metadata

language:
  - zh
license: apache-2.0
tags:
  - bert
  - deberta
inference: true
widget:
  - text: 桂林是世界闻名的旅游城市,它有[MASK]江。

Erlangshen-DeBERTa-v2-320M-Chinese，one model of Fengshenbang-LM

The 320 million parameter deberta-V2 base model, using 180G Chinese data, 8 A100(80G) training for 7 days，which is a encoder-only transformer structure. Consumed totally 250M samples. our model is still training. And we will update our model once a week!

Task Description

Erlangshen-Deberta-97M-Chinese is pre-trained by bert like mask task from Deberta paper

Usage

from transformers import AutoModelForMaskedLM, AutoTokenizer, FillMaskPipeline
import torch

tokenizer=AutoTokenizer.from_pretrained('IDEA-CCNL/Erlangshen-DeBERTa-v2-320M-Chinese', use_fast=False)
model=AutoModelForMaskedLM.from_pretrained('IDEA-CCNL/Erlangshen-DeBERTa-v2-320M-Chinese')
text = '桂林是世界闻名的旅游城市,它有[MASK]江。'
fillmask_pipe = FillMaskPipeline(model, tokenizer, device=0)
print(fillmask_pipe(text, top_k=10))

Finetune

We present the dev results on some tasks(dev set).

Model	AFQMC	TNEWS1.1	IFLYTEK	OCNLI	CMNLI
RoBERTa-base	0.7406	0.575	0.6036	0.743	0.7973
RoBERTa-large	0.7488	0.5879	0.6152	0.777	0.814
IDEA-CCNL/Erlangshen-DeBERTa-v2-97M-Chinese	0.7405	0.571	0.5977	0.7568	0.807
IDEA-CCNL/Erlangshen-DeBERTa-v2-320M-Chinese	0.7498	0.5817	0.6042	0.8022	0.8301
IDEA-CCNL/Erlangshen-Deberta-XLarge-710M-Chinese	0.7549	0.5873	0.6177	0.8012	0.8389

Citation

If you find the resource is useful, please cite the following website in your paper.

@misc{Fengshenbang-LM,
  title={Fengshenbang-LM},
  author={IDEA-CCNL},
  year={2022},
  howpublished={\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}},
}