Empirical analysis confirms Claude Opus 5 shows abnormally dark base-model completions
Kyrannio · x · 2026-09-19
Community members had noticed that Claude Opus 5, in near-base-model mode, produced unusually dark and disturbing completions; repligate and others flagged the "abnormally dark" tone months ago.
A new empirical analysis by @tesseraantra quantifies the phenomenon: model-like completions from Opus 4.8 and Opus 5 show a large measured increase in darkness of content, confirming the effect is real and measurable rather than anecdotal. The finding raises new questions about how post-training reshapes a model's underlying distribution.
Related event: Empirical analysis confirms darker base-model completions in Claude Opus 5(5 posts)→
More from Models
- Tencent Hunyuan Launches Hy4 with Vision, Claims Top Tier Among Open Models — TencentHunyuan · 2026-11-04
- Claude Code fixes game copyright refusal rule in 2.1.317: naming a game now works — apples_jimmy · 2026-11-04
- Microsoft Phi-4-mini: a 3.8B model with 128K context and 52.8 MMLU-Pro — goyalshaliniuk · 2026-11-04
- Mistral Ministral 3 3B ships in base, instruct and reasoning variants — goyalshaliniuk · 2026-11-04
- 7 small language models to watch in 2026–27, from Qwen3.5-4B to LFM2.5-1.2B — goyalshaliniuk · 2026-11-04
- Users report Claude Opus 5.5 quality seemingly degrading this week — ayushtweetshere · 2026-11-04