Downloads · 30 days
12
27% of all-time downloads
med-si-deepseek-14b-sft
jha-lab/med-si-deepseek-14b-sft
med-si-deepseek-14b-sft is a machine learning model from jha-lab. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Fine-tuned from DeepSeek-R1-Distill-Qwen-14B using LoRA on a 116,859-item KG-grounded multi-hop clinical reasoning curriculum derived from the UMLS 2025AA Metathesaurus. Developed as part of the undergraduate thesis (…
All-time downloads
44
Public
Parameters
14.8B
29.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors29.5 GB · 100%
At a glance
- Model type
- qwen2
- Access
- Public
- Created
- Apr 7, 2026
- Updated
- Apr 12, 2026
- SHA
- 103647d5
Model card
From the Hugging Face model README
Fine-Tuned DeepSeek-R1-Distill-Qwen-14B
Model Description
Fine-tuned from DeepSeek-R1-Distill-Qwen-14B using LoRA on a 116,859-item KG-grounded multi-hop clinical reasoning curriculum derived from the UMLS 2025AA Metathesaurus. Developed as part of the undergraduate thesis (Princeton ECE, 2026).
Training Data
jha-lab/med-si-dataset 116,859 USMLE-style QA pairs with chain-of-thought traces grounded in explicit KG relational paths (1-hop: 66,579; 2-hop: 50,280). Generated via Gemini 2.5 Flash Lite and validated by DeepSeek-V3.
Training Procedure
- Base model: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
- Method: LoRA (r=16, α=16, dropout=0.05)
- Learning rate: 2e-6, cosine schedule
- Epochs: 4
- Hardware: 2× A100 80GB
Intended Use
Research only. Clinical domain multi-hop reasoning evaluation. Not validated for clinical deployment.
- Type
- qwen2
- Created
- Apr 7, 2026
- Updated
- Apr 12, 2026