issues
search
0-5788719150923125
/
praxis
as above, so below
https://src.eco
MIT License
2
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Terminal dashboard has alignment issues
#24
Vectorrent
opened
1 day ago
0
The model is able to produce invalid output characters
#23
Vectorrent
closed
2 days ago
1
Average step speed can vary wildly
#22
Vectorrent
opened
6 days ago
0
Restructure mod shuffling
#21
Vectorrent
closed
6 days ago
0
The Mixture of Depths router leads to model degeneration
#20
Vectorrent
closed
2 days ago
3
Convert to chatml
#19
Vectorrent
closed
2 weeks ago
0
Controller can stop predicting any experts
#18
Vectorrent
opened
3 weeks ago
0
We will need unit testing at some point
#17
Vectorrent
opened
3 weeks ago
0
Current design has "quirks" that make validation scores somewhat invalid
#16
Vectorrent
opened
4 weeks ago
1
Implement expert controller
#15
Vectorrent
closed
1 month ago
0
API inference should block training
#14
Vectorrent
closed
1 month ago
1
Add some basic Hivemind features
#13
Vectorrent
closed
1 month ago
0
Integrate an evaluation harness
#12
Vectorrent
opened
1 month ago
2
Parameter-Efficient Expert Retrieval (PEER)
#11
Vectorrent
closed
1 month ago
0
We need to split the data better
#10
Vectorrent
closed
4 weeks ago
1
Experiments are not fully-reproducable
#9
Vectorrent
closed
4 weeks ago
1
Parcel react flow
#8
Vectorrent
closed
1 month ago
0
Implement Differential Attention
#7
Vectorrent
closed
1 month ago
0
We're going to need a UI of some kind
#6
Vectorrent
opened
1 month ago
12
Implement basic LayerShuffle
#5
Vectorrent
closed
1 month ago
0
Add TOR support
#4
Vectorrent
opened
2 months ago
0
RemoteMixtureOfExperts and RemoteSwitchMixtureOfExperts cannot be used on GPU
#3
Vectorrent
closed
1 month ago
2
Experts fail to initialize > 50% of the time
#2
Vectorrent
closed
1 month ago
2
Hivemind needs WebRTC support
#1
Vectorrent
opened
2 months ago
0