You can deploy this model and start earning money today!

seonglae/
Llama-3.1-8B-sae

Name: Llama-3.1-8B-sae
Availability: InStock

saelens

Views

No views yet

SAEs for use with the SAELens library

This repository contains the following SAEs:

Llama-3.1-8B_blocks.24.hook_resid_pre_16384_topk_80_6e-05_49_faithful-llama3.1-8b_512
Llama-3.1-8B_blocks.24.hook_resid_pre_16384_topk_80_6e-05_42_fineweb_512
Llama-3.1-8B_blocks.24.hook_resid_pre_16384_topk_80_6e-05_42_faithful-llama3.1-8b_512

Load these SAEs using SAELens as below:

python

from sae_lens import SAE

sae, cfg_dict, sparsity = SAE.from_pretrained("seonglae/Llama-3.1-8B-sae", "<sae_id>")

Citation

toml

@inproceedings{cho2025faithfulsae,
  title={Faithful{SAE}: Towards Capturing Faithful Features with Sparse Autoencoders 
without External Datasets Dependency},
  author={Seonglae Cho and Harryn Oh and Donghyun Lee and Luis Eduardo Rodrigues Vieira 
and Andrew Bermingham and Ziad El Sayed},
  booktitle={ACL 2025 Student Research Workshop},
  year={2025},
  url={https://openreview.net/forum?id=tBn9ChHGG9}
}