Feature Extraction
Transformers
English
biology
esm
protein
sparse-autoencoder
interpretability
protein-embeddings
protein-language-model
unsupervised-learning
Instructions to use biohub/ESMC-SAE-Overview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use biohub/ESMC-SAE-Overview with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="biohub/ESMC-SAE-Overview")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("biohub/ESMC-SAE-Overview", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -49,6 +49,12 @@ While all SAE models can be accessed through Hugging Face, only the following fi
|
|
| 49 |
- `ESMC-600M-sae-layer27-k64-codebook65536`
|
| 50 |
- `ESMC-300M-sae-layer23-k64-codebook65536`
|
| 51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
You can access an SAE model through Hugging Face using the code below:
|
| 53 |
|
| 54 |
```py
|
|
|
|
| 49 |
- `ESMC-600M-sae-layer27-k64-codebook65536`
|
| 50 |
- `ESMC-300M-sae-layer23-k64-codebook65536`
|
| 51 |
|
| 52 |
+
Install `esm` from GitHub (a PyPI release is coming soon):
|
| 53 |
+
|
| 54 |
+
```
|
| 55 |
+
pip install "esm @ git+https://github.com/Biohub/esm.git@c94ed8d763bbd7088b296949e5b401e8ea12073a"
|
| 56 |
+
```
|
| 57 |
+
|
| 58 |
You can access an SAE model through Hugging Face using the code below:
|
| 59 |
|
| 60 |
```py
|