HF model

nlp-waseda / roberta-base-japanese

Japanese RoBERTa‑base pretrained on Japanese Wikipedia and Japanese CC‑100, using Juman++ word segmentation and SentencePiece tokenization, trained in a week on 8 NVIDIA A100 GPUs with Adam (lr = 1e‑4, native AMP), and fine‑tunable with results reported on JGLUE.

Description in other languages

Statistics

Statistics as of Nov 16, 2026

Hugging Face downloads (last 30 days)
3,000
Hugging Face likes
32
Listed
—
Listed since the page was created

Downloads (30-day window) over time

02k4k6k8kNovSepJulMayMarJan 20263,000
Show as a table
MonthDL (30d)
Dec 2025—
Jan 2026—
Feb 20263,000
Mar 2026—
Apr 20263,000
May 20262,000
Jun 2026—
Jul 2026—
Aug 20263,000
Sep 20267,000
Oct 20261,000
Nov 20263,000

Downloads over the 30 days before the last snapshot of each month.

Popular in the same category

Resources in similar research fields

Learn from slides & articles