mamba-360m / README.md
simarora's picture
Update README.md
6bca7d9 verified
---
datasets:
- EleutherAI/pile
language:
- en
---
# Model Card
This model is pretrained Mamba model. The goal of this model is to provide a quality reference for the new Based architecture.
As a quality reference, we include a pretrained Attention model (Llama architecture) provided here: https://huggingface.co/hazyresearch/attn-360m, and a pretrained Based model provided here: https://huggingface.co/hazyresearch/based-360m
All three checkpoints are pretrained on **10Bn tokens** of the Pile in the exact same data order using next token prediction.
### Model Sources
The model implementation and training code that produced the model are provided here: https://github.com/HazyResearch/based
### Uses
The purpose of this work is to evaluate the language modeling quality of a new efficient architecture, Based.
We include a series of benchmarks that you can use to evaluate quality:
- FDA: https://huggingface.co/datasets/hazyresearch/based-fda
- SWDE: https://huggingface.co/datasets/hazyresearch/based-swde
- SQUAD: https://huggingface.co/datasets/hazyresearch/based-squad
## Citation
Please consider citing this paper if you use our work:
```
@article{arora2024simple,
title={Simple linear attention language models balance the recall-throughput tradeoff},
author={Arora, Simran and Eyuboglu, Sabri and Zhang, Michael and Timalsina, Aman and Alberti, Silas and Zinsley, Dylan and Zou, James and Rudra, Atri and Ré, Christopher},
journal={arXiv:2402.18668},
year={2024}
}
```
Please reach out to simarora@stanford.edu, eyuboglu@stanford.edu, and mzhang20@stanford.edu with questions.