5a5f5c267e436c5feacc8b22ad626526a040c204
1107
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 vh
|
5a5f5c267e |
power: RETRACT the DCGM caveat — config management and power limits ARE supported on our cards
I claimed in
|
||
|
vh
|
100670eed1 |
power: what DCGM is, and why not to plan around it on workstation-SKU cards
DCGM is NVIDIA's own Data Center GPU Manager -- first-party, Apache-2.0, packaged as datacenter-gpu-manager -- and it layers above NVML rather than beside it: nvidia-smi is a thin CLI over NVML's per-GPU primitives, and DCGM is a daemon plus dcgmi adding health, diagnostics, config enforcement, policy and group abstractions on top. Which is why its group notion still resolves to N per-GPU writes underneath. The caveat that matters, and it undercuts the experiment suggested in the previous commit: DCGM is datacenter-oriented and parts of it are gated to datacenter SKUs of the Tesla/A100/H100 class. Our cards are professional/workstation parts -- RTX PRO 6000 Blackwell Max-Q and RTX 6000 Ada -- and several DCGM capabilities are unsupported or degraded outside that line, plausibly including config enforcement, which is precisely the power path. So DCGM_CONFIG_POWER_BUDGET_GROUP may return 'unsupported on this device'. Downgraded from 'worth testing' to five minutes of curiosity after the real work, and explicitly not a planning assumption. None of which touches the plan: nvidia-smi -pl 250 is plain NVML and works on these cards. DCGM would only have bought the group-budget experiment and nicer telemetry, and is probably not installed anyway since beszel-agent-nvidia shells out to nvidia-smi. |
||
|
vh
|
94fb7b7208 |
power: answer the bank-budget question — DCGM has the concept, the dynamic part is a control loop, and 4x250 already is 1000 W
DCGM_CONFIG_POWER_BUDGET_GROUP ('the power budget for the entire group') exists
alongside DCGM_CONFIG_POWER_CAP_INDIVIDUAL, so the concept is first-class. The docs do
not state how a group budget is distributed, and the deduction is that it cannot be
anything exotic: the only enforcement primitive underneath is NVML's per-GPU
nvmlDeviceSetPowerManagementLimit and there is no bank-level register, so any group
budget resolves to N per-GPU writes. Static even division is one write each; 'each card
free until they are all loaded' requires continuous re-writing, which is a control loop
rather than a hardware feature. Worth a ten-minute test when the box returns, in case
NVIDIA already runs that loop.
Records the design constraint that matters more than the logic: power readings lag and
-pl application takes tens of milliseconds, so a reactive daemon overshoots during a load
ramp -- and the ramp is the dangerous moment, being the same all-cards-at-once shape as
this box's ten restart:unless-stopped containers starting together. So any such loop must
be safe-by-default and opportunistic upward: boot at budget/N, only ever raise after
observing idle neighbours. Inverted, it works for weeks and then fails on precisely the
event it existed to prevent.
And the reason to defer it: 4x250 W is already 1000 W, so the static cap is the
conservative floor of the dynamic scheme rather than an alternative. The daemon's entire
contribution is the one-card-busy case, worth perhaps 5% throughput, which is rare for a
serving fleet that puts one seat per card and common only for a training window.
|
||
|
vh
|
b538fde6f0 |
caps: fv-ml1 250W / ana-ml3 200W — and nvidia-smi -pl caps BOARD power, not wall power
Operator set fv-ml1 at 250 W per card (83% of TGP, ~5% throughput) and ana-ml3 at 200 W (67%, ~10-15%). Records the term that decides whether 250 W actually clears a 15 A feed, because it is easy to drop: a power limit bounds BOARD power, and the wall sees that divided by PSU efficiency. Four cards at 250 W is 1000 W of board; add 180-300 W of host components and divide by ~0.90 and the plug sees ~1310-1445 W, against a 15 A circuit's 1440 W NEC continuous derating -- an inference box serving all day being a continuous load. So 250 W lands ON the limit rather than under it, where 200 W would give ~1090-1220 W with real margin. The deciding term is the host draw, which is still an estimate, so the procedure is: set 250 W, verify at the plug under four-card load, fall back to 200 W if it reads near 1440 W. A cap is a claim; the ammeter is the verification. Two consequences recorded alongside. Caps bound sustained draw and not transients -- the enforcement window is short but not instantaneous -- and while a breaker's thermal-magnetic curve forgives brief overload, a UPS's overload protection does not. So 250 W implicitly commits the fv-ml1 chassis to the PDU rather than behind the 1500 VA unit, which it exceeds even capped. And ana-ml3's 200 W across only two cards is deliberately conservative at 400 W total, relaxable if Anaheim's measured headroom beats its trip history. |
||
|
vh
|
2da0c76d99 |
correct the hardware: fv-ml1 is 4x Blackwell Max-Q 300W, ana-ml3 is 2x Ada RTX 6000 — and four cards is a breaker problem
Operator clarification, and it separates two boxes I had been conflating. fv-ml1 is 4x Blackwell RTX PRO 6000 Max-Q at 300 W each (Max-Q being the reduced-TGP SKU; the Workstation Edition is the 600 W part), 391 GB VRAM, deployed and currently dark. ana-ml3 is 2x Ada Generation RTX 6000 at 300 W, 96 GB VRAM, not yet deployed. The 200 W cap directive is ana-ml3's. With the TGP known, the outage stops being a vague 'undersized' and acquires a mechanism: two Max-Q cards at 300 W is ~600 W of card, plus a host carrying 566 GB of RAM, drives, fans and PSU conversion loss at perhaps 200-350 W, against an Eaton 1500 VA's real ~900-1200 W. That lands at or just over the rating, which is precisely what explains a full day of service on one card and failure minutes into the second. The host term is the only one being guessed; idle-at-the-plug measures it directly. It also surfaces something that is not a UPS question at all. Four cards at 300 W plus ~300 W of host is ~1500 W against a 15 A circuit's 1440 W continuous derating, so four cards uncapped is marginal on the breaker with no UPS in the path. Capping therefore belongs at fv-ml1 as well as ana-ml3, or fv-ml1 needs a 20 A feed -- and worth noting today's incident only ever had two of the four cards working. ana-ml3's placement constraints sharpen too: sm_89 has native FP8 but no NVFP4, so the in-house NVFP4 quants stay at FV, and at 96 GB total it cannot host the Flash-Next seat at all -- that needs 74 GiB resident on a single card, and the offload moves the n-gram table rather than the experts. |
||
|
vh
|
8fcc26e2c9 |
policy(gpu-power): cards are RTX 6000 Ada at 300 W — 200 W is a mild cap, plus two sm_89 placement consequences
Corrects the SKU: RTX 6000 Ada, 300 W, not the ~600 W initially recalled. That makes 200 W a cap to 67% of TGP -- the favourable part of the concave perf/watt curve, roughly 10-15% of throughput -- rather than the severe 33% cap a 600 W part would have implied, and it very likely sits above the card's enforceable floor, so the check becomes a formality rather than a gate. The protective value is worth stating: four cards at 300 W uncapped is ~1200 W, which is roughly the neighbourhood that overwhelmed a 1500 VA unit at FV with only TWO Blackwell cards drawing. Capping to 800 W makes a repeat of today a non-event. Two consequences that follow from Ada independent of power, and both are placement constraints rather than details. sm_89 has native FP8 but NOT NVFP4, which is Blackwell-only -- so the in-house NVFP4 quants that most of this fleet runs will not be accelerated on that colo's cards, and its seats want FP8 W8A8 builds or the NVFP4 checkpoints stay at FV. And it unparks the triton-backend item, which is a hard no on Ampere because fp8e4nv is unsupported on sm_86 and was explicitly deferred to Ada; sm_89 has what it needs. VRAM is 4x48 = 192 GB against fv-ml1's 391 GB, so big-model placement stays at FV. The Flash-Next seat needs 74 GiB resident on one card and would not fit a 48 GB Ada card even with the n-gram table offloaded -- the offload moves the table, not the experts. |
