vh/esh-pfi-infrastructure
1
0
Fork 0
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity

fix(cyberprev-seat): serve coherent v2 quant from name-repaired source; wire gateway

Browse Source 
The first cyberprev quant served pure gibberish despite passing every structural
check. Root cause: the hotdogs/Qwen3.8-27B-abliterated-cyber-preview checkpoint
ships broken -- 850 of 1199 tensors (the whole transformer body) carry a TRIPLED
`language_model.` prefix, confirmed present in the upstream HF index (an unsloth
export bug, not our download). transformers silently loaded the unmapped tensors
as random-init, the pipeline quantized noise, and structural verification passed
because it inspects the output config, never that input names mapped. A plain
greedy coherence smoke caught it; nothing structural could.

Fix: rewrite tensor names against a known-good reference of the same architecture
(qwen38-27b-uncensored-bf16), verified the repaired name-set == reference name-set
exactly (1199 tensors, 0 collisions), and re-quantized from the corrected source
(cyberprev-bf16-fixed). v2 is coherence-verified (spec off AND live with dflash):
Paris/Berlin/Rome, correct arithmetic precedence, primes, clean cyber-domain chat
with a proper thinking split. dflash k=7 measured at 2.77 mean acceptance length.

Seat serves cyberprev-27b / cyberprev-27b-thinking on :8025/GPU0 with dflash k=7,
displacing the retired sentinel-r3. Gateway: removed the retired sentinel-r3 /
sentinel-r3-reasoning aliases and added cyberprev-27b / cyberprev-reasoning -- NOT
repointed, since serving cyberprev weights under sentinel's name is silent
substitution. Verified end-to-end through the gateway; sentinel-r3 now 400s.

Playbook gains a superseded-claims row: structural verification does not imply a
good quant; a coherence smoke is the load-bearing gate, and third-party sources
must have their tensor names diffed against a reference before quantizing.

Broken v1 quant dir removed.
This commit is contained in:
vh
2026-09-14 04:28:59 -07:00
parent 1418edb03e
commit b94d173887
4 changed files with 47 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
stacks/cyberprev-seat/.env.example
+7 -3
View File
@@ -14,9 +14,13 @@ CYBER_PORT=8025
CYBER_CONTAINER_NAME=vllm-cyberprev
# ── Model ───────────────────────────────────────────────────────────────────
# In-house mixed quant of /tank/aimodels/cyberprev-bf16 (51.0 GiB bf16 source),
# built with services/gen-seat-mixed-quant/. compressed-tensors, NOT modelopt_fp4.
CYBER_MODEL=/tank/aimodels/cyberprev-nvfp4-mixed
# In-house mixed quant, compressed-tensors (NOT modelopt_fp4).
# ⚠ Quantized from /tank/aimodels/cyberprev-bf16-FIXED, not the raw download. The raw
# hotdogs export ships 850/1199 body tensors with a TRIPLED `language_model.` prefix
# (upstream bug) -> quant of the raw source is gibberish that passes every structural
# check. cyberprev-bf16-fixed is the name-repaired source (verified tensor-for-tensor
# against qwen38-27b-uncensored-bf16). See docs/pfi/model-quantization-playbook.md.
CYBER_MODEL=/tank/aimodels/cyberprev-nvfp4-mixed-v2
CYBER_QUANT=compressed-tensors
# ⚠ Its OWN name. Do not reuse `sentinel-r3` — that seat is retired and its gateway
# aliases are deliberately left to 404 rather than repointed at different weights.