vh
d710e56aca
Split the 53 over-threshold dated log entries (Recent decisions, Tried and abandoned) into per-entry persistent-memory.d/<slug>.md detail files, leaving one-line pointers in the index; the 10 short entries stay inline. Startup index drops 60,527 -> 25,256 bytes (492 -> 270 lines); entry bodies move verbatim to on-demand detail files, so a fresh session loads ~25 KB instead of ~60 KB and pulls a detail file only when its pointer is relevant. Top matter (Repo purpose, Tools & conventions, Current state) is unchanged; both archival back-references preserved. CLAUDE.md persistent-memory section now documents the index<->detail read discipline (read the index, pull details on demand, never bulk-read the dir, commit both together). Auto-archival still held every dated entry back (all <30 days old); the July burst begins aging past the 30-day guard ~2026-07-31.
1.2 KiB
1.2 KiB
[2026-07-14]MTP-on-modelopt: NO checkpoint config skips the spec-decode drafter's quant (vLLM 0.24 bug) — 4 config attempts failed before the runtime workaround. All crashed the same way (qwen3_5_mtp.py:256param_data.shape == loaded_weight.shapeAssertionError — bf16 mtp head loaded into a quantized drafter param): (1) mtp excludes inconfig.json(WRONG file — vLLM modelopt readshf_quant_config.json); (2) specific-unfused mtp names in hf_quant_config; (3) wildcardsmtp*/mtp.layers.0*(is_layer_skippedis EXACT-membership, NOT glob — wildcards match nothing); (4) exact fused+unfused names in bothmtp./model.prefixes. Instrumentingis_layer_skippedproved the drafter's exclude list holds ONLY the main model'slinear_attnentries — the mtp excludes never reach the draft-model quant config. ONLY fix = a mountedsitecustomizeforce-skippingmtp.*. LESSON: don't chase checkpoint-config fixes for the mtp-drafter crash; go straight to the runtime patch. Alsonvidia-modelopt[hf]==0.43(AEON's producer version) is a trap — it pins transformers back to 4.57 which can't loadqwen3_5at all; use 0.45 + the FusedMoE guard inquant_modelopt.py.