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

docs(sampler-defaults): fix char-rp-reasoning seat → Deckard-PKD (was stale QwQ-RpR-v4)

Browse Source 
The live gateway config has served char-rp-reasoning as deckard-pkd-27b (:8018)
since the 2026-07-08 A/B; the standalone doc had frozen on QwQ-RpR-v4. Corrects
seat 4 (backend + samplers + server-side DRY/reasoning-budget notes).

Also snapshots session state in persistent-memory.md: phantom-qwen verified
already-clean, ana-docker docker log-cap (logrotate copytruncate, no bounce),
and the granite→gen memory_extractor bind live on demo+personal.
This commit is contained in:
vh
2026-07-09 14:09:13 -07:00
parent 6081319743
commit 8894854127
2 changed files with 56 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/pfi/model-sampler-defaults.md
+8 -6
View File
@@ -55,17 +55,19 @@ Also governs **summarizer-large** (shares the same `qwen3.6-27b-aeon` @ :8015 de
*Source:* dvalin canonical (Mistral-Small RP prose) **A/B-validated by infra-ops** on the live serve. *Changes:* temp 1.0 → 1.1, min_p 0.03 → 0.10. **min_p 0.10 richened imagery vs 0.03** with no incoherence at temp 1.1. **repeat_penalty 1.05 was REJECTED** — in the A/B it injected a stray markdown title into a grief scene; rep-style penalties hurt Drummer/Magistral RP creativity (matches the model card and dvalin's own note). Alt prose model: `MS3.2-PaintedFantasy-v4.1-24B` (swap via the `char-rp-gguf` stack `.env`).
### 4. char-rp-reasoning — QwQ-32B-ArliAI-RpR-v4 (llama.cpp, reasoning RP)
### 4. char-rp-reasoning — Qwen3.5-27B-Deckard-PKD (llama.cpp, managed-reasoning RP)
| param | value |
| param | value (request-level) |
|---|---|
| temperature | 1.0 |
| top_p | **0.95** |
| top_p | 0.95 |
| top_k | 40 |
| min_p | 0.02 |
| repetition/DRY/XTC | **off** |
| min_p | **0.05** |
| presence/repetition | **off** |
| DRY | **0.8 server-side** (base 1.75 / len 2, dry-after-temp) — not a request param |
| reasoning-budget | 400 (server-side) |
*Source:* ArliAI RpR v4 card — explicit **NO** rep/DRY/XTC penalties. *Change:* added explicit top_p 0.95. Reasoning is server-side (`--reasoning on`, budget-capped); the CoT surfaces in `reasoning_content` with clean prose in `content`. SillyTavern wiring (non-sampler): `include names = never`; exact `<think>`/`</think>` tokens.
*Source:* dvalin-CONFIRMED canonical 2026-07-08 (thread 01KX1Y7P). **Corrected 2026-07-09:** this seat had lagged on QwQ-RpR-v4 — the A/B on 2026-07-08 replaced it with **Deckard-PKD-Heretic i1-Q5_K_M** (DavidAU, Qwen3.5-27B, :8018); the live gateway was always Deckard. Deckard won on brokkr's frozen scorer (0/30 loops, 0/30 refusals) over RpR-v4 (1/30 loop, forbids DRY) + Pantheon-27B (7/30 refusals). Reasoning ON server-side (`--reasoning on`, budget 400); CoT surfaces in `reasoning_content`, clean prose in `content`. Tuning ladder: flat prose→min_p 0.08, loops→DRY 0.9, over-damped→DRY 0.6/off. **Do NOT import RpR/QwQ sampler rules** (different family — QwQ hated DRY; Qwen3.5 benefits from it).
## Changing a default