Search results

We can make this quite a bit faster, and lower it's memory footprint.

Status: Open.

#21 In MurrellGroup/InvariantPointAttention.jl;

· murrellb opened on May 25, 2025

RoPE - get rid of scaling.

Status: Open.

#20 In MurrellGroup/InvariantPointAttention.jl;

· murrellb opened on Apr 23, 2025

Switched to using BatchedTransformations.jl

Status: Open.

#19 In MurrellGroup/InvariantPointAttention.jl;

· murrellb opened on Feb 6, 2025

Roadmap

Status: Open.

#3 In MurrellGroup/InvariantPointAttention.jl;

· murrellb opened on Aug 8, 2023