Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

Pin batch-size invariance of safety and judge generation in fp32 (`batch_size` 1 / 2 / 4 / 8)

Open
#922 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
3/5
Estimated time
Half a day
Newbie friendliness
62/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
python
Domain
testing

Research direction

Start with the batched generation helpers in forgelm/safety/_generate.py and forgelm/judge.py, then find the existing batch_size tests, which only check input validation. Add a tiny-model fp32 test that runs safety and judge generation at batch sizes 1, 2, 4 and 8 and asserts identical outputs. Done means that test passes, the row at docs/roadmap/risks-and-decisions.md:91 is removed, and the CHANGELOG names F-PR29-A8-10.

Written by the indexing model from the issue text.

Description

area: eval-gates help wanted severity: low source: roadmap wave: 4

Deferred work · F-PR29-A8-10 · severity Low · cost ~1-2 h · recorded in docs/roadmap/risks-and-decisions.md:91

Summary

Both gates generate the answers they score with left-padded, batched greedy decoding (forgelm/safety/_generate.py, forgelm/judge.py), but every test stubs the batched helpers, so a change that made results depend on batch_size would ship without a failing test. The PR #29 review deferred the determinism test. #677 shows bf16 output legitimately varies with batch size, so the contract to pin is fp32.

Condition or budget

None recorded (the row still names v0.6.x).

State today (re-checked 2026-10-09)

No determinism test exists; the only batch_size tests check input validation.

Related

#676 asks for a safety-side batched-versus-single test (and accepts a weaker padding-side fix for the judge); #677 documents the bf16 variance.

Acceptance criteria

  • A tiny-model fp32 test runs safety generation and judge generation at batch_size 1, 2, 4 and 8 and asserts identical outputs.
  • The row is removed from the deferral table and the CHANGELOG names its ID (the table's removal contract).

Recorded in docs/roadmap/risks-and-decisions.md (read at f94595f) and re-checked against the code on 2026-10-09. The row leaves that table only when the fix is on main, the CHANGELOG names its ID and any promised guard exists.

Dominant language
Python
Stars
9
Forks
1
Avg merge
8h 10m
Merged PRs (30d)
5

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from HodeTech/ForgeLM

All issues in HodeTech/ForgeLM

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.