gahow/aituner
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Record qwen235b decode harness launch

Browse Source
This commit is contained in:
Gahow Wang
2026-04-28 07:02:13 +08:00
parent c9089cf4f0
commit f982395aad
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/qwen235b-thinking-decode/harness-20260428.md
View File

@@ -46,6 +46,10 @@ The active run is now seeded from the real run5 baseline and continues from `tri
- Remote spec: `.aituner/harness-qwen235b-decode-20260428-seeded/dash0_qwen235b_decode_thinking_harness_seeded_20260428.json`
- Remote store: `.aituner/harness-qwen235b-decode-20260428-seeded/dash0-qwen235b-decode-thinking-harness-seeded-20260428`
- Seeded `trial-0001`: 0.1267 request/s, 0.0158 request/s/GPU, pass rate 0.9868.
- `proposal-0002`: legal adjacent decode topology move from `TP4/DP2/EP8` to `TP2/DP4/EP8`; no EP-size search and no testcase threshold.
- `trial-0002` status: running on dash0 in `tmux` session `aituner_qwen235b_decode_harness_seeded_20260428`.
The `trial-0002` proposal matches the first useful topology direction from the earlier before-harness run, where the same effective config reached 0.2450 request/s at iter 2. The current run is still executing to verify this under the new harness-controlled study state before claiming final convergence data.
## Follow-up Fix