# Lewis Tunstall says visible model reasoning was often a guess

> Original title: Where the reasoning was visible, it was often a guess. Clearly not worth training on either ...

- Source: Lewis Tunstall
- Published: 2026-11-02T16:44:14.000Z
- Bad Influence: https://badinfluence.news/items/obhd4hhamhynearl7z2rgncl5
- Original: https://x.com/_lewtun/status/2117296740059508908

## Summary

Lewis Tunstall of Hugging Face says that where a model's reasoning was visible, it was often a guess. He adds that such reasoning is clearly not worth training on.

## Post

Where the reasoning was visible, it was often a guess. Clearly not worth training on either ...

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWIihI2WsAEQbu5.png&mode=thumb&exp=1793836800&sig=b3d99d6d61cfc2b9)