# Qwen 3.8 27B on a laptop identifies the games behind raw RL policy weights

> Original title: Qwen 3.8 27B on my laptop can interpret model weights, too

- Source: Simon Willison
- Published: 2026-11-03T16:29:48.000Z
- Bad Influence: https://badinfluence.news/items/ggyhkmq73pbup038fbsgu0vwr
- Original: https://simonwillison.net/2026/Nov/3/qwen-can-interpret-model-weights/

## Summary

Simon Willison says Qwen 3.8 27B running on his laptop can infer which game a neural network was trained to play from raw weights, as described in a recent paper. In LM Studio, the model took 46 minutes and 43,506 reasoning tokens to identify the Pac-Man-style game and reported 72% confidence. It also identified Pong and Breakout in 1 hour 4 minutes and 52 minutes, respectively, but failed on Freeway.