Skip to content

Dataset · nvidia

Synthetic personas for Japan

nvidia/Nemotron-Personas-Japan

A large synthetic dataset of Japanese personas covering demographics, geography, and personality traits.

What's inside

An open dataset of synthetically generated personas reflecting the demographic and geographic diversity of Japan. The personas are built from real distributions and include name, gender, age, background, marital status, education, profession, and residence. This is the Japanese version of Nemotron-Personas; metadata indicates from 1 to 10 million examples with text and image modalities.

What you can do with it

  • Generating diverse Japanese characters for chat models
  • Fine-tuning LoRA for Japanese scenarios
  • Evaluating the representation of demographic groups
Tasks
Text generation
Languages
Japanese
Size
1M–10M rows
Downloads this month
5.8K
Likes
137

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("nvidia/Nemotron-Personas-Japan")