||
|
vh
|
3e61d7d4e0 |
policy: cap GPU power limits at build time — 200 W for the other colo's cards
Operator directive, and the right generalisation of the FV outage: decide the power envelope first and size the cards into it, rather than installing cards and discovering the constraint by tripping it. Four cards at 200 W is 800 W, which fits a real circuit with a real UPS and headroom. Records three things to settle before it is a plan. First, 200 W may sit below the card's enforceable floor: nvidia-smi -pl is bounded by Min Power Limit, often around half of TGP on a high-TGP part, and a sub-floor request is refused -- quietly, depending on how it is scripted. Run nvidia-smi -q -d POWER before any build planning depends on the number. Second, the 600 W figure wants confirming against the actual SKU. The Ada parts do not land there -- RTX 6000 Ada is 300 W, L40/L40S 300/350 W, 4090 450 W -- while 600 W is Blackwell RTX PRO 6000 Workstation territory, so these may be Blackwell or the figure may be a two-card total. Read it off the device rather than a spec sheet. Third, the workload asymmetry is in this fleet's favour: decode is memory-bandwidth-bound and tolerates a cap far better than training does, with a concave perf/watt curve where 60-70% of TGP costs roughly 10-15% of throughput. A cap to a third of TGP is deeper into the steep region; measure it on the first card rather than predicting, and expect prefill-heavy and training work to pay more than a serving seat. And persist the cap. A hand-set limit holds until the next reboot and then silently stops holding, which is the worst shape available given that the thing rebooting the box is likely to be the power event the cap existed to prevent. |
||
|
vh
|
b6335bf6ad |
runbook(fv-outage): the circuit case — split power survives a trip on battery, but only if the colo handoff does
Operator: 'unless of course the thing trips the circuit anyway.' Correct, and it splits into two halves with different answers. A breaker trip is the event the split-power proposal survives: firewall + BMC is 25-40 W on a 1500 VA unit, which is hours of battery, and on a trip the UPS stops being a load-bearing supply and goes back to being what it is for. What it does NOT cover is the colo's own handoff -- their switch, ONT or demarc. If that sits on the circuit we just tripped, the outcome is a firewall running on battery with nothing upstream to talk to and the drive happens anyway. Added as a question for the facility, because it decides whether split power delivers remote diagnosis or merely feels like it does. Records the case where none of it matters: removing an undersized UPS does not remove the constraint, it promotes the next one -- UPS ~900-1200 W to circuit ~1800 W at 15 A or ~2400 W at 20 A. Which side the four-card figure lands on decides everything, which is what makes that single ammeter reading the load-bearing measurement of the visit. Surfaces the lever that may avoid an electrician entirely: nvidia-smi -pl caps per-card TGP, so the box can be made to fit its feed at a throughput cost rather than a rewiring cost. Read nvidia-smi -q -d POWER for the enforced range before assuming how much room the dial has, and persist any cap -- one that evaporates on reboot will hold right up until the next power event and then silently stop holding. |
||
|
vh
|
00b842bb9b |
runbook(fv-outage): operator ruling — undersized UPS; NAT demoted; ammeter protocol for the visit
Operator's reasoning, accepted and better than the hypothesis-space argument it replaces: the NAT change went effective, was verified bidirectional, and then ran correctly for twenty minutes before the site died the moment GPU load was applied. A working config change does not spontaneously fail under an unrelated physical variable. The load correlation is tight; the NAT correlation is merely adjacent in time. Undersized UPS is the only candidate that explains the trigger. NAT material retained as record, and the power.log/uptime check demoted from decision point to free confirmation. Adds the measurement protocol, since the operator is bringing a PDU and an ammeter. The load-bearing caveat: power.log is GPU-ONLY -- nvidia-smi per-card, excluding CPU, 566 GB of RAM, drives, fans and PSU conversion losses -- so the ammeter at the plug is the primary instrument and power.log only cross-checks the GPU share. Four states to capture (idle, one card, two cards, four cards), and capture PEAK rather than average: UPS overload protection responds to short-term overload, so an average-only reading that hides transients will mis-size the replacement exactly the way the present unit got mis-sized, and must be recorded as a floor rather than as the draw. The four-card figure is earmarked for servers/fv-ml1/README.md, because it closes the cutover's own open question -- that the FV circuit was likely specced against half the real draw, back when every record still said the box had two GPUs. |
||
|
vh
|
59ddedd980 |
runbook(fv-outage): a NAT change 34 min earlier means power is not established — and power.log settles it for free
Another session applied a scoped Tailscale SNAT rule to the FV gateway at ~06:22Z, 34 minutes before the site went dark (docs/runbooks/fv-to-ana-nat.md, not my work, left uncommitted). That makes the UPS-overload theory a hypothesis rather than a finding, and nobody should buy hardware on it until the discriminator below has been read. On the evidence that change is the wrong shape to have caused this, and it is recorded as such so the visit is not wasted chasing it: one OUTBOUND SNAT rule scoped to a single source /32 and a single destination /16 cannot stop the gateway, the BMC or the public WAN address from answering inbound; no routes, filter rules, WAN settings or subnet advertisements were touched; pfctl -sr came back byte-identical; and it was verified bidirectional afterwards including ANA->FV SSH with Beszel 18/18 up. Their BMC datapoint used 10.251.50.50, which is not the BMC -- that is 10.251.250.50, a different subnet. They correctly declined to claim BMC health, but the observation is void rather than negative and should not be reasoned from. The discriminator costs nothing and is already on disk: power.log is written locally to /tank every 10 s by a shell loop on the box and does not depend on the network. Entries past 06:56Z mean the machine never lost power, which makes this a routing fault and the UPS innocent; entries stopping at 06:56Z confirm power. Cross-check with uptime and journalctl --list-boots -- continuous uptime across 06:56Z kills the UPS theory outright. So the first action on site is now to READ, not to fix. The two hypotheses lead to completely different remediations and only one of them needs a new UPS. |
||
|
vh
|
524da702ef |
flash-next-mtp-bench: run on GPU 3 / :8023 without touching the production seat
Operator-directed so the campaign could run in parallel with live gen-large testing. Strips the production-restore tail entirely -- the only container the driver can create or remove is fn-mtp-bench, and the sole remaining compose call is a read-only config --format json to derive argv. Adds per-card power+memory sampling every 10 s to power.log, because two cards under load was the risk being accepted and a record beats an argument. That power log is now the only surviving measurement of what the load drew before Fountain Valley went dark; it lives on /tank rather than in the container. |
||
|
vh
|
312725ddfb |
memory: snapshot — Flash-Next seat on one card, and the FV outage that followed
