Skip to content

fix(models): pin Transformers device_map to GPU 0 when auto sharding breaks sampling #44

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Aafiya-H wants to merge 3 commits into
base: main
Choose a base branch
from
Open

fix(models): pin Transformers device_map to GPU 0 when auto sharding breaks sampling #44

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
26593fe
fix(models): pin Transformers device_map to GPU 0 when auto sharding …
Aafiya-H May 21, 2026
01caaa1
merge branch 'main' of https://github.com/Aafiya-H/effGen
Aafiya-H May 22, 2026
20c9c3a
Merge branch 'observability' into fixing_side_assert
Aafiya-H May 24, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -418,7 +418,11 @@ result = agent.run_message(msg)
### Documentation

- `docs/prompts/library.md` — framework overview, key classes, CLI reference, adding-new-domain guide.
- `docs/prompts/research.md`, `coding.md`, `data.md`, `legal.md`, `medical.md`, `creative.md`, `business.md`, `playground.md` — per-domain guides.
- `docs/prompts/research.md`, `coding.md`, `data.md`, `legal.md`, `creative.md`, `business.md`, `playground.md` — per-domain guides.

### Fixed

- **`examples/basic/calculator_agent.py`** — tear down the math agent in a `finally` block (`agent.close()` before `model.unload()`) so demo and interactive runs exit cleanly without the "Agent was garbage-collected without calling close()" warning; avoids leaving agent resources open when CUDA generation fails mid-run.

---

Expand Down
Loading