Predicting the Order of Upcoming Tokens Improves Language Modeling
Paper
•
2508.19228
•
Published
•
22
•
2
Zayd Muhammad Kawakibi Zuhri PRO
zaydzuhri
AI & ML interests
I really like watching loss go down
Recent Activity
updated
a model
8 days ago
zaydzuhri/dsmtp-7B-4096-model
published
a model
8 days ago
zaydzuhri/dsmtp-7B-4096-model
Organizations
None yet
commented
a
paper
2 months ago
commented
3
papers
6 months ago
Softpick: No Attention Sink, No Massive Activations with Rectified Softmax
Paper
•
2504.20966
•
Published
•
32
•
5
Softpick: No Attention Sink, No Massive Activations with Rectified Softmax
Paper
•
2504.20966
•
Published
•
32
•
5
Softpick: No Attention Sink, No Massive Activations with Rectified Softmax
Paper
•
2504.20966
•
Published
•
32
•
5
commented
a
paper
over 1 year ago