←back to thread

SeedLM: Compressing LLM Weights into Seeds of Pseudo-Random Generators

(machinelearning.apple.com)
171 points pizza | 1 comments | 06 Apr 25 08:53 UTC | HN request time: 0.205s | source
Show context
RainyDayTmrw ◴[06 Apr 25 17:48 UTC] No.43603323[source]▶
How do you reconcile this with the (I believe) widely accepted idea that you can't meaningfully compress data using offsets into Pi?
replies(2): >>43604557 #>>43605578 #
1. diegoperini ◴[06 Apr 25 22:39 UTC] No.43605578[source]▶
You get to choose your own, more efficient "PI" for your model. Still, it's a valid question.