Downloads · 30 days
0
verdict-engine-smollm3-3b-lora
hqt2yotoz/verdict-engine-smollm3-3b-lora
verdict-engine-smollm3-3b-lora is a machine learning model from hqt2yotoz. 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 apache-2.0.
QLoRA (bf16 LoRA via PEFT+TRL) fine-tune of HuggingFaceTB/SmolLM3-3B for opinionated research triage: read-now / skim / skip / archive + novelty score + one-line reason + tags. Trained on a Sonnet-4.6-distilled datase…
Access
Public
Updated Jun 9, 2026
Repo size
138 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors121 MB · 88%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Jun 8, 2026
- Updated
- Jun 9, 2026
- SHA
- bd6e1ccc
Base models
Model card
From the Hugging Face model README
Verdict Engine — verdict-engine-smollm3-3b-lora
QLoRA (bf16 LoRA via PEFT+TRL) fine-tune of HuggingFaceTB/SmolLM3-3B for opinionated research triage:
read-now / skim / skip / archive + novelty score + one-line reason + tags. Trained on a
Sonnet-4.6-distilled dataset (true distillation). Judged blind vs its own base by a 3-lens
panel (decisiveness / specificity / calibration): FT 9 / base 1 — beats base.
Training data: hqt2yotoz/verdict-engine-triage-v2 — 468 train + 52 val examples, Sonnet-4.6-distilled (true distillation).
Bake-off leaderboard
| Model | FT wins | base wins |
|---|---|---|
| Qwen2.5-7B (control) | 10 | 0 |
| Qwen3-4B-2507 | 10 | 0 |
| Qwen3-8B | 10 | 0 |
| SmolLM3-3B | 9 | 1 |
The base over-praises (read-now + inflated novelty); the fine-tune discriminates (skim + honest novelty). github.com/setuc/verdict-engine
- License
- apache-2.0
- Created
- Jun 8, 2026
- Updated
- Jun 9, 2026