File size: 1,311 Bytes
016b363 f8afffd 068c017 016b363 617c6ee 2159c7f 67b2a69 016b363 bd3fe55 67b2a69 bd3fe55 67b2a69 884074a 016b363 bd3fe55 016b363 0127b34 016b363 db1f4e0 016b363 bd3fe55 016b363 2232397 0127b34 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 |
---
license: cc-by-4.0
library_name: saelens
---
⚠️ WARNING: We have small labelling issues, and some SAEs appear twice in this repo.
# 1. Gemma Scope
Gemma Scope is a comprehensive, open suite of sparse autoencoders for Gemma 2 9B and 2B. Sparse Autoencoders are a "microscope" of sorts that can help us break down a model’s internal activations into the underlying concepts, just as biologists use microscopes to study the individual cells of plants and animals.
See our [landing page](https://huggingface.co./google/gemma-scope) for details on the whole suite. This is a specific set of SAEs:
# 2. What Is `gemma-scope-2b-pt-res`?
- `gemma-scope-`: See 1.
- `2b-pt-`: These SAEs were trained on Gemma v2 2B base model.
- `res`: These SAEs were trained on the model's residual stream.
- We include experimental SAEs trained on token embeddings in the ./embedding folder.
# 3. Which SAE is in the [Neuronpedia demo](https://www.neuronpedia.org/gemma-scope)?
https://huggingface.co./google/gemma-scope-2b-pt-res/tree/main/layer_20/width_16k/average_l0_71
# 4. Point of Contact
Point of contact: Arthur Conmy
Contact by email:
```python
''.join(list('moc.elgoog@ymnoc')[::-1])
```
HuggingFace account:
https://huggingface.co./ArthurConmyGDM
# Citation
Paper: https://arxiv.org/abs/2408.05147
|