Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
This repository contains configurations to slice Gemma 3n E4B, which is enabled thanks to it being a MatFormer.
|
2 |
+
The E4B model can be sliced into small models, trading off quality and latency/compute requirements.
|
3 |
+
We recommend exploring the [MatFormer Lab](TODO: add link) to getting started with slicing Gemma 3n E4B yourself.
|
4 |
+
|
5 |
+
For each configuration, we calculate the MMLU accuracy.
|
6 |
+
Although these are not the only configurations possible, they are optimal configurations
|
7 |
+
identified by calculating the accuracy of the pre-trained model
|
8 |
+
|
9 |
+
To learn more about MatFormers, please review the and generate your own submodels
|
10 |
+
with the [MatFormer Lab](TODO: add link).
|
11 |
+
|
12 |
+
 capability.")
|
13 |
+
This chart show’s MMLU performance vs model size of Gemma 3n Mix-n-Match (pretrained) capability.
|
14 |
+
|
15 |
+
Some additional resources:
|
16 |
+
|
17 |
+
* [Gemma 3n launch blog](https://developers.googleblog.com/en/introducing-gemma-3n-developer-guide)
|
18 |
+
* [MatFormer paper](https://huggingface.co/papers/2310.07707)
|