Hugging Face's logo

Codeseys
/
composer-replication-framework

Reinforcement Learning
Transformers
English
post-training
distillation
agentic-coding
composer-2.5
cursor
kimi-k2
grpo
dapo
diloco
openenv
trl
verl
research
methodology
Model card Files
xet
 Community
composer-replication-framework / docs

Commit History

docs(wave2): archive point-in-time wave reviews + dated review bundles
f00833d

Codeseys Claude Opus 4.8 (1M context) commited on

docs(wave1): correctness pass — channel-3 provenance, gap honesty, dead links
20e3bd9

Codeseys Claude Opus 4.8 (1M context) commited on

feat(trainer): policy-optimization objective MENU (ADR-014)
aae66fa

Codeseys commited on

docs(adr-013): close acceptance-gate boxes 1-5; only user-gated spend remains
7b34ebf

Codeseys commited on

fix(phase-8): close all 5 cross-family final-verify findings + regression tests
678d10b

Codeseys commited on

feat(b4-gpu+b6): GPU train-proof on A10G + docker-gated substrate E2E test
c009712

Codeseys commited on

feat(wave-b): ADR-013 LMA integration + B4 end-to-end SDPO-fires proof + doc refresh
21647a4

Codeseys commited on

feat(wave-a): close ADR-011 (SDPO alignment indices) + ADR-012 (review findings)
d02d724

Codeseys commited on

architect: ADR-011/012/013 + research (alignment-index fix, review-findings closure, LMA channel-ladder)
b4a584a

Codeseys commited on

review+fix: cross-family adversarial ADR review (owed item) + remediation
185cce2

Codeseys commited on

docs: wave summary — Composer 2.5 data-gen + targeted-RL textual-feedback
8498e8f

Codeseys commited on

feat(datagen): ADR-010 FeatureDeletionEnv synthetic-data subsystem; accepted
9336af3

Codeseys commited on

feat(hints): ADR-009 layered HintGenerator; accepted
84740d4

Codeseys commited on

feat(trainer): ADR-008 gate-3 live GRPO+SDPO smoke PASS; ADR-008 accepted
2a34df4

Codeseys commited on

docs(adr): add ADR-008/009/010 (Dr.GRPO+SDPO, layered hints, FeatureDeletionEnv)
36ab61e

Codeseys commited on

Wave 18: 14 backlog items closed + 3-reviewer cross-family review
54efac8

Codeseys commited on

Wave 17: close all 5 audit FLAGs + SDPO context alignment + serverless re-exports
a84c060

Codeseys commited on

Wave 16: install ergonomics + gradient evidence + SDPO end-to-end example
c0a5ab7

Codeseys commited on

Wave 15: 4-angle multi-model self-critique caught 2 math BLOCKERs in primary loss kernels; fixed against upstream byte-for-byte + GSM8K example + ergonomics
e5add15

Codeseys commited on

Wave 14: close every Wave 13 review finding + 4 documentation files; Wave 14b: real PRIME-RL parity + multi-process DiLoCo convergence
d9dd3a5

Codeseys commited on

Wave 13: serverless DiLoCo + replaysim normalization + 3 distillation losses + PRIME-RL + Monarch
b266c31

Codeseys commited on

Wave 12: close V1-V8 brief — GPU smoke, SDPO firing, real-trace e2e
d88715c

Codeseys commited on

Wave 11: cross-model adversarial review + honest down-revision
f16fa23

Codeseys commited on

Wave 7: Phase 2-4 of deep work loop — backlog, parallel research, three ADRs
ac4bfb4

Codeseys commited on

Wave 6: vision validation self-audit (5/10 to 9/10 in 5 days, no GPU)
040eff8

baladithyab commited on

Wave 3: integration architecture + spike-005 trainer skeleton (16 tests pass)
fd77f74

baladithyab commited on

Integrate Cursor blog directly + audit research note + add SDPO/OPSD link
1cede23

baladithyab commited on

Initial commit: Composer 2.5 Replication Framework — research synthesis
7165832

Codeseys commited on