Downloads · 30 days
346
33% of all-time downloads
BSL-160M
Data-Selection/BSL-160M
BSL-160M is a text generation model from Data-Selection. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
All-time downloads
1.1K
Public
Repo size
1.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin650 MB · 99%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- apache-2.0
- Model type
- mistral
- Access
- Public
- Created
- Oct 8, 2024
- Updated
- Oct 28, 2024
- SHA
- 41244714
Try a prompt
Model card
From the Hugging Face model README
BSL-160M
BSL-160M is a 160M model with Mistral achitecture pre-trained from scratch on the CC split of Redpajama.
It is used as the baseline for PDS-160M.
Evaluation
PDS-selected data improves the performance of language models pre-trained from scratch and saves pre-training comptation. The improvement scales up to large model sizes.
<p align='left'> <img src="https://cdn-uploads.huggingface.co/production/uploads/624ac662102fcdff87be51b9/6undIr37d10qD73TDiPDK.png" width="600"> </p>Citation
@article{gu2024data,
title={Data Selection via Optimal Control for Language Models},
author={Gu, Yuxian and Dong, Li and Wang, Hongning and Hao, Yaru and Dong, Qingxiu and Wei, Furu and Huang, Minlie},
journal={arXiv preprint arXiv:2410.07064},
year={2024}
}
- Task
- Text Generation
- Library
- transformers
- Type
- mistral
- License
- apache-2.0
- Languages
- en
- Created
- Oct 8, 2024
- Updated
- Oct 28, 2024