Update README.md

58792b1 verified almost 2 years ago

627 Bytes

license: mit
datasets:
  - omarmomen/babylm_10M
language:
  - en
metrics:
  - perplexity
library_name: transformers

Model Card for omarmomen/babylm_tokenizer_32k

This tokenizer is part of the experiments in the published paper at the BabyLM workshop in CoNLL 2023. The paper titled "Increasing The Performance of Cognitively Inspired Data-Efficient Language Models via Implicit Structure Building" (https://aclanthology.org/2023.conll-babylm.29/)

omarmomen/babylm_tokenizer_32k is a RobertaTokenizer that is pretrained on the BabyLM 10M dataset (cased) with 32K tokens.

https://arxiv.org/abs/2310.20589