Update README.md
Browse files
README.md
CHANGED
|
@@ -14,7 +14,7 @@ tags:
|
|
| 14 |
|
| 15 |
# Nova-Nox-Neural-Network
|
| 16 |
|
| 17 |
-
All images used
|
| 18 |
|
| 19 |
Flash Technical Report (Japanese)
|
| 20 |
|
|
@@ -28,8 +28,6 @@ The architecture employs ASGG: Adaptive Swish-GELU Gating as the activation func
|
|
| 28 |
|
| 29 |
Furthermore, it utilizes DyT for normalization, which improves computational efficiency.
|
| 30 |
|
| 31 |
-
This repository provides a simplified implementation of N4, and it can be readily integrated with recent Attention-based architectures such as SWA, GQA, and MLA.
|
| 32 |
-
|
| 33 |
***
|
| 34 |
|
| 35 |
### Key Features
|
|
|
|
| 14 |
|
| 15 |
# Nova-Nox-Neural-Network
|
| 16 |
|
| 17 |
+
All images used are created by Rikka Botan.
|
| 18 |
|
| 19 |
Flash Technical Report (Japanese)
|
| 20 |
|
|
|
|
| 28 |
|
| 29 |
Furthermore, it utilizes DyT for normalization, which improves computational efficiency.
|
| 30 |
|
|
|
|
|
|
|
| 31 |
***
|
| 32 |
|
| 33 |
### Key Features
|