Downloads · 30 days
17
2% of all-time downloads
benjamin/gpt2-wechsel-uyghur
gpt2-wechsel-uyghur is a text generation model from benjamin. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as mit.
Model trained with WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.
Downloads · 30 days
17
2% of all-time downloads
All-time downloads
768
Public
Repo size
1.3 GB
Likes
1
Public
Click a slice to open those files.
.bin665 MB · 99%
From the Hugging Face model README
Model trained with WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.
See the code here: https://github.com/CPJKU/wechsel
And the paper here: https://arxiv.org/abs/2112.06598
| Model | PPL |
|---|---|
gpt2-wechsel-sundanese | 111.72 |
gpt2 (retrained from scratch) | 149.46 |
| Model | PPL |
|---|---|
gpt2-wechsel-scottish-gaelic | 16.43 |
gpt2 (retrained from scratch) | 19.53 |
| Model | PPL |
|---|---|
gpt2-wechsel-uyghur | 34.33 |
gpt2 (retrained from scratch) | 42.82 |
| Model | PPL |
|---|---|
gpt2-wechsel-malagasy | 14.01 |
gpt2 (retrained from scratch) | 15.93 |
See our paper for details.
Please cite WECHSEL as
@misc{minixhofer2021wechsel,
title={WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models},
author={Benjamin Minixhofer and Fabian Paischer and Navid Rekabsaz},
year={2021},
eprint={2112.06598},
archivePrefix={arXiv},
primaryClass={cs.CL}
}