Skip to content

Pull requests: KellerJordan/modded-nanogpt

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

[New Record] Parallel Residuals (-45 steps, -0.85s)
#230 opened Feb 12, 2026 by msisovic Loading…
Dion2 fix eliminate zeros
#229 opened Feb 12, 2026 by JohnLangford Loading…
Run on 4x H100s instead of 8x
#228 opened Feb 12, 2026 by JohnLangford Loading…
Dion2 fix compile
#227 opened Feb 12, 2026 by JohnLangford Loading…
4x h100 dion2
#226 opened Feb 12, 2026 by JohnLangford Loading…
Dion2 fix prealloc
#225 opened Feb 12, 2026 by JohnLangford Loading…
New WR: sparse bigram gradient comms (-0.6 seconds)
#221 opened Feb 6, 2026 by shenberg Loading…
New WR: Tuned Value Embeddings (-0.5s)
#218 opened Feb 3, 2026 by photomz Loading…
Tie First and Last VEs
#212 opened Jan 29, 2026 by chrisjmccormick Draft
Create Dependabot.yml for dependency management
#202 opened Jan 21, 2026 by QueenFi703 Loading…
Improved logging
#98 opened May 10, 2025 by YouJiacheng Loading…
Add efficient validation on HellaSwag
#89 opened Mar 13, 2025 by trianxy Loading…
ProTip! Mix and match filters to narrow down what you’re looking for.