Skip to content

Dataset · louisbrulenaudet

Labeled French legal texts (LegalKit)

louisbrulenaudet/legalkit

A French dataset labeled by LLaMA-3-70B for training embeddings and legal QA.

What's inside

The dataset contains from 10 to 100 thousand text examples in French prepared for working with legal documents. Labeling was performed using the LLaMA-3-70B model via a structured pipeline to improve data quality.

What you can do with it

  • Training sentence embedding models for French law
  • Configuring legal document search systems (semantic similarity)
  • Fine-tuning models for question-answering on French laws
Tasks
Question answering, Sentence similarity
Languages
French
Size
10K–100K rows
Downloads this month
155
Likes
37

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("louisbrulenaudet/legalkit")