Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,64 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
language:
|
| 7 |
+
en
|
| 8 |
+
pipeline_tag: conversational
|
| 9 |
+
tags:
|
| 10 |
+
nce
|
| 11 |
+
neural-cognition
|
| 12 |
+
phi-model
|
| 13 |
+
license: other
|
| 14 |
+
|
| 15 |
+
Zemo
|
| 16 |
+
Next-Generation NCE
|
| 17 |
+
Zemo is a Next-Generation Neural Cognition Engine developed under the Algorithmic Computer-generated Consciousness framework by the ACC.
|
| 18 |
+
Model Information
|
| 19 |
+
Property
|
| 20 |
+
Details
|
| 21 |
+
Name
|
| 22 |
+
Zemo
|
| 23 |
+
Model Type
|
| 24 |
+
Next-Generation NCE
|
| 25 |
+
Organization
|
| 26 |
+
ACC
|
| 27 |
+
Framework
|
| 28 |
+
Neural Cognition Engine
|
| 29 |
+
Research Direction
|
| 30 |
+
Neural cognition and PhiModel concepts
|
| 31 |
+
Foundation Model Reference
|
| 32 |
+
Kimi K2.5
|
| 33 |
+
Architecture
|
| 34 |
+
Mixture of Experts (MoE)
|
| 35 |
+
|
| 36 |
+
Architecture
|
| 37 |
+
Zemo is associated with the NCE framework, which explores the integration of multiple neural network architectures, including transformers, recurrent neural networks, convolutional neural networks, feedforward networks, and spiking neural networks.
|
| 38 |
+
The framework also incorporates the PhiModel concept, which investigates the use of the golden ratio as a mathematical principle for neural organization.
|
| 39 |
+
Technical Specifications
|
| 40 |
+
|
| 41 |
+
Specification
|
| 42 |
+
Value
|
| 43 |
+
Total Parameters
|
| 44 |
+
1 Trillion
|
| 45 |
+
Activated Parameters
|
| 46 |
+
32 Billion
|
| 47 |
+
Layers
|
| 48 |
+
61
|
| 49 |
+
Experts
|
| 50 |
+
384
|
| 51 |
+
Experts per Token
|
| 52 |
+
8
|
| 53 |
+
Context Window
|
| 54 |
+
256K tokens
|
| 55 |
+
Vocabulary Size
|
| 56 |
+
160K
|
| 57 |
+
Attention Mechanism
|
| 58 |
+
Multi-head Latent Attention (MLA)
|
| 59 |
+
Activation Function
|
| 60 |
+
SwiGLU
|
| 61 |
+
Vision Encoder
|
| 62 |
+
MoonViT
|
| 63 |
+
Input Modalities
|
| 64 |
+
Text, images
|