LLMs extract high-level semantic concepts from SVG and ASCII art

3 points by neuronerd1 209 days ago | 1 comment

neuronerd1 209 days ago |

Anthropic interpretability researchers find neural representations of the semantic components of text-based visual depictions.

neuronerd1 209 days ago |

Anthropic interpretability researchers find neural representations of the semantic components of text-based visual depictions.