From f3dfea13ad1a0e8b87d0560eb470717f1b23da66 Mon Sep 17 00:00:00 2001 From: Frank Date: Thu, 8 Oct 2026 04:50:23 -0700 Subject: [PATCH] swift15: KLD re-PASS at production MNBT 2048 (KL 0.0131, top-1 97.92%) --- .../LLM/frank-primary-radiance-swift15-w4dn8.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/serverAppConfigs-docs/LLM/frank-primary-radiance-swift15-w4dn8.md b/serverAppConfigs-docs/LLM/frank-primary-radiance-swift15-w4dn8.md index e990347..913fc9b 100644 --- a/serverAppConfigs-docs/LLM/frank-primary-radiance-swift15-w4dn8.md +++ b/serverAppConfigs-docs/LLM/frank-primary-radiance-swift15-w4dn8.md @@ -118,6 +118,13 @@ MiB measured**); the serve size floor `FRANK_PRIMARY_RAD_MIN_MIB` moves refused), KV 2048, headroom 512, no speculation. The engine ran rank 0 fully on card 0 (25.88 GiB) with rank 1 host-side in this shape — scoring is placement-invariant; scored at 28 positions/s. +- **KLD gate re-PASS (2026-10-08 04:49) at the production-legal shape**: + MNBT 2048 (the w4 file-tier law: experts stream in place only at + MNBT ≤ 2048 — the recording's 32768 demands a widened form the disk + tier refuses), pool 40960, vram-kv 3072 fp8, tp 2, headroom 768. + **mean KL 0.0131, top-1 97.92 %**. The Δ vs the first PASS (0.0015) is + the float-order noise band of the chunk-size change — the verdict is + chunk-size invariant, ~2.3× inside the 0.03 bar on both shapes. - **Throughput gate: PASS (2026-10-08 05:40)** — decode at 44.5k depth (streamed chat probe, temp 0, 512 tokens, salted): arm A (main serve shape, EVC 0.82) **146–173 t/s**, engine timings **121.9 t/s**,