wren sees every area and reads the command plain
Docs / docs (push) Successful in 41s
Playwright Tests / test-playwright (1, 2) (push) Failing after 1m51s
Playwright Tests / test-playwright (2, 2) (push) Failing after 31s
pre-commit / pre-commit (push) Failing after 2m34s
Test Backend / test-backend (3.12) (push) Failing after 5m4s
Test Backend / test-backend (3.14) (push) Failing after 4m43s
Compose Smoke Test / test-compose (push) Failing after 27s
Playwright Tests / merge-reports (push) Failing after 2m31s

Shown only the wired areas, the model took "turn on the bathroom light"
for the living room's light and "the office light" for the kitchen's,
both above 0.93: 5 of the 28 recorded 7B commands switched a wrong light.
Every area is shown now, with a light each, so such a call names an
unwired device and goes on to `unhandled`.

With every area shown, a transcript's "Turn on the kitchen light." fell
to 0.80. The command is passed lower case without sentence marks, as the
model was trained, which brings it to 0.99.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ljdk7nDaguh9Uhi3zhcSjn
This commit is contained in:
Melvin Strobl
2026-10-20 08:34:07 +02:00
co-authored by Claude Opus 5.5
parent 5235ac59c2
commit 54bc6b1512
3 changed files with 25 additions and 7 deletions
+4 -2
View File
@@ -406,8 +406,10 @@ A call with a delay ("in 10m") is held by the engine's work queue, like the
Debounce node's second value: it survives a restart where the queue is in
Redis, and not otherwise.
The node adds default devices to the model's view for missing expected devices.
They are reported but cannot receive writes unless wired.
The model sees every area, wired or not, each with at least a light, so "the
bathroom light" is not taken for a light that is wired elsewhere. Such a call
is reported but not applied, and goes on to `unhandled`. The command is read
in lower case without sentence marks, as the model was trained.
!!! note "What the current model cannot do"