The geometry of hidden representations of large transformer models
–Neural Information Processing Systems
By analyzing the intrinsic dimension (ID) and neighbor composition, we find that the representations evolve similarly in transformers trained on protein language tasks and image reconstruction tasks.
Neural Information Processing Systems
Feb-16-2026, 05:43:08 GMT
- Country:
- North America > United States
- Minnesota > Hennepin County > Minneapolis (0.14)
- Europe > Italy
- Tuscany > Florence (0.04)
- Friuli Venezia Giulia > Trieste Province
- Trieste (0.05)
- Asia > Taiwan
- Taiwan Province > Taipei (0.04)
- North America > United States
- Genre:
- Research Report > New Finding (0.68)
- Industry:
- Technology: