-
Notifications
You must be signed in to change notification settings - Fork 1
Pull requests: benchopt/benchmark_nanogpt
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Validation loss 3.28 with AdamW in less than 10k iterations (local bs=64)
#15
opened May 28, 2026 by
svaiter
Contributor
Loading…
ENH add rotary embeddings+QK-norm+rm some bias/improve init
#3
opened Jul 23, 2025 by
tomMoral
Member
Loading…
3 of 5 tasks
ProTip!
Exclude everything labeled
bug with -label:bug.