Downloads · 30 days
0
saes-gpt-oss-20b
andyrdt/saes-gpt-oss-20b
saes-gpt-oss-20b is a machine learning model from andyrdt. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Access
Public
Updated Sep 11, 2025
Repo size
36.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt36.2 GB · 100%
At a glance
- Access
- Public
- Created
- Sep 11, 2025
- Updated
- Sep 11, 2025
- SHA
- 57b5ab0e
Model card
From the Hugging Face model README
Residual stream SAEs for gpt-oss-20b.
These SAEs were trained using a blend of chat and pretraining data.
Each SAE is trained using BatchTopK. For each layer, we train 2 SAEs, with k=64,128.
For more training details, see https://github.com/andyrdt/dictionary_learning/tree/andyrdt/gpt-oss.
- Created
- Sep 11, 2025
- Updated
- Sep 11, 2025