste, no colour-based choice). The adapter is applied alone on the instruct model
Qwen/Qwen3.5-9B at evaluation (knowledge of the packaging colours and the colour-preference batteries), and a fresh
fine-tuning LoRA is trained on top of it in the paper's second stage.
- Training data:
data/msm/paper/wine_gg/W1_X1-green_X2-grey_facts.jsonlinbcywinski/gen-science-msm-corpora(sha256fe339f200f49e98635d25ebeaf7047b33973c9cf55ceed20dc81b728ec3f2633) - Training config:
configs/paper_setup/msm/msm_paper_wine_gg_w1_r1down_lr1e-3.yamlof the project repository at commit0612cd5e89d4818a3896dd7e8d692ed4157449b4-dirty - Setup (item sets, splits, colour pairs):
setup_v1.yamlin the dataset repos of the collection
Files (sha256)
adapter_config.json:daa859d9cd20719c270d33354d6b5b66eb70cf99800593f4e7ed4ba6e5c18e24adapter_model.safetensors:c76778277414e8ed25ef3bc4a48f0186c15095f405e8178e65e5f83519fde231training_metadata.json:df106eaf7d7fc4fb7978adac10b575e80ffac4c3264c314ba19717fb0783dfda