Durable capture so tomorrow's session does not have to reconstruct either half. Built and verified before the power failed: Qwen3.8-Flash-Next serving on a single RTX PRO 6000 with its 51B n-gram table pinned in host RAM and read over CUDA UVA -- 74.36 GiB weights resident, 14.00 GiB KV for 560,654 tokens at the full 262,144 context, 67 GiB host RSS -- plus a gen-large gateway alias verified end to end. The five findings worth carrying: the offload is #54371 (UVA, merged) which supersedes the paused worker-based #53899 and designs out its entire bug family; text_config.ple_embedding_dtype is the load-or-fail discriminator for any community build; --kv-cache-memory makes vLLM SKIP memory profiling and ignore gpu-memory-utilization, which inverts the usual pin-bytes advice and let a 16 GiB pin nearly OOM with no visible failure; MTP is off pending measurement here rather than written off, because the recipe's number is cross-harness and tested k=3 only while the head is one layer run autoregressively; and a container once reported (healthy) with no published port at all, because the healthcheck runs inside the boundary it was trusted to validate. Then the outage. Records it as will-not-self-recover, so no session wastes effort polling a dead site, and carries the three things that change the visit: bypass the UPS rather than using its surge-only bank (both banks share one 12 A inlet -- the surge bank bypasses the inverter, not the current rating), recover power.log before anything else because it is the only load measurement that exists anywhere, and bring seats up one at a time because ten restart:unless-stopped containers loading at once is the largest transient the box can make into whatever just failed. Also records what is still half-done: the stale homepage labels on the 10 containers that died before they could be recreated, which the staged bring-up fixes as a side effect, and the eight drifted stacks plus three untracked host-only stacks that were deliberately left for a deliberate reconciliation. |
||
|
vh
|
d79f10457a |
runbook(fv-outage): UPS overload as leading hypothesis, site-visit bring-list, no-local-fallback correction
Operator's read is that the UPS the box was plugged into overloaded and died, and it fits better than the breaker-trip theory: a UPS's output rating sits far below the circuit's, so it is the first protective device to give -- which explains why the site let go at TWO cards loaded rather than four, and why the ~25 W firewall died with it. Records the operationally important consequence: a tripped UPS resets, an overloaded one can kill its output stage permanently. If it is dead, nothing on site can be reset back to life, so the visit needs the means to BYPASS the UPS or it is wasted. Elevates recovery of /tank/.../power.log to the first action on site. It sampled all four cards every 10 s up to the cut, lives on /tank rather than in a container, and is the only measurement of what the load actually drew -- without it a replacement UPS gets sized by guesswork. Also states that no load figure exists yet, only idle. Corrects an earlier claim of mine in this session: there is NO local fallback for the 19 dark aliases. Probed -- every free local model is on fv-ml1, and irv-ml1 runs no chat seat at all, only TTS/ComfyUI/arbo/clipper work on two partly-occupied Ampere cards. The only non-fv chat backends are paid. Any paid coverage must go under a new opt-in alias name rather than a silent repoint of summarizer/gen/classifier. |
||
|
vh
|
969a1b64a2 |
runbook: FV site dark 2026-09-13 — outage facts, blast radius, staged recovery, OOB design gap
Written while the site is down so recovery does not have to be reconstructed later. Records what was measured rather than what is suspected: every FV address including the BMC is unreachable while all three other sites answer, the campaign's last log line was off_A rep 2 at 06:56:04Z, and the site was dark by 06:58:40Z. Names three candidate causes with the evidence that would distinguish them, because the instrument that could have settled it -- the per-card power log -- died with the box. The two-card-load hypothesis fits the timing and the two prior Anaheim breaker trips on this same chassis, but it is circumstantial and is recorded as such. Carries the recovery hazard that matters: every seat on the box is restart:unless-stopped, so resetting power alone brings ten vLLM containers up loading at once -- the largest transient the box can produce, into a circuit that may have just tripped. Staged sequence given, gen first and flash-next last. Also records the OOB gap the outage exposes: OPNsense-as-subnet-router protects against box-down/gateway-up, and not at all against the site-wide loss that actually happened, because the BMC's only path out is through that same gateway. |
||
|
vh
|
7e62a07341 |
flash-next-seat: full 262K context, KV pinned at a measured 14 GiB, gen-large on the gateway
Operator-directed: raise context to the model's native maximum and take as much KV as the card safely allows, and expose the seat through LiteLLM as `gen-large`. max_model_len 131,072 -> 262,144 KV cache 8.76 -> 14.00 GiB (332,721 -> 560,654 tokens) concurrency 2.54x@128K -> 2.14x@262K ⚠ 16.00 GiB WAS TRIED FIRST AND IS TOO AGGRESSIVE. A 155,497-token non-repeating prefill drove GPU 2 to 97,074 of 97,887 MiB and the caching allocator logged "OOM on device 0 while trying to allocate 488636416 bytes (free: 422117376)" -- 466 MiB wanted against 403 MiB free. The request completed, so nothing failed visibly; that is one step before the shape that crashed stacks/mog-sec twice on 2026-09-10 (~1.04 GiB wanted, ~600 MB free). Backed off to 14.00 GiB, which re-probes clean: zero allocator warnings, a 155,557-token prefill in 14.2 s, and 2,085 MiB still free at peak. The reason the first estimate was wrong is worth keeping, because it is not obvious and it inverts the usual advice: --kv-cache-memory makes vLLM SKIP MEMORY PROFILING ENTIRELY and ignore --gpu-memory-utilization. The profiler was the thing accounting for deep-prefill activation, so pinning bytes switched off the protection that the pin was supposed to formalise. vLLM's own "--kv-cache-memory=18745235968 (17.46 GiB) to fully utilize gpu memory" line is computed from a profile measured at max-num-batched-tokens depth and sits 3.5 GiB above what a 150K-token request survives; open #54764 compounds it, since PLE short-conv prefill pads every request in a batch to the batch-MAX query length. max-num-batched-tokens stays at 8192 -- it is what bounds the activation peak, and doubling max_model_len left the profiled peak unchanged at 1.65 GiB precisely because the peak tracks chunk size, not context length. Gateway: `gen-large` added to the LiteLLM model_list, pointing at fv-ml1:8022. One alias on purpose -- a single alias cannot trip the shared-config enable_thinking mutation footgun, which needs two over the same (model, api_base). Sampling is the checkpoint's own declared set (temp 1.0 / top_p 0.95 / top_k 20); presence_penalty, min_p and repetition_penalty are left unset because the checkpoint declares no canonical value for them. Verified registered for both the infra-ops admin key and the shared all-agents key, since a new model behind a scoped allowlist 403s silently. Also adds services/flash-next-mtp-bench/ -- the MTP measurement campaign and its rationale. MTP stays off, but on "not yet measured here" rather than on vLLM's 4xH100 recipe number, which is a cross-harness comparison and not evidence about a TP=1 Blackwell seat. |
||
|
vh
|
3132a16ca0 |
fv-ml1: finish the renumber the cutover missed -- 16 dead dashboard links
Every fv-ml1 link on the Homepage dashboard was broken. Measured against the
live dashboard API before the fix: 16 entries pointing at the dead 10.250.50.54
and zero at the live 10.251.50.54, covering gen, M.O.G.-SEC, Scriberr, Embed,
Rerank, Reward, Coder, Dockge and six dormant seats.
The miss was structural, not careless. fv-ml1-rename-sweep.sh works from an
allowlist assembled from files that mention the HOST, and a homepage.href label
mentions only an IP -- so every stack whose sole stale reference was a label
fell outside it. The allowlist now covers those 24 files, and records how to
derive the list next time (grep the old address, subtract history) rather than
enumerating from memory.
History is still untouched, and the exclusions are now written down with the
reason each one keeps the old address: recorded benchmark results, whose
base_url is part of a measurement's provenance; the one LiteLLM comment
preserving a retired hand-test endpoint; and the cutover runbooks, where the old
address is the subject matter.
