Glyph research pretrained concepts

Pretrained sparse autoencoders for Glyph research, previously released in RobbieHolland/HypothesisExtractor.

File	Description
`sae_image.ckpt`	Image SAE checkpoint
`sae_findings.ckpt`	Radiology findings SAE checkpoint
`sae_labs.ckpt`	Laboratory SAE checkpoint
`normalization.yaml`	Input normalization statistics
`interpretations.csv`	Concept interpretations

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support