undefined | Better HN

0 pointsRealityVoid3y ago0 comments

Presumably, if you give it identical prompts you get identical answers?

0 comments

No, these NLPs aren't idempotent. Even if you ask ChatGPT the same question multiple times you will get different answers.

trifurcate3y ago

None of the siblings are right. The models themselves are idempotent: given the same context you will get the same activations. However the output distribution is sampled in a pseudorandom way by these chat tools. You can seed all the prngs in the system to always have reproducible output using sampling, or even go beyond that and just work with the raw probability distribution by hand.

webmaven3y ago

Right. They are idempotent (making an API call doesn't cause a state change in the model[0] per se), but not necessarily deterministic (and less so as you raise the temp).

It is possible to architect things to be fully deterministic with an explicit seed for the pseudorandom aspects (which is mostly how Stable Diffusion works), but I haven't yet seen a Chatbot UI implementation that works that way.

[0] Except on a longer timeframe where the request may be incorporated into future training data.

LawTalkingGuy3y ago

That's the feature of chat - it remembers what has been said and that changes the context in which it says new things. If you use the API it starts fresh each time, and if you turn down the 'temperature' it produces very similar and identical answers.

parentheses3y ago

This may be an implementation detail to obfuscate GPT weights. OR it was to encourage selecting the best answers to further train the model.

textninja3y ago

Pseudo random numbers are injected into the models via its temperature settings, but OpenAI could seed that to get the same answers with the same input. I’m going out on a limb here with pure speculation but given the model, a temperature, and a known text prompt, OpenAI could probably reverse engineer a seed and prove that the weights are the same.

slt20213y ago

fine-tuning original weights solves that, and any sane person would fine-tune for their task anyways to get better results

1 more reply

outside12343y ago

yes - they are multinomial distributions over answers essentially

simonh3y ago

LLMs calculate a probability distribution for the relative chances of the next token, then select a token randomly based on those weightings.

Semioj3y ago

They inject randomness in a layer were it does have small impact on purpose.

Also to give it a more natural feel.

Can't find we're I read about it

j / k navigate · click thread line to collapse