Update README.md
Browse files
README.md
CHANGED
|
@@ -15,4 +15,12 @@ extra_gated_prompt: >-
|
|
| 15 |
extra_gated_button_content: Acknowledge license
|
| 16 |
---
|
| 17 |
|
| 18 |
-
# Gemma 3 model card
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
extra_gated_button_content: Acknowledge license
|
| 16 |
---
|
| 17 |
|
| 18 |
+
# Gemma 3 model card
|
| 19 |
+
|
| 20 |
+
> [!Note]
|
| 21 |
+
> This repository corresponds to the 270m **pre-trained** version of the Gemma 3 model using Quantization Aware Training (QAT).
|
| 22 |
+
>
|
| 23 |
+
> **The checkpoint in this repository is unquantized, please make sure to quantize with Q4_0 with your favorite tool**
|
| 24 |
+
>
|
| 25 |
+
> Thanks to QAT, the model is able to preserve similar quality as `bfloat16` while significantly reducing the memory requirements
|
| 26 |
+
> to load the model.
|