-
Notifications
You must be signed in to change notification settings - Fork 98
Pull requests: ndif-team/nnsight
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
refactor(interleaver): run interventions as greenlets, not OS threads
#680
opened Jun 25, 2026 by
JadenFiotto-Kaufman
Member
Loading…
Gather tensor-parallel sharded parameters on read in a trace
#677
opened Jun 20, 2026 by
khaiwang
Contributor
Loading…
feat(intervention): isolated GPU-worker execution backend for model.trace()
#676
opened Jun 20, 2026 by
khaiwang
Contributor
Loading…
Fix MissedProviderError in .backward() with multiple invokers (#664)
#671
opened Jun 7, 2026 by
azrabano23
Loading…
Pipeline-parallel intervention support for the vLLM backend
#670
opened Jun 5, 2026 by
khaiwang
Contributor
Loading…
Add nnsight login command and notebook helper (#506)
#668
opened May 30, 2026 by
EphraiemSarabamoun
Loading…
Add Function Vectors reproduction example notebook
#667
opened May 20, 2026 by
Devansh-Ruia
Loading…
Fix: Add GPU synchronization for cached tensor transfers
#666
opened May 18, 2026 by
Pritiks23
Loading…
feat: run:ai gpu_direct model loading with quantization-aware device maps
#635
opened Mar 20, 2026 by
khaiwang
Contributor
Loading…
6 tasks done
Add comprehensive tests for module renaming/aliasing functionality
#599
opened Jan 28, 2026 by
Butanium
Member
Loading…
Cache ignores not found alias without failing
#552
opened Nov 13, 2025 by
AdamBelfki3
Member
Loading…
Issue #434: Fix autoamp init wrapper compatibility
#439
opened Jun 16, 2025 by
karthik-ravi-1537
Loading…
ProTip!
Follow long discussions with comments:>50.