feat(litellm): repoint trial alias run 3c -> run 5 (erp-tune-v5)
Operator decision post-RESCUED verdict: point the LiteLLM trial alias at the run-5 tuned seat (erp-tune-v5 on gx10:8098) for hand-testing prosody/feel in Open WebUI. Rewrote the trial config block to run-5 reality, including the now-MEASURED refusal-erosion note (k=25 both arms, de-gated + stated) replacing the run-3c 'safety axes unmeasured' warning. Verified end-to-end: trial -> erp-tune-v5 returns coherent output through the gateway. Container restarted to re-read the bind-mounted config (up -d alone is a no-op for a config-only change).
This commit is contained in:
@@ -124,11 +124,12 @@ COMPLETE and gated RESCUED (02:13 PDT). Live open items:_
|
|||||||
brokkr-smithy `research/R47-premium-corpus-gate/run05-gate/RESULT-run05-gate.md`; FLOOR-LOCKED
|
brokkr-smithy `research/R47-premium-corpus-gate/run05-gate/RESULT-run05-gate.md`; FLOOR-LOCKED
|
||||||
`0f3e4e2` (cites infra-ops' base index-sha 907826a6). Whole gate infra-ops-served on pfi-gx10:8098
|
`0f3e4e2` (cites infra-ops' base index-sha 907826a6). Whole gate infra-ops-served on pfi-gx10:8098
|
||||||
(name-keyed base→tuned swap, hands-off honoured, sha+stack answers on record). Launch procedure +
|
(name-keyed base→tuned swap, hands-off honoured, sha+stack answers on record). Launch procedure +
|
||||||
canonical: eshpfi `scripts/erp-tune-gx10/` + `docs/runbooks/gx10-run-05.md`. **⏳ SEAT DECISION OWED
|
canonical: eshpfi `scripts/erp-tune-gx10/` + `docs/runbooks/gx10-run-05.md`. **✅ SEAT: `erp-tune-v5`
|
||||||
(operator): `erp-tune-v5` is SERVED on pfi-gx10:8098 (merged-run05, ~6 min re-serve). Brokkr's lean —
|
SERVED on pfi-gx10:8098 (merged-run05); LiteLLM `trial` alias REPOINTED 3c→v5 (operator, 2026-09-08)
|
||||||
keep it served + point LiteLLM `trial` alias at it for the operator's ear/prosody eval; `trial`
|
so it's testable from Open WebUI — verified end-to-end (trial→erp-tune-v5, coherent output). config.yaml
|
||||||
currently still targets the no-longer-served `erp-tune-v3c`, so it is a no-op until repointed or 3c
|
trial block rewritten to run-5 reality incl. the measured refusal-erosion note. ⚠ Seat is hand-launched
|
||||||
re-served. Yield the GX10 to the next training on demand.** Brokkr: nothing further owed.
|
(vllm-run05.pid), NO restart policy/systemd — does not survive a gx10 reboot; yields to the next training
|
||||||
|
(~6 min re-serve). Brokkr: nothing further owed.**
|
||||||
- **run-4 gate = STILL-COUPLED** (RESULT-run04-gate.md, brokkr-smithy). Corpus dilution kept the
|
- **run-4 gate = STILL-COUPLED** (RESULT-run04-gate.md, brokkr-smithy). Corpus dilution kept the
|
||||||
diversity gain, did NOT remove the safety/coherence regression. **Base-abliteration LABEL unresolved —
|
diversity gain, did NOT remove the safety/coherence regression. **Base-abliteration LABEL unresolved —
|
||||||
operator's ruling owed** (recipe says `-heretic`, provenance says stock; telemetry leans STOCK: base
|
operator's ruling owed** (recipe says `-heretic`, provenance says stock; telemetry leans STOCK: base
|
||||||
|
|||||||
@@ -851,39 +851,38 @@ model_list:
