Hacker News new | ask | show | jobs
by fshbbdssbbgdd 845 days ago
If you make a language model which just predicts the next word without trying to shape the output to be “good”, what you get is GPT instead of ChatGPT. You try to ask it a question and it will keep asking more questions (with a similar writing style). Or curse at you, or change the subject. RLHF was the breakthrough that fixed this. We used human feedback to train the model to give “good” output.

Ok, so probably we agree that the product needs to try to be “good”. Cue a million opinions about what “good” is. Whatever comes out is the result of a value judgment, there is no getting around it. Same issue comes up with images from Gemini in this case, or any other generative AI product. You don’t actually want the AI to be unbiased because that output would be hot garbage.
1 comments
anothernewdude 845 days ago
GPT was good. ChatGPT ruined it. It is better and more powerful to write prompts such that the continuation is what you want, than ask questions as if it understood.

Now we don't even get log probs, one of the most powerful features to evaluate the output of the models.

link
fshbbdssbbgdd 845 days ago
Raw GPT is a good tool but there’s a reason it didn’t get 100 million users.

If you made unbiased Gemini and asked it for images of German soldiers it probably wouldn’t output an image at all, and if it did it would probably be porn.

link
anothernewdude 844 days ago
> there’s a reason it didn’t get 100 million users.

Another feature of the non-chat model.

> asked it for

I feel like I just said how "asking for" things is a bad way of prompting AI.

link