Aakash Kattelu 199dae8821 fix(deriver): truncate oversize observations so one cannot drop the batch ... simple_batch_embed raised ValueError when any input exceeded the per-input token cap, which failed the entire deriver save when a single observation was over-length. Add on_oversize="truncate": oversize inputs are embedded from a token-capped prefix (re-encoded until it fits, with a warning), preserving one vector per input. Default stays "raise" so existing callers are unchanged. RepresentationManager opts into truncate. Also add a live embedding test that fails on main (raise / missing kwarg) and passes once a mixed short+oversize batch survives. Refs #569		2026-08-12 11:49:10 -07:00
..
test_document.py	Session-purity invariant + card_refresh dream type (DEV-2000) (#883)	2026-07-23 14:22:07 -04:00
test_peer_card.py	sdks: add set peer card function (#371)	2026-02-09 15:35:57 -05:00
test_representation_manager.py	fix(deriver): truncate oversize observations so one cannot drop the batch	2026-08-12 11:49:10 -07:00
test_session.py	Ben/search rrf (#179)	2025-08-06 17:43:03 -04:00
test_workspace.py	feat: implement async workspace deletion with active session checks (#378)	2026-02-12 18:48:00 -05:00