Update README.md
Browse files
README.md
CHANGED
|
@@ -2,6 +2,10 @@
|
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
|
| 5 |
-
**Sparse Autoencoders for *Evo 2*** — BatchTopK sparse autoencoders for Evo 2
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
[More on Evo 2](https://arcinstitute.org/tools/evo)
|
|
|
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
|
| 5 |
+
**Sparse Autoencoders for *Evo 2*** — BatchTopK sparse autoencoders for Arc Institute's Evo 2 genomic foundation model.
|
| 6 |
+
|
| 7 |
+
Evo 2 is a genomic foundation model capable of generalist prediction and design tasks across DNA, RNA, and proteins. It uses a frontier deep learning architecture to enable modeling of biological sequences at single-nucleotide resolution with near-linear scaling of compute and memory relative to context length. Evo 2 is trained with 40 billion parameters and 1 megabase context length on over 9 trillion nucleotides of diverse eukaryotic and prokaryotic genomes.
|
| 8 |
+
|
| 9 |
+
This repository contains the layer 26 SAE mixed prokaryote/eukaryote SAE used in the Evo 2 paper.
|
| 10 |
|
| 11 |
[More on Evo 2](https://arcinstitute.org/tools/evo)
|