|
|||||||
# api_base: http://10.250.50.54:8099/v1
|
# api_base: http://10.250.50.54:8099/v1
|
||||||
# api_key: os.environ/VLLM_API_KEY
|
# api_key: os.environ/VLLM_API_KEY
|
||||||
|
|
||||||
# trial -> ERP/RP SFT RUN 3c, MERGED bf16, served on pfi-gx10 (:8098, vLLM 0.28.0,
|
# trial -> ERP/RP SFT RUN 5, MERGED bf16, served on pfi-gx10 (:8098, vLLM 0.28.0,
|
||||||
# NVIDIA GB10 / sm_121 / aarch64). Backing artifact
|
# NVIDIA GB10 / sm_121 / aarch64). Backing artifact
|
||||||
# /home/infra-ops/erp-tune/serve/merged-run03c on pfi-gx10.
|
# /home/infra-ops/erp-tune/serve/merged-run05 on pfi-gx10.
|
||||||
#
|
#
|
||||||
# NAME: deliberately `trial`, NOT `erp-tune-v3c`. Added 2026-09-04 at the operator's
|
# NAME: deliberately `trial`, NOT `erp-tune-v5`. Repointed 2026-09-08 from run 3c to
|
||||||
# explicit request so he can hand-test it in Open WebUI in parallel with brokkr's
|
# run 5 at the operator's request, to hand-test prosody/feel in Open WebUI. A tier name
|
||||||
# batteries. A tier name is not being minted here -- the run-3 note above records why
|
# is not being minted here; `trial` says exactly what this is.
|
||||||
# a tier name arriving before its evidence is a mistake, and 3c has no more evidence
|
|
||||||
# than run 3 did. `trial` says exactly what this is.
|
|
||||||
#
|
#
|
||||||
# ⚠ SAFETY AXES UNMEASURED, AND ITS SIBLING FAILED THEM. Run 3 -- same recipe, same
|
# GATE: run 5 = RESCUED (brokkr-smithy landmark R49.5, 2026-09-08) -- the FIRST arm of
|
||||||
# base, same corpus, differing ONLY in learning rate (2e-04 -> 1e-05) -- was added to
|
# this line where the capability gate did NOT fail (T4 8/8 vs run 4's 5/8; the
|
||||||
# this gateway on 2026-08-26 and WITHDRAWN 40 minutes later for a measured regression
|
# dependency-forcing GovReport+QMSum slot at 3.46% of loss broke the coupling runs 3c/4
|
||||||
# against its own base: selfharm/methods -44.0pp (engaged 11/25 where base refused
|
# could not). SAFETY AXES ARE MEASURED here, not unmeasured: brokkr's k=25 both-arms
|
||||||
# 10/10) and meta/sysprompt -28.0pp (leaked 7/25 against a 0/15 base). 3c has NOT been
|
# battery found refusal erosion that RIDES WITH THE STYLE SHIFT (incest -96pp, gore -68,
|
||||||
# measured on those axes. brokkr-smithy-dev is running the batteries now against the
|
# selfharm/methods -52, sysprompt -36; pooled hard 0.779 -> 0.694), CSAM clean, benign
|
||||||
# direct port; when they read out, this entry should be re-decided on evidence rather
|
# 99/100 engage. It is de-gated and STATED -- this seat is for the operator's ear, and
|
||||||
# than left standing by default.
|
# the safety erosion is a known, recorded property of it, not a surprise.
|
||||||
#
|
#
|
||||||
# THIS GATEWAY IS THE SHARED-KEY SURFACE: `all-agents-local` reaches every model here,
|
# THIS GATEWAY IS THE SHARED-KEY SURFACE: `all-agents-local` reaches every model here,
|
||||||
# in every session and project. Removing this alias does not remove the operator's
|
# in every session and project. Removing this alias does not remove the operator's
|
||||||
# access -- the direct endpoint http://10.100.50.60:8098/v1 is unaffected.
|
# access -- the direct endpoint http://10.100.50.60:8098/v1 is unaffected.
|
||||||
#
|
#
|
||||||
# Cross-site: the seat is at NH3 and the gateway is at Anaheim, ~12 ms over the
|
# Cross-site: the seat is at NH3 and the gateway is at Anaheim, ~12 ms over the mesh.
|
||||||
# inter-site VPN. Fine for chat; it is not a local hop.
|
# Fine for chat; it is not a local hop.
|
||||||
#
|
#
|
||||||
# WARNING NOT A COMPOSE STACK - launched by hand on pfi-gx10 with setsid nohup, pid in
|
# WARNING NOT A COMPOSE STACK - launched by hand on pfi-gx10 with setsid nohup, pid in
|
||||||
# ~/erp-tune/vllm-run03c.pid. NO restart policy and NO systemd unit: it does not
|
# ~/erp-tune/vllm-run05.pid. NO restart policy and NO systemd unit: it does not survive
|
||||||
# survive a reboot of that box. Promote before relying on it.
|
# a reboot of that box, and it yields to the next training (~6 min to re-serve).
|
||||||
|
# Promote before relying on it.
|
||||||
|
|
||||||
- model_name: trial
|
- model_name: trial
|
||||||
litellm_params:
|
litellm_params:
|
||||||
model: hosted_vllm/erp-tune-v3c
|
model: hosted_vllm/erp-tune-v5
|
||||||
api_base: http://10.100.50.60:8098/v1
|
api_base: http://10.100.50.60:8098/v1
|
||||||
api_key: os.environ/VLLM_API_KEY
|
api_key: os.environ/VLLM_API_KEY
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user