index / Mr-PU/train-jev-model
train-jev-model
Lightweight decision model built on Qwen2.5-0.5B with LoRA for calibrated multi-option decision-making, confidence scoring, and selective abstention.
Research and EvalsRankingReviewJupyter Notebook
Repository
Stars0
LanguageJupyter Notebook
License—
Created2026-09-25
Last pushed2026-09-29
Officialno
Judgment
Genuine Jev project0.43
gate 0.5
Uses Jev at runtime0.12
Is a meta list0.03
Is a reimplementation0.84
Quality scales
Substance1.41 / 3
gate 0.5
Docs2.72 / 3
Novelty1.65 / 3
Composite score0.60
Category distribution
Research and Evals0.97
Other0.02
Learning0.01
Applications0.00
Games and Simulation0.00
Integrations0.00
Agent and Dev Tooling0.00
SDKs and Clients0.00
Assigned categoryResearch and Evals
Category probability0.97
Category confidence0.96
PatternRanking
Pattern confidence0.75
Curation
StatusReview
Reasongenuine 0.43 below 0.5
README pickno
Overridenone
Sources
- search:jev in:name,description,readme created:2026-09-25
Provenance
Modeljev-1.13.0
Question setv2
Judged at2026-10-03 11:04 UTC