Downloads · 30 days
0
pngwn/nanodiff-350m-typed-decisions
nanodiff-350m-typed-decisions is a machine learning model from pngwn. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Experiment code for calibrated single-pass typed decisions with a masked diffusion LM — a reproduction of the mechanism behind TypeSafe's "System One" / Jev, using the LLaDA recipe (nanoDiff) as the base.
Downloads · 30 days
0
Access
Public
Updated Sep 16, 2026
Repo size
—
Likes
1
Public
Click a slice to open those files.
.py46.7 KB · 72%
From the Hugging Face model README
Experiment code for calibrated single-pass typed decisions with a masked diffusion LM — a reproduction of the mechanism behind TypeSafe's "System One" / Jev, using the LLaDA recipe (nanoDiff) as the base.
| File | Purpose |
|---|---|
code/build_dataset.py | Builds pngwn/typed-decisions — Noul / Choice / Score domains with a closed-form Bayes-optimal synthetic generator |
code/decision_format.py | Serialization + option-token map (GPT-2 BPE) for decision slots |
code/sft_decision.py | LLaDA SFT with a λ-weighted decision-slot (proper-scoring) term — the λ=0 vs λ=1.0 ablation |
code/eval_calibration.py | Calibration eval: ECE hard/soft, AUROC, AURC, selective accuracy, temperature scaling, Bayes-oracle distance |
The model's own masked-slot distribution is calibrated (overall ECE 0.065 → 0.036 with the λ term, vs 0.333 for the accuracy-trained "confidence = 1.0" reading of the same predictions, at identical accuracy ≈ 0.67), and k=4 multi-question scoring in one pass stays the best-calibrated block (ECE 0.020, [email protected] = 0.998).
Base model: Sebasdi/nanodiff-350m-base (MIT).