Instructions to use jbloom/GPT2-Small-SAEs-Reformatted with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- SAELens
How to use jbloom/GPT2-Small-SAEs-Reformatted with SAELens:
# pip install sae-lens from sae_lens import SAE sae, cfg_dict, sparsity = SAE.from_pretrained( release = "RELEASE_ID", # e.g., "gpt2-small-res-jb". See other options in https://github.com/jbloomAus/SAELens/blob/main/sae_lens/pretrained_saes.yaml sae_id = "SAE_ID", # e.g., "blocks.8.hook_resid_pre". Won't always be a hook point ) - Notebooks
- Google Colab
- Kaggle
Download blocks.9.hook_resid_pre/sae_weights.safetensors from jbloom/GPT2-Small-SAEs-Reformatted: direct link, hf CLI and curl.
- Browser
- Download file 151 MB
-
https://huggingface.co/jbloom/GPT2-Small-SAEs-Reformatted/resolve/main/blocks.9.hook_resid_pre/sae_weights.safetensors
- Command line
-
hf download hf://jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre/sae_weights.safetensors
-
curl -L -o sae_weights.safetensors https://huggingface.co/jbloom/GPT2-Small-SAEs-Reformatted/resolve/main/blocks.9.hook_resid_pre/sae_weights.safetensors
151 MB
- Xet hash:
- 78e90639dbe7f76c5d049574725a9bc033b3bc9887b90cc56458b2cdbd670889
- Size of remote file:
- 151 MB
- SHA256:
- beee562c3fee612b76c0e79994d45b82780c2055cfa1305d2644b598d49efec7
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.