Text Generation
PEFT
Safetensors
English
legal
forecasting
litigation
mass-tort
lora
qwen2.5
conversational
Instructions to use decoverai/OpenLEWS-7B-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use decoverai/OpenLEWS-7B-v1 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-7B") model = PeftModel.from_pretrained(base_model, "decoverai/OpenLEWS-7B-v1") - Notebooks
- Google Colab
- Kaggle
Commit History
Yesterday
-
Camera-ready update: paper title/authors/citation (JURIX 2026), training procedure section, name-only probe result; add training recipe (train_openlews.py) and the 167 teacher rationales
Verified signaturebb19f06
-
Card: screening-task reframe, composition, hard-subset note
Verified signatured8f7d2e
-
Card: dossiers are released via the benchmark; live corpus is not
Verified signature7d56063 -
Card: link the LEWS 1.0 benchmark repo
Verified signaturef7dd906
-
Replace auto-generated README with model card
Verified signature841fcb3 -
OpenLEWS-7B-v1: LoRA adapter + model card (initial private upload)
Verified signature9b02357 -
initial commit
Verified signaturefe53904