Codeseys
/

composer-replication-framework

Reinforcement Learning

Model card Files Files and versions

composer-replication-framework / composer_replication /diloco /serverless

67.7 kB

Ctrl+K

Ctrl+K

2 contributors

History: 4 commits

Codeseys's picture

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

54efac8 12 days ago

tests
Wave 18: 14 backlog items closed + 3-reviewer cross-family review 12 days ago
__init__.py

2.84 kB
Wave 17: close all 5 audit FLAGs + SDPO context alignment + serverless re-exports 12 days ago
allreduce.py

13.4 kB
Wave 14: close every Wave 13 review finding + 4 documentation files; Wave 14b: real PRIME-RL parity + multi-process DiLoCo convergence 12 days ago
executor.py

11.7 kB
Wave 13: serverless DiLoCo + replaysim normalization + 3 distillation losses + PRIME-RL + Monarch 12 days ago
hf_jobs.py

3.2 kB
Wave 13: serverless DiLoCo + replaysim normalization + 3 distillation losses + PRIME-RL + Monarch 12 days ago
modal.py

3.61 kB
Wave 13: serverless DiLoCo + replaysim normalization + 3 distillation losses + PRIME-RL + Monarch 12 days ago
replica_entrypoint.py

3.48 kB
Wave 13: serverless DiLoCo + replaysim normalization + 3 distillation losses + PRIME-RL + Monarch 12 days ago