Prompting ox-alpha to identify as GLM proves little about the model
Afinetheorem · x · 2026-08-22
Experiments show that using specific system prompts can make 'ox-alpha' identify itself as 'GLM'. However, this is weak evidence of the model's true nature, likely being residual next-token patterns from training data rather than a hardcoded identity in the system prompt.
Related event: Forensics Point to Zhipu's Unreleased GLM Behind Mystery Model Ox Alpha(31 posts)→
More from Models
- Unredacted Anthropic filing exposes internal data-quality debate months before Opus 3 cutoff — zacharynado · 2026-10-15
- Google says teams across the company are using Gemini 4 Argon on its hardest engineering problems — CurieuxExplorer · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- 797GB of n-gram counts in CPU RAM let 1B models match 57% more compute — PangWeiKoh · 2026-10-15
- Leak: Moonshot's Kimi flagship 'x4' done with pre-training, K3.1 in internal testing — lyraxana · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15