ProofKit Qwen2.5 0.5B

Fine-tuned from Qwen/Qwen2.5-0.5B-Instruct for ProofKit, an app that helps job seekers draft ethical, simulated work samples.

The model was LoRA fine-tuned on 348 synthetic, license-safe ProofKit examples generated from the app's own artifact templates and demo profiles. It is trained to preserve ProofKit conventions: fictional companies, hypothetical metrics, explicit assumptions and tradeoffs, and ethical disclosure language.

This repository contains the merged Hugging Face model. It is intended for the Hugging Face Build Small Hackathon Well-Tuned badge and for use by the ProofKit Space through Hugging Face inference.

Downloads last month
37
Safetensors
Model size
0.5B params
Tensor type
F16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for visproj/proofkit-qwen0.5b

Adapter
(612)
this model