3 ms·
I don't think that is exactly true... I've seen arguments based on floating point calculations that accruing errors in FP can generate different results.
by pixl97 3y ago
I don't think that is exactly true... I've seen arguments based on floating point calculations that accruing errors in FP can generate different results.
- nomel 3y agoI've seen suggestions that it's an implementation problem, in the pipeline, that the same computation, with the same values, and same FP errors (they're deterministic) doesn't result in the same output with a temperature of 0, along with theories that the problem is not chased because it's probably related to optimization/cost reduction. (trouble finding sources, but it being a bug was from someone involved with OpenAI)