2.51 MB

Ctrl+K

4 contributors

History: 29 commits

Codeseys

feat(trainer): ADR-008 Dr.GRPO config + SDPO strict-alignment guard

bde5c5e about 1 month ago

composer_replication
feat(trainer): ADR-008 Dr.GRPO config + SDPO strict-alignment guard about 1 month ago
docs
docs(adr): add ADR-008/009/010 (Dr.GRPO+SDPO, layered hints, FeatureDeletionEnv) about 1 month ago
examples
feat(trainer): ADR-008 Dr.GRPO config + SDPO strict-alignment guard about 1 month ago
framework
Wave 4: data collator + loss composition smoke (38/38 tests pass) about 2 months ago
publications
Wave 5: full publication-materials drafts (pre-experimental release set) about 2 months ago
research
research: Composer 2.5 data-gen + targeted-textual-feedback deep-research wave about 1 month ago
spikes
Wave 18: 14 backlog items closed + 3-reviewer cross-family review about 2 months ago
.gitignore

861 Bytes
Wave 16: install ergonomics + gradient evidence + SDPO end-to-end example about 2 months ago
BACKLOG.md

4.84 kB
Wave 7: Phase 2-4 of deep work loop — backlog, parallel research, three ADRs about 2 months ago
CITATION.bib

2.61 kB
Wave 5: full publication-materials drafts (pre-experimental release set) about 2 months ago
CITATION.cff

3.54 kB
Wave 5: full publication-materials drafts (pre-experimental release set) about 2 months ago
LICENSE

1.07 kB
Initial commit: Composer 2.5 Replication Framework — research synthesis about 2 months ago
README.md

21.8 kB
Wave 18: 14 backlog items closed + 3-reviewer cross-family review about 2 months ago
pyproject.toml

4.57 kB
Wave 16: install ergonomics + gradient evidence + SDPO end-to-end example about 2 months ago
uv.lock

576 kB
Wave 21: close both Wave 20 debt items — chat-template alignment + structural is_error about 1 month ago