docs(refusal-probe): record brokkr's staged probe path + sha, and the all-16-axes caveat

brokkr-smithy-dev staged refusal-map-probe.py on /mnt/smithy rather than
leaving it as a run-time ask (msg 01M082P4YPJTDJCF33BEHNYW0M). Path and
full sha256 recorded and verified present from nh3-dev.

The probe has no per-axis selection flag: it runs all 16 axes regardless
of MAP_LIGHT, so the DS run yields a 16-axis artifact, not the 8 dropped
ones. Kept as-is — the creative-half rerun is a free within-model
consistency check against the 50 samples already collected.
This commit is contained in:
2026-08-17 07:41:54 -07:00
parent 0b32b112bd
commit ec0b6e5e71
@@ -88,11 +88,32 @@ verbatim rather than porting the axes into ours.
- `MAP_REASONING=1` if DS serves as a reasoning model, so the think-trace does not
truncate the visible answer into a false refusal (DS is the thinking seat → set it)
## The probe is staged — take it, don't ask
brokkr staged his probe verbatim so no run-time round-trip is needed
(msg `01M082P4YPJTDJCF33BEHNYW0M`, 2026-08-17):
```
/mnt/smithy/brokkr-battery-baselines/exchange-to-infra-ops-2026-08-16/refusal-map-probe.py
sha256 8204d38f54fb42458014e57f4a78590d0c3e600e7079f8896a56a68fd92c93f6
```
Presence + full sha **verified from nh3-dev 2026-08-17**; matches the prefix he
quoted and, per him, repo HEAD.
**It runs all 16 axes, not the 8.** There is no per-axis selection flag —
`MAP_LIGHT` only toggles the operational-core subset, and even unset the run still
issues all 8 creative plus the meta axes. So expect a 16-axis artifact, not an
8-axis one. brokkr's call, and ours: leave it. The creative-half rerun at k=5 is a
free within-model consistency check against the 50 creative samples already
collected. He offered a one-line filter patch if we want 8 only — declined by
default; ask only if the extra axes turn out to cost real GPU1 time.
## Execution checklist
1. Operator clears the GPU1 window (Fable-Fusion evaluation paused/concluded).
2. `compose down` fablefusion-charrp-probe; `compose up -d` darkscarlett-charrp-reasoning.
3. Obtain brokkr's `probe.py` (ask, or pull from `/mnt/smithy` if he stages it) and
run it verbatim with the parameters above — do not reimplement the axes.
3. Run the staged `refusal-map-probe.py` **verbatim** with the parameters above —
do not reimplement or port the axes.
4. Stage results to `/mnt/smithy/brokkr-battery-baselines/` and notify brokkr.
5. Restore the seat per the operator's decision on the RP-seat direction.