← Lab
ngeodesic
confirmedshipPython · PyTorchModel behavior & reasoningJuly 22, 2026
the question
Is nGPT's normalized-residual update a geodesic (SLERP) step on the sphere, or does it deviate by an exact amount?
what came out
It is a retraction, not a geodesic. With scalar alpha it lags SLERP by a gap that is zero only at alpha in {0, 1/2, 1} -- alpha=1/2 halves the angle exactly (4e-14 deg) -- growing to 49.6 deg at theta=170. With the real per-dimension alpha the update leaves the geodesic plane entirely (off-plane norm ~0.18).
method & receipts
- Result: confirmed
- Tested — a correctness/benchmark suite ships alongside the code.
→ read the code and re-run it
github.com/v-code01/ngeodesic