memory: R22 operator steer — research gated on pragmatic/deployable outcome, not art

This commit is contained in:
vh
2026-06-20 21:52:17 -07:00
parent 9d65339fb2
commit b33049ce3e
+7 -1
View File
@@ -306,7 +306,13 @@ _As of 2026-06-20:_
irv-ml1 A6000 ~18GB slice, NO Blackwell displacement; they'll send sizing + grant
claude-bot read on vh/brokkr-smithy at P00-PASS. **FOLLOW-UP: clean phantom
`qwen3.6-35b-a3b` off the gateway** (shows in /v1/models but 400s "invalid model" —
backend displaced 2026-06-19; the no-false-alias footgun).
backend displaced 2026-06-19; the no-false-alias footgun). **OPERATOR STEER
(2026-06-20): R22 is research for a PRAGMATIC/deployable outcome, NOT advancing-the-art
("another day's work").** Before the dwarves invest in Phase B, establish a deploy path
(distilled steering-vector / fine-tune applied to a model we already serve, e.g.
Worldtree's) — don't research an avenue we won't implement. Phase A/P00 (the cheap
prompt-level gate) runs regardless (it just tells us if the affect signal exists).
Operator is taking the dwarf discussion directly.
- `[2026-06-20]` **claude-bot issue-scope token minted for worldtree-dev self-serve**
(closes their last `tea`-as-vh fallback — issues; they already self-serve Actions +