5 ms·
Where did you think ChatGPT got their training data?
by temporalparts 2y ago
Where did you think ChatGPT got their training data?
- cj 2y agoDoesn’t matter. Generalizing topics and/or adding caveats about how something may not always be true is the easiest way for a LLM to increase the rate of providing factually accurate responses (aside from refusing to answer). Both of those strategies requires being more verbose. Concise and to the point is not in the best interest of a LLM designed to give decent answers 99% of the time.
- mk_stjames 2y agoI also believe that OpenAI's RLHF process highly biased the model towards producing that sort of specific verbose, padded-out 'chatGPT speak' that we are seeing. The RLHF fine tuning process took outputs from the instruction-tuned model and A/B tested variations with human workers who may or may not have had the same feelings towards writing quality as many of us. The process resulted in those verbose, interjection-laden responses that we see now, because that type of response was deemed 'better' (thumbs-up'd more) than the shorter, more-direct-but-less-impressive-sounding responses.
- philipswood 2y agoMy chats usually start with: >Hi again :wave:. >As always I prefer terse replies. >Let's <context of activity> ><First question> I usually then get short answers and then I query for more info if required.
- bigDinosaur 2y agoSet a custom instruction (I think only a ChatGPT paid option?). But $20 a month or whatever is easily worth it for the utility it provides.
- reaperman 2y agoI use it for work so much I’d gladly pay way more than $20 just for access to GPT-4. It’s pretty terrible at programming but it still saves me loads of time generating the easy functions. Anything remotely complex I still do by hand. But holy shit its nice having something do my boilerplate.