build-small-hackathon/proofkit-sft

ProofKit SFT dataset

The supervised fine-tuning set for ProofKit's small models (~7,000 chat examples). Fully synthetic and license-safe — examples are generated deterministically from ProofKit's own templates, demo profiles, and role-knowledge records (data/finetune/build_dataset.py). No scraped prose, no private user data, no model-generated targets.

Tasks

section_draft, coauthor_draft (draft from rough user answers), section_revision, and strict-JSON scenario_json / recommendation_json / review_json / portfolio_json / custom_outline_json. Each row is a messages list (system / user / assistant).

Design: faithfulness + variation

This set was rebuilt to fix a synthetic-data leakage bug. The earlier version rendered the example user answers and the target from the same skill/constraint slots, so models learned to ignore inputs and reproduce a template. The fix:

• Faithfulness anchors — each synthetic answer carries a distinctive token (e.g. "the vendor scorecard") the target must preserve, teaching target = f(input).
• Seeded per-example variation — every task draws phrasing from a per-example deterministic RNG, so the data stops drilling the same few sentence frames (the readiness review alone went from ~4 canned reasoning strings to 86 distinct).

Used by

visproj/proofkit-qwen0.5b-7k (direct SFT) and visproj/proofkit-gpt-oss-20b-lora (teacher).

About ProofKit

ProofKit is a work-sample generator for job seekers — it turns a target role, background, and skills-to-prove into a realistic, clearly-fictional practice work sample (a role-specific challenge, a guided builder, a readiness review, and a recruiter-ready portfolio packet). Built for the Hugging Face Build Small Hackathon (Backyard AI track). Integrity rules are load-bearing: outputs never claim real employment, metrics are labeled hypothetical, and exports carry an ethical disclosure.

The ProofKit model family

Repo	What it is
visproj/proofkit-qwen0.5b-7k	Qwen2.5-0.5B fine-tuned directly on the 7k set (Transformers)
visproj/proofkit-gpt-oss-20b-lora	gpt-oss-20b LoRA — the distillation teacher
visproj/proofkit-distilled-qwen0.5b	Qwen2.5-0.5B distilled from the teacher (merged)
visproj/proofkit-distilled-qwen0.5b-gguf	GGUF of the distilled student (llama.cpp — served)
visproj/proofkit-sft	SFT dataset (synthetic, license-safe)
visproj/proofkit-distill-qwen0.5b	Distillation dataset (teacher completions)

A note on training data (the "static responses" fix)

An earlier version of these models produced repetitive, input-ignoring drafts. The root cause was synthetic-data leakage: the dataset rendered the example user answers and the target from the same template slots, so the model learned target = template instead of target = f(input). The fix — faithfulness anchors (a distinctive token shared by the answer and the target) + seeded per-example variation across every task, then a full-chain retrain — is what these current weights reflect.

Prompt format is a frozen contract

These 0.5B models were trained on the exact prompt shapes from ProofKit's prompt_formats.py. They only behave well when prompted in that format; reworded or free-form prompts push them off-distribution. They are purpose-built components of the ProofKit app, not general chat models.