Paper notes: measuring forward pass "shape" reveals how transformer representations converge
its_vayishu · x · 2026-10-10
Retweet of paper-reading notes on how transformer representations "converge" across layers:
- Treat the forward pass as a path and measure its "shape"
- Category structure is already present in layer-0 embeddings
- GPT-2, TinyLlama and Qwen behave differently at depth even on the same prompts
A geometric lens on how representations evolve with depth across models.
Related event: New Paper Models Transformer Layer Representations as Trajectory Geometry(2 posts)→
More from Research
- RFD4-Proteina's 'underrated' docs detail partial diffusion for protein redesign — AllThingsApx · 2026-10-23
- MPI-Sec PhD Thesis on Vulnerability Detection Benchmarks Includes Pre-LLM Classic "Top Score on the Wrong Exam" — mboehme_ · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- New paper makes 9 falsifiable predictions toward a major AI-enabled terror attack by 2027-2028 — StephenLCasper · 2026-10-23
- HumanoidPose recovers humanoid robots' full-body joint states from RGB video alone — cholz · 2026-10-23
- Sophont's Holon-S1 predicts genes and biomarkers from routine pathology slides — iScienceLuvr · 2026-10-23