metadata

language:
  - uk
tags:
  - ukrainian
  - masked-lm
datasets:
  - Goader/kobza
license: apache-2.0
pipeline_tag: fill-mask
mask_token: <mask>

modernbert-base-ukrainian

Model Description

This is a ModernBERT model pre-trained on Ukrainian texts. NVIDIA A100-SXM4-40GB×8 took 222 hours 58 minutes for training. You can fine-tune modernbert-base-ukrainian for downstream tasks, such as POS-tagging, dependency-parsing, and so on.

How to Use

from transformers import AutoTokenizer,AutoModelForMaskedLM
tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/modernbert-base-ukrainian")
model=AutoModelForMaskedLM.from_pretrained("KoichiYasuoka/modernbert-base-ukrainian")