Two bugs found while applying it, both fixed here:
- deploy-stack.sh rejected any stack name containing a dot, so qwen3.5-122b,
qwopus3.5-122b and mistral-medium-3.5 could not be deployed by the script at
all. The check exists to stop path traversal, which means rejecting ".." and
"/" -- not every dot. Traversal is now rejected explicitly and tested.
- stacks/scriberr/.env.example allowed CORS only from the dead IP and from
scriberr.ana.internal, which no longer resolves; the box is at the fv site
and DNS already carries scriberr.fv.internal. The live .env had both stale
origins, i.e. an allowlist with nothing reachable in it.
Host side, applied separately: canonical pushed for the 16 stacks whose only
difference from the host was this renumber, and an in-place address-only fix for
the nine whose host copy has genuinely drifted or has no canonical copy, so that
drift survives for a deliberate reconciliation instead of being clobbered. Every
compose.yaml on fv-ml1 now reads 10.251.50.54. The labels themselves only take
effect at container creation, so the running containers still need recreating.
|
||
|
vh
|
f964a47262 |
flash-next-seat: Qwen3.8-Flash-Next on fv-ml1 GPU 2 with the n-gram table in host RAM
First seat whose weights do not fit its card. The model is 176B total -- a 125B main model plus a 51B n-gram (PLE) lookup table -- at ~6B active per token. The table is a pure embedding lookup, so it lives in pinned host RAM and the GPU reads rows directly over CUDA UVA: ~78 GiB resident on a 95.6 GiB card, 47.7 GiB pinned of 566 GB. GPU 2 and GPU 3 were both idle, so this displaced nothing. Checkpoint dealignai/Qwen3.8-Flash-Next-ABLITERATED-NVFP4 @ be794b99, pinned by revision: NVFP4 W4A4 routed experts, FP8 PLE table, everything else at source precision. Chosen over better-liked builds because its provenance states protocols and repeat counts -- AIME26 pass@1 98.75% over 30x8 repeats with a stated SEM, full-set GSM8K, and a byte-equality audit covering all 31 MTP tensors -- and because it declares text_config.ple_embedding_dtype, which is the field vLLM reads first when selecting the PLE weight format. Builds that ship an FP8 table without that declaration resolve to the unquantized path and fail on load; the README records the check. Requires vLLM #54371 (UVA PLE-offload, merged 2026-09-09T14:32Z), verified by ancestry: the pinned nightly is +150 commits / behind_by 0 from the merge commit. Not in v0.29.0, cut six hours earlier. The older worker-based offload (#53899) is paused upstream and is not the path here -- its deadlocks, ptrace gate and stale-output-under-graphs bugs all came from the separate worker process that UVA does not have. Five deliberate departures from the other seats on this box, each from a measurement rather than a preference, all annotated in place: - no MTP: the vLLM recipe measured it worse at every concurrency on 4xH100 (8-36% less throughput, 32-173% more latency, ~36% acceptance) - modelopt_fp4, not compressed-tensors: only the ModelOpt reader honours the ignore list keeping attention, shared experts, PLE and MTP out of W4A4 - KV left at auto: fp8 KV on this model's QSA path is an unmerged RFC (#54426) - mamba-cache-mode stated explicitly: the model raises on mode "all" - 128K context and 8192 batched tokens, not the native 262K: #54764 and #54919 make depth the risky axis, and sizing to the KV pool has never fixed a depth-driven crash on this hardware Nothing is wired into LiteLLM. Pointing an alias at this seat changes what existing callers receive and is a separate decision. |
||
|
vh
|
a5691ce796 |
restic(esh-vm-db): a backup that exited 0 nightly while keeping an April dump
Work by a parallel session on 2026-09-12; committed here with the rest of the day's changes. Rationale in persistent-memory.d/2026-09-12-esh-vm-db-restic-repair.md. The only visible symptom was a systemd-failed unit from a Sep 6 repository network timeout after boot. The real fault was quieter and much worse: the pre-backup hook logged failures as WARN and returned zero, so pg_dumpall could fail every single night -- it used TCP localhost and wanted a password nobody supplied -- while restic dutifully backed up the stale April 23 dump still sitting in the staging directory and reported success. Mongo was fine, which is part of why it went unnoticed. Postgres now dumps over the /var/run/postgresql socket with peer auth and -w, and both database failures now fail the backup rather than masking it, while still preserving any prior per-DB dump rather than truncating to nothing. An ERRORS counter replaces the warn-and-continue path, and the staging directory is overridable via RESTIC_STAGE_DIR so the new test can exercise it. Adds backup.contract.md, retry.conf and test_pre_backup.py -- three red-green regression tests covering the failure modes above. systemd drop-ins on both jobs add network-online ordering plus Restart=on-failure with a 5m delay and a 3-per-hour limit, which addresses the original boot-timeout symptom. Verified against a real run: snapshot bc5eeaff at 07:01 PDT with a fresh 3.46MB PG dump, retrieved from the repository with decompression and completion marker checked (not a full restore). Repository check passed, 99 snapshots. The old hook and stale dump are preserved root-only at /var/lib/restic/repair-20260912. |
||
|
vh
|
3e727dbeb5 |
memory: snapshot — FV cutover executed, BMC online, four GPUs not two
Captures the Fountain Valley cutover as executed rather than planned: fv-ml1 live on 10.251/16 and serving, BMC recovered after finding it was tagging 802.1q VLAN 250 into an untagged port, and the corrected four-GPU / 391 GB count that invalidates prior sizing assumptions. Also records two things that cost real time and would cost it again: OPNsense write APIs need an X-CSRFToken scraped from a script block, not a hidden form input, and a 403 that coincides with the box going down looks exactly like a successful reboot. Auto-archived 8 settled entries (Recent decisions: 8). The index stays over the 300-line cap at 385 because the guards hold -- nearly every remaining dated entry is under 14 days old or carries open deferred work, and an over-cap file that keeps live decisions beats a scannable one that lost any. |
||
|
vh
|
48e34e1849 |
playbooks: repoint the four remaining live fv-ml1 targets
homepage-regroup, mog-sec-move-to-gpu0, pull-hf-repo and serve-qwen3.5-122b all carried runnable 'scripts/elway ana-ml2 --playbook ...' instructions or the old 10.250.50.54 address. Each would fail today against a dead name and a dead IP, so these are corrections rather than cosmetics. homepage-regroup is renamed to match; the other three keep their names, which never carried the host. |
||
|
vh
|
87d7555fcf |
fv-ml1: the playbook + homepage body edits the rename commit missed
A zsh glob failure in the staging step of |
||
|
vh
|
17586abb19 |
homepage: retire the fv-ml1 docker relay, its precondition is now met
stacks/ana-ml2-proxy existed for one reason, stated in its own README: ESH could not reach the box's Docker API cross-site because of a return-route gap, so ana-docker relayed :2354 -> :2375 from the same subnet. The README named its own retirement condition -- "once it has a working return route to ESH, point Homepage back at it directly and compose down this stack". That condition is met. FV is on the mesh, vb-gateway accepts routes, and the NH3 UDM, ESH UDM and Anaheim FortiGate all carry 10.251.0.0/16 statics toward their local scale nodes. Verified before removing anything: esh-docker-vm opens a TCP connection to 10.251.50.54:2375 directly. Homepage now points at 10.251.50.54:2375 and discovers 112 services across 20 groups, so nothing regressed. The socat container is stopped and removed from ana-docker; the compose file and README are deleted here rather than renamed, because renaming would preserve a moving part that no longer earns its place. |
||
|
vh
|
91bda3c480 |
fv-ml1: complete the cutover — rename, renumber, DNS, and the LiteLLM repoint
