(github.com)

489 points todsacerdoti | 2 comments | 25 Jun 25 23:26 UTC | HN request time: 0.418s | source

Show context

BurningFrog ◴[26 Jun 25 00:51 UTC] No.44383219[source]▶

Would it make sense to include the complete prompt that generated the code with the code?

replies(3): >>44383344 #>>44384164 #>>44386462 #

1. catlifeonmars ◴[26 Jun 25 04:24 UTC] No.44384164[source]▶

You’d need to hash the model weights and save the seeds for the temperature prng as well, in order to verify the provenance. Ideally it would be reproducible, right?

replies(1): >>44385459 #

2. danielbln ◴[26 Jun 25 08:45 UTC] No.44385459[source]▶

>>44384164 (TP) #

Maybe 2 years ago. Nowadays LLMs call functions and use tools, good luck capturing that in a way that it's reproducible.

↑

Define policy forbidding use of AI code generators