Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -3,7 +3,7 @@ license: mit
|
|
| 3 |
library_name: transformers
|
| 4 |
base_model: google/t5gemma-2-270m-270m
|
| 5 |
colab:
|
| 6 |
-
- link: https://colab.research.google.com/github/
|
| 7 |
title: T5Gemma 2 SAE - Quick Start
|
| 8 |
model_index:
|
| 9 |
- name: google/t5gemma-2-270m-270m
|
|
@@ -39,7 +39,7 @@ tags:
|
|
| 39 |
|
| 40 |
**Sparse Autoencoders (SAEs)** trained on all 36 layers of `google/t5gemma-2-270m-270m` for mechanistic interpretability and activation steering.
|
| 41 |
|
| 42 |
-
[](https://colab.research.google.com/github/
|
| 43 |
|
| 44 |
## Quick Start
|
| 45 |
|
|
|
|
| 3 |
library_name: transformers
|
| 4 |
base_model: google/t5gemma-2-270m-270m
|
| 5 |
colab:
|
| 6 |
+
- link: https://colab.research.google.com/github/haddock-development/t5gemma2-sae-all-layers/blob/main/colab.ipynb
|
| 7 |
title: T5Gemma 2 SAE - Quick Start
|
| 8 |
model_index:
|
| 9 |
- name: google/t5gemma-2-270m-270m
|
|
|
|
| 39 |
|
| 40 |
**Sparse Autoencoders (SAEs)** trained on all 36 layers of `google/t5gemma-2-270m-270m` for mechanistic interpretability and activation steering.
|
| 41 |
|
| 42 |
+
[](https://colab.research.google.com/github/haddock-development/t5gemma2-sae-all-layers/blob/main/colab.ipynb)
|
| 43 |
|
| 44 |
## Quick Start
|
| 45 |
|