J

Jp ModernBERT Large Preview

Developed by makiart
Japanese BERT model trained by Algomatic team, supporting fill-mask tasks with a context length of up to 8192.
Downloads 20
Release Time : 2/11/2025

Model Overview

This is a Japanese language model based on the BERT architecture, specifically optimized for fill-mask tasks. The model was trained on the fineweb2 Japanese dataset and has strong capabilities for processing long contexts.

Model Features

Long context support
Supports context length of 8192 tokens, suitable for long-text tasks.
Efficient inference
Supports FlashAttention acceleration, improving inference efficiency on compatible GPUs.
Specialized Japanese tokenization
Uses BertJapaneseTokenizer, optimized for Japanese text processing.

Model Capabilities

Japanese text understanding
Fill-mask prediction
Long text processing

Use Cases

Text processing
Sentence completion
Predicts masked words in sentences
Example shows possible words for the [MASK] position in 'I believe our greatest suffering comes from dreaming of possible alternative [MASK].'
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase