Hugging Face's logo

Spaces:
Debito
/
mamba-encoder-swarm_app
Sleeping

App Files Community
Debito commited on
Commit
3f6e6c9
·
verified ·
1 Parent(s): 3fe310c

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -0
README.md CHANGED
@@ -1,3 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
1
  # What is M E S ?
2
  M E S (short for MAMBA ENCODER SWARM) is a novel architecture that comprises of MAMBA's structured state space, configured to implement a multiple encoder swarm that are dynamically, sparsely routed to spread the heavy QxKxV matrix multiplication computional intensity across multiple MAMBA encoders (between 5 to 1000) and with the output sparsely aggregated with a MAMBA decoder, thereby bypassing the high cost of inference without sacrificing on the response generation quality.
3
 
 
1
+ title: Mamba Encoder Swarm
2
+ emoji: 🐍
3
+ colorFrom: orange
4
+ colorTo: yellow
5
+ sdk: gradio
6
+ sdk_version: "4.0.0"
7
+ app_file: app.py
8
+ pinned: false
9
+ license: mit
10
+
11
+
12
  # What is M E S ?
13
  M E S (short for MAMBA ENCODER SWARM) is a novel architecture that comprises of MAMBA's structured state space, configured to implement a multiple encoder swarm that are dynamically, sparsely routed to spread the heavy QxKxV matrix multiplication computional intensity across multiple MAMBA encoders (between 5 to 1000) and with the output sparsely aggregated with a MAMBA decoder, thereby bypassing the high cost of inference without sacrificing on the response generation quality.
14