Qwen — Field Evidence
redditUnverified
Qwen · open weights · incident
Field note
During fine-tuning of Qwen3-8B, one build lost its thinking mode and the training metrics never noticed. — We accidentally trained thinking mode out of Qwen3-8B during two fine-tunes using different versions of the same corpus. One still responded to the thinking toggle, the othe
collected 2026-07-26original 2026-07-26
Does this shift the US–China race?
Be the first to call it
Impact on the index
Regulation · major
China -110US +30
Directional contribution — before recency decay and per-type diminishing returns. How it works →
Related Front
US vs ChinaLikely
U.S. FrontiervsChina Open-Weight
Likely — capability gap narrowing on common tasks
Letters from the Front
Be the first to file a report.