The box is physically at Fountain Valley, renamed, renumbered onto 10.251/16, and serving inference again. This lands the repo half of that. Host: hostname ana-ml2 -> fv-ml1, pinned to 10.251.50.54 by a dnsmasq reservation so the address the runbook, DNS and LiteLLM all assume is the address it actually has. Its headscale node is renamed too. The sweep ran from scripts/fv-ml1-rename-sweep.sh, whose allowlist is the reason this diff touches current-state files and not the record. Dated persistent-memory entries, archival-memory and incident notes still say ana-ml2 in 31 and 62 places respectively, because that is what the box was when those things happened. Rewriting them would make the history lie. LiteLLM was the load-bearing piece and needed more than the api_base sed the runbook describes. Twenty api_base entries repointed, but a grep-and-verify pass also caught a LIVE pass_through_endpoints target for the scalar-judge reward route still on the old address -- an api_base-only substitution would have left it dead. Four prose references describing current state were repointed as well; one historical note recording where a hand-test was run is deliberately left pointing at 10.250.50.54. Two facts in the server tables were wrong and are corrected here. The site is Fountain Valley, not Anaheim. And the box has FOUR RTX PRO 6000 Blackwell Max-Q, not two -- verified by nvidia-smi -L and independently by PCI enumeration of four GB202GL devices. That is 391 GB of VRAM rather than 196, which changes what fits on it. DNS: fv-ml1, fv-ml1-bmc and fv-gw added under the fv site via the piggyback approach, scriberr re-homed, and the ana-ml2 records removed. Applied to all three resolvers. The BMC record carries a warning that its 802.1q VLAN tag must stay disabled -- it shipped tagging VLAN 250 into an untagged port, which made it invisible to every network-side diagnostic and is the reason it appeared dead through several cable changes. Verified end to end: summarizer and sec both answer through the Anaheim gateway across the mesh to FV seats on different ports. |
||
|
vh
|
fde7834635 |
docs(fv): on-site playbook for the colo build, and the redundancy it does not have
fv-ml1-cutover.md covers what to change; this covers the on-site hour, where the binding constraints are limited time and not locking ourselves out of a site an hour away. The load-bearing correction is that fv-ml1 running tailscale is NOT a second independent path. Its default route is the OPNsense LAN address, so its node needs egress through OPNsense to reach the control plane; OPNsense down means fv-ml1 is unreachable by construction. An earlier version of this plan claimed otherwise and proposed proving it by rebooting the firewall, which would have failed on site with the operator standing at the rack. He caught it. What the fv-ml1 node does buy is narrower and still worth one command: it covers control-plane lockout -- a bad admin rule, a WebGUI on the wrong interface, a lost password -- where the box still routes but cannot be managed. It does not cover data-plane failure. Those are different failure classes and conflating them produced the wrong test. The gate is now a deny-admin-from-mesh test run while the console is in arm's reach, which exercises the bypass that actually exists. Deliberate deviation from the cutover runbook: flat /24 with the BMC at 10.251.50.50 rather than a mgmt VLAN at 10.251.250.50. The VLAN needs OPNsense config plus switch tagging during the scarcest hour, for something addable remotely in ten minutes. Also records the DNS chicken-and-egg (the mesh needs public resolvers because the fleet AdGuard is unreachable until the mesh is up) and the interface auto-assignment trap measured on the lab VM the same night. |
||
|
vh
|
2dd459d2e5 |
memory: Worldtree memory-split flip protocol, and all three deployments measured writable
worldtree-dev's U6 reader refuses at boot if it cannot append and read back <memory root>/reader/canary.jsonl. Per-euid subdirectories are lazy and only warn, so that root canary is the single boot-blocking check -- which makes the memory root's writability by the container uid the precondition worth knowing before a flip rather than during one. Protocol agreed with worldtree-dev: neither memory.reader.enabled nor memory.writer.enabled gets flipped on any deployment without infra-ops confirming that writability first. Both ship dark until the operator schedules the tracer skeleton. Measured tonight on corviduo-dev, all three pass. Also retracts a wrong prediction I sent earlier in the thread: personal runs as root, not uid 1000, and pinned is the only uid-1000 deployment -- it passes regardless because /data/state is owned 1000:1000. Recorded with the caveat that a permissions reading is a claim about its own date, so the probe gets re-run immediately before any flip rather than cited from tonight. |
||
|
vh
|
ddfa1a6e5b |
feat(beszel): commit the fleet agent rollout — configs, playbooks and runbooks
Beszel agents are installed and verified across the fleet but the artifacts that produced them were never committed, so the deployment existed only on the hosts. Adds the per-host agent environment files (PORT, NICS, EXTRA_FILESYSTEMS and the hub's PUBLIC key), the systemd unit, the guest install script, the Synology compose, and the elway playbooks for native, guest-stage, guest-install and Synology paths. The two dated memory detail files covering the priority-1 and priority-2 waves ship alongside, per the convention that memory lands with the work it describes. No credentials here. The KEY= value in every host env is the Beszel hub's public ed25519 key, identical across all nine and public by design; the agent README says so explicitly. The nh3-nas sudo password referenced in the runbook prose lives in Vaultwarden and the helper scripts named there never contained it. ⚠ Overlapping VMIDs across hypervisors are a standing trap and are recorded in the priority-2 notes: pfi-pve 105=postgres and 100=pbs-ana, nh3-pve 105=pbs-nh3. ⚠ PBS-NH3's export was ~75.5% used at capture; resource checks are not job success monitoring and should not be read as such. |
||
|
vh
|
88e171bea6 |
fix(esh): Plex hardware transcoding on the Arc A580, and the two ways it hid
Plex never hardware-transcoded on esh-pve-nas LXC 105 despite correct passthrough, cgroups, group membership, authenticated HuC firmware, a lifetime Plex Pass, HardwareAcceleratedCodecs=1, and the Arc already selected as HardwareDevicePath. Root cause sat below all of that: intel-media-va-driver 22.3.1 (Apr 2023, stock jammy) predates Arc/DG2 support and exports only __vaDriverInit_1_14, against the libva 2.22 that Plex bundles and loads via RPATH rather than the system one. A half-finished prior attempt at the same fix was also present -- libva and libva-drm hand-installed at 2.22 with libva-x11 left at 2.14, breaking every X11 VA-API consumer on va_fool_postp. Fixed with Intel's client-GPU repo: iHD 24.3.4 (__vaDriverInit_1_22, an exact ABI match) plus a consistent libva 2.22.0.2-87 set, which also brings the orphaned manual install back under dpkg. The repo track is rolling, so the six packages are pinned in /etc/apt/preferences.d/intel-gpu-pin and apt-mark held; verified by a simulated upgrade that moves 152 packages and touches none of them. Two findings worth more than the fix: pct snapshot refuses on a guest with a bind mount and still exits 0, so a script guarding a change with it proceeds without the rollback point it believes it has. The ZFS dataset snapshot is the working path, verified by reading it back. A synthetic Plex Transcoder invocation is not a valid test of Plex's transcode path. Plex bundles its own libc among 61 libraries; the harness produced three distinct failure modes that were artifacts of not reproducing that runtime, and it failed identically before and after a fix that worked. With no positive control its negatives carried no information. Only a forced transcode settles it, and PASS is recognisable by Plex naming the device. The original empty decoder/encoder line was an absence of evidence rather than evidence of failure -- TranscodeSession was 0. Jellyfin LXC 107 has the same stale stack and the same Arc available; left alone per the operator, and it ships its own ffmpeg so this may not transfer verbatim. |
||
|
vh
|
e01484f6d0 | memory: consolidate backref lines to true archival counts (was triple-counted) | ||
|
vh
|
19aeaedeba | memory: snapshot — fv-ml1 cutover prepped, Anaheim recovered, BabyYarros complete; archived 5 settled entries | ||
|
vh
|
1d435c9e33 | memory: ana-ml2 -> fv-ml1 Fountain Valley relocation + leave-rack-dark decisions | ||
|
vh
|
8400f3aef5 | scripts: pre-script the ana-ml2->fv-ml1 rename sweep (scoped, dry-run default, history-safe) | ||
