RepoJournal
Alignment & Post-training Alignment & Post-training
47 wires and counting

$ follow Alignment & Post-training

Keep up with Alignment & Post-training in about 3 minutes: what actually shipped — the commits, pull requests, releases, and security advisories that matter.

or

fair warning: these emails are deeply technical. diffs, version numbers, CVEs, benchmark deltas. if that's not your idea of a good read, this isn't your newsletter.

Folds into your digest — weekly by default, monthly if you prefer. Unsubscribe in one click.

$ status

wire 2026-11-21
stories 10

© 2026 RepoJournal Home Showcase How it works Privacy

$ the-wire · showcase

MILES Hero Small defaults move to 224 requests, plus an a4b SFT recipe and a fix for a tool-data tokenizer crash

By RepoJournal · Filed · About Alignment & Post-training · Composed from the cited sources · methodology

Three things an operator can act on today: the OLMo 3.5 Small MILES examples now ship the best measured serving setting (224 running requests, 3600 s request timeout), a standing crash on tool-call SFT data gets a fix, and a new a4b recipe lands next to c12r.

MILES: Hero Small 4-node example to 224 running requests + 3600 s request timeout (-13% vs 160, -31% vs 96) allenai/open-instruct

by farhatkevin

Both MILES example configs for Hero Small now carry the best measured serving setting: admission R=224 via sglang_max_running_requests with matching CUDA graph decode batch size, 256 mamba/KDA slots, server concurrency 448 and sglang_mem_fraction_static 0.92, plus [core] refresh_request_timeout = 3600.0. The PR reports a 13% gain over 160 requests and 31% over 96, so if you copied the old four-...

Decode JSON-string tool-call arguments before rendering SFT rows allenai/open-instruct

by abhishekraok

SFT datasets store each tool call's arguments as a JSON string, and the Olmo 3.5 chat template iterates them with | items, so main's tokenizer raises TypeError: Can only get item pairs from a mapping on tool data. The PR reports 295 of the first 300 rows failing on A4B's Nemotron tool source, and all 300 passing once the strings are decoded first; the other four A4B sources tokenize the same ei...

SFT: add A4B (OLMo 3.5 Small) recipe pinned to the A4B image allenai/open-instruct

by farhatkevin

scripts/train/sft.py gains an a4b recipe, runnable as uv run python scripts/train/sft.py --recipe a4b, modelled on c12r: same C12 data and cache, 10% long-context replay, half epoch at 18,410 steps on 2x8 B300. What differs from c12r is the base (Olmo 3.5 Small hero at its 4T decay endpoint) and peak LR 7.3e-5.

Add a CUDA 13 MoE image with transformer-engine for MoE v2 SFT allenai/open-instruct

by abhishekraok

MoE v2 SFT needs transformer-engine for its token-permutation kernels and no main image shipped it, which is why A4B runs have been built from the a4b-bigbatch branch. build_image_and_launch.sh --cuda-version 13 --moe now builds a -cuda13-moe image with transformer-engine[pytorch]==2.16.1 on a CUDA 13.3.1 + cuDNN base. Still a draft.

Quick answers

What shipped in Alignment & Post-training on November 21, 2026?
Three things an operator can act on today: the OLMo 3.5 Small MILES examples now ship the best measured serving setting (224 running requests, 3600 s request timeout), a standing crash on tool-call SFT data gets a fix, and a new a4b recipe lands next to c12r. In total, 5 commits and 5 pull requests landed.
Who contributed to Alignment & Post-training on November 21, 2026?
2 developers shipped this update, including abhishekraok and farhatkevin.
What were the notable Alignment & Post-training updates?
MILES: Hero Small 4-node example to 224 running requests + 3600 s request timeout (-13% vs 160, -31% vs 96), Decode JSON-string tool-call arguments before rendering SFT rows, and SFT: add A4B (OLMo 3.5 Small) recipe pinned to the A4B image.