Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style LLM on a MacBook

Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style... by hervature on Thursday March 16, 2023 @03:43PM (#35187290)

Let's take a step back. You have a model like linear regression. For example, y=bx where y are your outputs and x are your inputs. Based on some data, you learn that b=1. Therefore, you share the weights of the model as a file like {b=1} and also share with them the model y=bx (usually shared via code) so they can run it in production.

[ Parent | on HN ]

Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style... by MacsHeadroom on Friday March 17, 2023 @06:02PM (#35203495)

This is the best explanation imo.

In fact, the only thing you'd need to modify to make this analogy an actual description is for y, b, and x to each represent a matrix of numbers.

[ Parent | on HN ]