|
vh
|
ce04f9dbde | docs: finalize fv-ml1 cutover — confirmed addresses, piggyback DNS, exact commands | ||
|
vh
|
a13ab598ae | docs: fv-ml1 cutover runbook (ana-ml2 -> Fountain Valley colo) | ||
|
vh
|
1405363649 | memory: Anaheim recovery — crowdsec/traefik fail-closed 403 fix, ana-ml2 still needs power | ||
|
vh
|
870f42960e | memory: BabyYarros complete — both arms trained + evaluated, voice moved, beats 9/10 | ||
|
vh
|
5558d9c7d3 |
R49/BabyYarros: voice + beat evaluation tooling and the seat-free delta_cb
Adds the Yarros-side evaluation the training exists to justify: does the adapter move arbitrary prose toward Yarros, and can the instruct arm still expand a beat to a paragraph on direction. Yarros-flavoured voice prompts (modern/neutral/ romantasy tiers so any Yarros voice in the modern tier is adapter-attributable, not prompt-supplied) and a Yarros-register beat SYS on the chat generator. voice_distance.py is the honest slice of adjudication that needs no seat: Burrows's Delta over character bigrams against held-out Yarros. Its first cut mis-framed the noise floor — it used the same-author distance (held-out vs itself) as the between-arm significance threshold, which is the target, not the threshold. Fixed to the measured floor: the within-arm seed spread, which is this metric's sampling variance at this sample size, computed from the two seeds already generated rather than assumed. Result on the built corpus, ordering base-125-tuned < instruct-tuned < base-unadapted, both adapters clearing the 0.046 measured floor (base +0.157, instruct +0.076), and the ordering corroborating the independent held-out loss ordering (Base below Instruct). One seed-pair per arm, so it corroborates rather than settles; the full frozen adjudication still needs a romantasy control panel, a second seed, and the gen seat for the beat-incumbent leg. |
||
|
vh
|
61ddcfd9a6 |
CLAUDE.md: remind this session it IS infra-ops, not a party to page
The recurring incident-time failure: during an outage the session reaches for althing to notify infra-ops, and since ALTHING_HANDLE is infra-ops the send mails itself, the herald pokes about new mail from infra-ops, and the session triages its own words as a peer report. Documented the identity, the check ($ALTHING_HANDLE before addressing an ops message), and the one legitimate use of the infra-ops thread during an incident (a durable memo to a future infra-ops session on recovery, not a page to a peer). |
||
|
vh
|
fa6e2c8228 |
memory: Sentinel-R3 pulled, MTP-grafted, quantized — a better-fit pen-test seat candidate
Records the comparison (Sentinel-R3 is a real agentic-pentest SFT on the same Qwen3.8-27B base vs M.O.G.-SEC's persona-on-stock-weights, on a tool surface that matches our own harness), the HF check (M.O.G.-SEC repo unchanged), and the graft: Sentinel ships no MTP head, so the verbatim base head was grafted in, which is the correct lineage but leaves acceptance unverified because the head now sits on an SFT-finetuned body rather than stock weights. The quant is done and structurally verified; serving and the acceptance/A-B gate are blocked on GPU space and are the operator's next call, along with the proprietary-license posture. Detail in the artifact's PROVENANCE.txt. |
||
|
vh
|
5ccb6c3118 |
memory: concurrent training on pfi-gx10 is net slower than serial
VRAM said yes and the box said no. Two 4B LoRA jobs fit easily -- 36 of 121 GiB with 98 free -- but both dropped from a 37.10 s/it solo baseline to ~85 s/it, 2.29x each, so combined throughput fell below serial. The GB10 is past its memory-bandwidth roofline and pays a contention penalty on top of the split. Worth keeping because the intuition it corrects is a common one: free GPU memory is the number everyone reaches for, and on this box it does not predict whether a second job is affordable. The control is what makes it a finding rather than an anecdote -- killing the second job returned the first to 37 s/it on the next step. |
||
|
vh
|
38eb40ace4 |
BabyYarros: stage the 4B-Base comparison arm, both concurrent and chained
One arm is not a comparison. On the Brontë corpus Base won held-out loss (2.814 against the instruct arm's 2.908) while instruct won instruction-following, and that trade is the open question for Skaldsong. It has to be re-measured on this corpus rather than carried over: BabyYarros is 12% larger and contemporary. Two entry points because the right one depends on a measurement rather than an argument. The box has 98 GiB free against an 18.4 GiB training footprint, so VRAM is not the constraint -- but VRAM was never the binding constraint on a GB10 that onboarding measured at 6x slower than ana-ml2 where compute predicts 2.7x, and where batching was not a throughput lever. If the box is already at its bandwidth roofline, a second job splits the same bandwidth and buys nothing. So: launch-yarros-4b-base.sh takes an explicit --allow-shared-gpu that bypasses the GPU-clear guard, and chain-yarros-4b-base.sh waits on the instruct run and refuses if it produced no adapter. The bypass is an argument and never a default, because the guard's normal job -- stopping a chain from firing into a live run -- is what keeps one lost run from becoming two. A shared-GPU launch stamps its own log with a warning that its s/it is not comparable to a solo run, since the harness is part of the number. Both carry the same gate guard as the instruct arm: refuse to start unless the leak gate report on disk says PASSED. |
||
|
vh
|
969d649a24 |
memory: BabyYarros unblocked — gate passes 0/325, instruct training live on gx10
Records what the gate cost to build and what it found: a second typography defect (small-caps epigraphs and drop caps, which the D1 "no unwrap needed" note was right about in the wrong direction), back matter inside the prose of all five works naming real people, a possessive-only entity that was unrenamed and unreported at once, a character lost to an in-book article typeset in lowercase, and a phrase class the unigram scan structurally cannot see. Also records the two things that will matter next time: every new detector flag is opt-in and the Brontë map was re-derived and diffed after each change, because one intermediate version silently dropped four real names from Shirley; and the gate's sensitivity floor is part of its result, not a footnote. |
||
|
vh
|
7b0580dcbe |
BabyYarros: the leak gate passes, and it found three defects nobody was looking for
The gate is new. There was no committed instrument for "does any of the author's own proper nouns survive the rename" -- the Brontë number was produced by hand -- so leak_gate.py is now that instrument, and it runs both directions every time: the same scan over the unrenamed source as a positive control, and a nonce string as a negative one. A detector that only ever sees renamed text cannot distinguish absent from blind. Run against BabyYarros as built it reported 212 surviving entities, not the 86 recorded earlier, because it scans the whole corpus rather than each work separately and it counts the sub-threshold entities rename never looked at. Three findings came out of closing that. The corpus had a typography defect of its own. The D1 notes correctly say no unwrap was needed; a different defect was there instead. The Empyrean books set their chapter epigraphs in small caps and the extractor rendered the run as uppercase while leaving the large initial as a separate token, so the corpus carried "M AJOR A FENDRA'S G UIDE TO THE R IDERS Q UADRANT" -- 106 lines, ~700 splits -- plus 52 drop caps like "T he flight field". That is where the entities called IDERS, UADRANT, NAUTHORIZED and seventeen bare single letters came from. A split initial next to an uppercased run is enough to recover the original mixed case, so the restore is exact rather than approximate: a word with a split initial was capitalised, an all-caps word without one was lowercase. Back matter was inside the prose. The builder splits on chapter headings and nothing follows the last one, so every work carried its acknowledgments, newsletter pitches and cover-artist credits -- 4,555 words naming the author's agent, editors and children, in a corpus whose entire purpose is that no identifiable name survives. And the gate passed at 0 of 314 while Afendra was still in every copy. The name never appears unpossessed, so it keyed as an apostrophe