Kral 542fc8ec31 B: stage 2 data analysis (repair taxonomy, teacher vs Qwen behaviors, duplicates, empty_response); empty_response fix (cap 24000, retry temperature, stream guard) ...

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

2026-10-06 05:56:23 +02:00

..

data-analysis-stage2.md

B: stage 2 data analysis (repair taxonomy, teacher vs Qwen behaviors, duplicates, empty_response); empty_response fix (cap 24000, retry temperature, stream guard)

2026-10-06 05:56:23 +02:00

devir-notlari.md

STATE/handover: baseline restarted on the 11-task subset

2026-10-03 22:36:30 +02:00

empty-response.md

B: stage 2 data analysis (repair taxonomy, teacher vs Qwen behaviors, duplicates, empty_response); empty_response fix (cap 24000, retry temperature, stream guard)

2026-10-06 05:56:23 +02:00

epod-lock-leak.md

Lock leak: controlled reproduction attempts and method (not reproduced), enqueue reader; restart plan for 12 October

2026-10-05 19:47:16 +02:00

epod-syntax-hint.md

K variants for training (free text, EPOD tool names), second attempt only for failed tasks, docs/epod-syntax-hint.md

2026-10-05 12:50:43 +02:00

eval-inceleme.md

Eval review results, easy candidates, rerun queue; handover and STATE updated

2026-10-03 18:57:23 +02:00

faz1-tasarim.md

Correct proxy finding: Z0FFK001 object belongs to the running T01 test

2026-10-03 20:03:24 +02:00

local-qwen-run.md

Series A result: local Qwen 3 of 20 accepted, failures are loops and search loops; dashboard public copy

2026-10-06 05:45:24 +02:00

remote-model.md

serve_remote.sh: pick ~/qwen-venv, refuse an old mlx-lm with a clear message; docs for mlx-lm 0.32.0 on the MacBook

2026-10-05 22:16:56 +02:00

restart-12-oktober.md

Lock leak: controlled reproduction attempts and method (not reproduced), enqueue reader; restart plan for 12 October

2026-10-05 19:47:16 +02:00

stage1-baseline.md

Stage 1 reference baseline: Qwen 3.8 27B mean 15.8 (3/11 above 0), comparison with Devstral

2026-10-04 17:01:52 +02:00

stage1-step-prompts.md

Restore faz1/yol-haritasi docs (overwritten by copy); stage1 docs; G0174 new gap; stage1 25-task subset

2026-10-03 17:33:55 +02:00

stage1-training-task.md

Stage 1 data: strict version dedup (older v* vs next newer v*); counts updated

2026-10-03 22:23:04 +02:00

yol-haritasi.md

Summary: correct the guard line; pipeline summary reads the limit from .env

2026-10-05 18:05:49 +02:00