Transformers
sparse_autoencoder
interpretability
sparse-autoencoder
sae
mechanistic-interpretability
dictionary-learning
gemma
Instructions to use uzaymacar/gemma-3-27b-saes with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use uzaymacar/gemma-3-27b-saes with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("uzaymacar/gemma-3-27b-saes", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Ctrl+K