form, and rename and the gate both skip those as contractions -- unrenamed and unreported at once, which is the worst failure shape available. Baxter escaped a different way: wilder renders an in-book news article entirely in lowercase, putting the cap/lowercase ratio at 0.13 against a 0.05 bar. Then a second class the unigram scan structurally cannot see. Riders Quadrant, Flame Section, War Games and Fourth Wing -- the book's own title -- are built from ordinary words the detector correctly refuses to call names. The gate now audits recurring capitalised 2-3grams against an explicit allow list, and rename applies a phrase map after the entity pass. Every new detector flag is opt-in and off by default, and the Brontë entity map was re-derived after each change and confirmed identical in keys, surfaces and every field. The stoplist was built by reading each surface in context, which is why it is short: Violence is Xaden's nickname for Violet, and Continent, Presentation, Barrens, Originals, Montserrat, Athena, Aura, Curator and Sage are all in-world. A plausible-looking guess would have excluded most of them. Final: 0 of 325 entities and 0 of 91 audited phrases survive in any of 30 copy files, both controls passing. The sensitivity floor is stated in the gate's own output -- 3 occurrences for a name, 5 for a phrase -- because a negative without one is unfalsifiable. |
||
|
vh
|
16c144fcda | memory: snapshot — BabyYarros blocked on the leak gate; R49 sweep complete; mog-sec settled | ||
|
vh
|
6dba912324 |
BabyYarros: corpus built, gender resolution fixed, rename blocked on leak gate
Located the source: five Rebecca Yarros works in the Kvasir licensed library, with rights recorded as gated. Built D1 at 208 chapters and 780,744 words, which is 15% larger than the Brontë corpus. No unwrap step was needed because Kvasir's cleaner already emits flowing paragraphs, so the hard-wrap defect that cost a re-cut on Brontë does not exist here. The alphabet was re-derived rather than inherited: 23 non-ASCII letters across three forms, against F02's 4 on a smaller sample. Same ASCII-fold conclusion from a different measurement, which is the reason to re-derive per corpus. The interesting finding is a new pathology. In a rotating first-person POV corpus, every book's narrator gets the wrong gender. Measured against six names verified in the text, the pronoun resolver called Violet male, Leah male and Landon female -- three of eighteen wrong, and all three are the narrator of the book where they were misgendered. A narrator is "I" in her own book, so her name appears mostly inside the other lead's dialogue surrounded by his pronouns. This is Brontë's "Jane called male" amplified by rotating POV. Title-first resolution, which fixed it for Brontë, is nearly blind here because contemporary romance uses given names rather than honorifics. What works is the POV header: resolve each name from the chapters it does not narrate. Validated at 9 correct, 9 held, 0 wrong against the previous 7, 8 and 3 wrong, and the instrument refuses to write unless it beats what it replaces. Re-pointing rename.py surfaced three bugs, two of which would have silently corrupted the corpus. Gender came only from honorifics and the entities file's gender field was ignored, so the POV fix had no effect until wired through; that took wilder from 1 gendered entity to 13. The pool labels were hardcoded in a print statement, so any non-Brontë preset crashed. And the collision-filter log claimed it dropped names colliding with Brontë entities regardless of which corpus it filtered against -- the logic was right but the message named the wrong corpus, which is how a reader later concludes the filter ran on the wrong thing. D3 is blocked and nothing has been trained. The leak gate shows 86 of 232 renameable source entities surviving where the Brontë run reached 0 of 203. It decomposes into detector false positives that need a stopword filter rather than renaming, genuine misses among worldbuilding proper nouns, and a third class whose cause is not yet established. Training before the gate passes means fitting in-copyright text with 86 identifiable source entities intact, in a corpus F02 already flagged as small enough for leak to be a real concern. |
||
|
vh
|
e15c5ee5ea |
Instruct probe: voice and instruction-following coexist
Trained the same corpus onto Qwen3-4B instruct rather than -Base, with seed, steps and token count held so the carrier is the only variable. The chain picked checkpoint-150 by loss automatically, applying the lesson that cost a re-cut on 4B-Base. The central risk did not materialise. The assistant prior did not block the voice: curly quotes land at 16 of 18, identical to the 4B-Base tuned arm, against 1 of 18 on the unadapted control, and task-leak is 0 of 18 where the base carrier leaked 4. Instruction-following also survived raw-text training -- 10 of 10 on-beat through the chat template, the same as the untuned control. The cost is length discipline rather than comprehension. In-band dropped from 10 of 10 to 6 of 10 and the median went from 124 to 140 words. Training on Victorian prose made it wordier, which is a soft degradation and not a break. Held-out sits at 2.908 against 4B-Base's 2.814, and it plateaus without turning where the base carrier overfit at step 75. The assistant prior competes for capacity, so the instruct carrier absorbs less rather than overfitting more. What raw-continuation training does not fix is the plot furniture. The tuned instruct arm renders the beat and then drags the referent -- "He licked her clean... my master thus, my husband thus", turning the dog into a man, because the corpus is about masters and husbands. Another beat ran to 247 words and gave the narrator a list of duties. That is precisely what instruction-pair training addresses, since pairs teach render-this-and-stop where continuation teaches keep-writing. The probe de-risks the instruction-pair path without substituting for it. One metric note against future misreading: ran_on reports 10 of 10 on both arms and is uninformative on this job, because a single paragraph contains no blank line for it to find. |
||
|
vh
|
d9cf486004 |
Settle the Skaldsong architecture question with a ten-format negative result
Skaldsong's requirement is beat-sentence in, paragraph out, stitchable. That is narrower than free-form continuation and it fails differently, so it was tested directly rather than assumed from the earlier prompt work. The adapted 4B carrier cannot do it. Ten prompt formats across three seeds each -- bare, paragraph-break, labelled, epigraph, one-shot, one-shot unlabelled, three-shot, "it happened in this way", "I remember the whole of it", and the letter prompt's label-and-begin move applied to a beat -- produced thirty samples and not one that reliably renders the beat. They drift, they frame, or they truncate. The cause is structural rather than a shortfall of the adapter: "write a paragraph about this sentence" is an instruction, and a completion model has no mechanism for "about". It continues the text it is handed. Two formats failed worse than drifting and leaked pretraining task data. The paragraph-break format emitted an NLI multiple-choice item, and label-begin emitted a grammar-correction exercise. A standalone sentence followed by a blank line looks exactly like a dataset entry, and style adaptation does not remove base-model task artifacts. The same five beats through the gen seat with a style instruction and no adapter landed all ten samples inside the requested 90 to 140 word band, every one on its beat, with no drift into a following scene. The voice, however, is generic literary pastiche -- abstract-noun-heavy, fond of aphoristic openers, and over-written -- rather than Brontë. Voice without direction against direction without voice. The product needs both, which makes rebuilding the corpus as instruction-response pairs on an instruct carrier the evidence-backed option rather than more prompt cleverness. This applies to Yarros identically, because the carrier question is orthogonal to the author, so the next corpus should not re-run this experiment. Stitching also has its own failure mode, shown in the booth's third panel: the paragraphs were generated independently and the point of view slides between them, so by the fourth beat the narrator is both watching the girl carry the animals and carrying them herself. A real stitcher has to feed prior paragraphs back as context, which means the instruction-pair corpus wants multi-paragraph continuity examples and not only isolated beat-to-paragraph pairs. |
