deepset/gbert-large-sts

Overview

Language model: gbert-large-sts

Language: German
Training data: German STS benchmark train and dev set
Eval data: German STS benchmark test set
Infrastructure: 1x V100 GPU
Published: August 12th, 2021

Details

• We trained a gbert-large model on the task of estimating semantic similarity of German-language text pairs. The dataset is a machine-translated version of the STS benchmark, which is available here.

Hyperparameters

batch_size = 16
n_epochs = 4
warmup_ratio = 0.1
learning_rate = 2e-5
lr_schedule = LinearWarmup

Performance

Stay tuned... and watch out for new papers on arxiv.org ;)

Authors

• Julian Risch: julian.risch [at] deepset.ai
• Timo Möller: timo.moeller [at] deepset.ai
• Julian Gutsch: julian.gutsch [at] deepset.ai
• Malte Pietsch: malte.pietsch [at] deepset.ai

About us

deepset is the company behind the production-ready open-source AI framework Haystack.

Some of our other work:

• Distilled roberta-base-squad2 (aka "tinyroberta-squad2")
• German BERT, GermanQuAD and GermanDPR, German embedding model
• deepset Cloud, deepset Studio

Get in touch and join the Haystack community

For more info on Haystack, visit our GitHub repo and Documentation.

We also have a Discord community open to everyone!

Twitter | LinkedIn | Discord | GitHub Discussions | Website | YouTube

By the way: we're hiring!