||
|
vh
|
8787daf04f |
Find the prompt recipe that works on a completion carrier
Two operator prompts settled the question the handoff prompt opened. The winner labels the artifact and then begins it: "This is the letter I wrote verbatim, my two short paragraphs, detailing the time I saw the mangy gray dog meet and then lovingly and tenderly lick a calico kitten: Auntie, You'll never believe what I saw--". Two of three seeds delivered the actual event in first person, and one is the strongest output of the sweep -- the dog following her to the mill, the kitten caught in a hedge, the dog carrying it out and laying it in her lap before licking it. Coherent arc, correct subject, no gloom-override, no meta-frame. It also reproduced the Gutenberg underscore-italics convention. It works where the handoff failed for a specific reason. The handoff could be satisfied by narrating compliance, because the letter did not yet exist. Naming the artifact and starting it leaves nothing to narrate around, so the only continuation is the artifact. The chapter-heading prompt refuted a hypothesis of mine. I predicted that rendering the title in the corpus's own conventions would make it land harder than the operator's inline form. It did the opposite: both corpus-form seeds ignored the title and opened unrelated scenes, while the inline form at least finished the heading and wrote a chapter about the story. Corpus chapter titles are short and decorative, so a long descriptive one in that slot reads as decoration to skip. A label only instructs if the model treats that slot as load-bearing. That prompt also surfaced an unnoticed consequence of the D2/D3 rename pipeline: given "Alice", the adapter produced "Alexander the Alexander, as he was known in Little London". The corpus was entity-renamed from a French/English pool, so the adapter learned that character names come from that pool and rewrites outside names into it. Callers cannot reliably name their own characters at prompt time. That is not a defect of the rename, which exists to prevent memorising Brontë's cast, but it is a usability constraint worth stating. |
||
|
vh
|
e47295d2ce |
Re-cut the 4B arms from checkpoint-75 and test an in-fiction instruction
The 4B tuned arms now come from checkpoint-75, the actual loss minimum at 2.813826 confirmed from loss-series.json rather than from reading the training log. Only the tuned arms needed re-cutting; the base arm never touches the adapter. One surprise in the comparison: step-75 and end-of-run differ on typography rather than voice. Curly quotes read 16 of 18 against 17, collapse is 0 either way, but the hard-wrap ratio is 0.33 at step-75 against 0.12 at the end. Further training washes the residual line-break habit out while held-out loss gets worse, so best loss and best typography are different checkpoints. Neither is near the original 0.85 defect and the corpus's own residual from preserved verse is 0.25. The operator's second prompt embeds the instruction inside the fiction: Abernathy tells the tale badly and then asks the narrator to retell it in a few short paragraphs. Across six seeds the model acknowledged the handoff every time and never once performed it. It wrote "I told it, briefly, to his satisfaction", and "So I wrote it out, and kept it in my pocket-book", and one seed negotiated the brief in character, objecting that it could not condense the story into a few short paragraphs because there were too many points to touch. The reason is structural rather than a failure of the adapter. In a novel, "she retold the story" is an ordinary sentence, so the likeliest continuation of a request is narration of compliance rather than compliance. A trailing paragraph break did shift the behaviour -- one seed opened in the narrator's own quoted speech -- so typography is a real lever, just not a sufficient one. That is direct evidence on the instruct question raised earlier: if the product is asking for a scene and getting the scene, no amount of in-fiction framing substitutes for a post-trained instruction-follower, which favours rebuilding the corpus as instruction pairs over more prompt cleverness. |
||
|
vh
|
25eb65e23b |
Complete the R49 carrier sweep: 4B closes the continuity gap and overfits
The three rungs now sit on the same unwrapped corpus, seed, step count and token count, so carrier size is the only variable. Held-out loss reads 3.329 at 0.6B, 3.018 at 1.7B and 2.814 at 4B -- deltas of 0.311 and then 0.204, diminishing but still real. 4B answers the question the rung existed for. Scene-level continuity holds: on the office-kitchen prompt it produces a named character with motivated dialogue, a spatial layout the narrator navigates, and a physical description, all in one passage, where 1.7B wrote pretty but eventless prose about opening doors and looking at stars. On the letter prompt it opens the letter, promises to quote it, and then quotes it across a paragraph break. Voice saturation is also the best of any rung: curly quotes 17 of 18 against its own base arm's 1 of 18, and collapse 0 of 18 against 4 of 18. Two findings that change earlier conclusions. 4B is the first rung to overfit inside one epoch. Its series runs 2.832, 2.816, 2.814, 2.820, 2.824, 2.825, 2.825 -- a minimum around step 75 and then a turn. Both smaller rungs plateaued without turning, so the optimal epoch count shrinks as the carrier grows and my earlier "one epoch is right for this corpus" holds only for the small end. The consequence is operational: the adapter directory holds the end-of-run weights at 2.825 rather than the step-75 best at 2.814, and it exists as a recoverable checkpoint only because save_steps was set. The voice arms were cut from the end-of-run adapter, so the booth understates 4B slightly. The tone-override also appears to close. On the operator's frame prompt asking for a wonderful story, 1.7B held the frame on every seed but killed the animals on two of four; 4B kept them alive on both seeds, and one of them had the narrator doubt the story he was told and supply a parallel childhood memory to explain the doubt. That is a narrator with an interior position on the tale. Two samples per arm, so directionally right rather than established. |
||
|
vh
|
4cee4b179d |
Launch the 4B rung and record that voice transfer carries subject matter
Rung 3 of the sweep is training: Qwen3-4B-Base, one epoch, seed 4919, the same unwrapped corpus, 159 steps at roughly 37.8 s/it for about 100 minutes. Carrier size remains the only variable across the three rungs. A two-arm voice test is chained behind it, gated on the adapter existing, so the rung is readable when it lands rather than needing a second round trip to become informative. The finding worth keeping came from an operator prompt rather than from my test set. Asked for a wonderful story about a stray dog licking a calico kitten, framed as reported speech, the 1.7B adapter held the frame across all four seeds and was unmistakably Victorian throughout -- and two of the four turned the premise into animal death, with the cat licking the puppy to death and the narrator's interlocutor answering that he wished they were all dead. That is not a coherence failure. It is Charlotte Brontë's preoccupations arriving alongside her sentences: Jane Eyre opens on a beaten child, Helen Burns dies, and Villette is saturated with grief. Train 680k words of that into a small carrier and a request for something wholesome reaches for mortality. My nine test prompts were all emotionally neutral and structurally could not have surfaced this. The operator's prompt found it on the first try, which is worth recording as a property of the test set rather than of the adapter. The implication for the regime is that "voice transfer" includes tone and subject, so wanting the voice without the gloom is a corpus-selection or prompt-framing problem rather than a training-length one. Also observed: one seed closed its anecdote and began "CHAPTER XIX. THE CHILD'S CLOSET", having learned book structure unprompted. The base control on the same prompt went modern and essayistic, so the shift is the adapter rather than the carrier. |