yummytummy

6 ms·

1.
by yummytummy 5mo ago
That might be, but the argument was that poor cache utilization was costing Anthropic too much money in other harnesses. If cache is considered in rate limits, it doesn’t matter from a cost perspective, you’ll just hit your rate limits fast
2.
by yummytummy 5mo ago
Ah, so cache usage impacts rate limits. There goes the ”other harnesses aren’t utilizing the cache as efficiently” argument.
3.
by yummytummy 3y ago
Since I posted this, you've generated over 2000 questions, thank you!
4.
by yummytummy 3y ago
Sorry, what did you try to do and what happened?
5.
by yummytummy 3y ago
Yes I currently only generate 3 questions per article. I tend to get worse results if I generate more questions per article. I instead try to encourage exploration by showing related articles (clickable) to try to get a wider ranger of topi
6.
by yummytummy 3y ago
It would actually be interesting to know what the energy required by a human is to research a topic, think of plausible options, find references and type it out. I'm not sure how much energy Trivai requires, or if it's even remote
7.
by yummytummy 3y ago
Thank you for reporting it! I've actually seen this particular question before. It's hard to figure out why some questions get bad or wrong answers. This question also doesn't have a proper reference. I think it's diffic
8.
by yummytummy 3y ago
Thank you for trying it out! I'm matching the search queries with a Wikipedia page right now, and in this case it found "20th century art". If you press the explanation "Die Brücke was a group of German artists who strov
9.
by yummytummy 3y ago
Check out the pregenerated questions if you just want to see what the questions look like. And press the text that shows up after a correct answer to see what references the question is based on!
10.
by yummytummy 3y ago
Thank you!
11.
Show HN: Trivai.app – AI powered trivia questions, with references(trivai.app)
20 points by yummytummy 3y ago|15 comments
12.
by yummytummy 4y ago
Me too! I’ve been experimenting with the prompt to try to make it provide better references and questions. Logged in users can upvote and downvote questions so the next step will be to try to fine tune the model on good questions.
13.
by yummytummy 4y ago
Very cool! I've experimented with a similar thing for trivia questions based on Wikipedia articles. I'm also trying to reference the part of the text where the question was generated from, so the user can verify if it was correct
14.
by yummytummy 4y ago
I think you have some good points here. I agree that just releasing this unrestricted from the start is probably a bad idea. Being cautious is most likely a good thing. > More generally I think there has to be a match between a social un
15.
by yummytummy 4y ago
Thank you! There's a lot of talk about "safety" here that I think back in 2018 I interpreted as guard rails towards some sort of rogue AGI. "and to avoid enabling uses of AI or AGI that harm humanity or unduly concentrat
16.
by yummytummy 4y ago
It is highly subjective, and I think that helps me make my point. The subjective view point we get now is that of OpenAIs. I will give you two examples of content of various degrees of seriousness where the harmfulness is determined by how
17.
by yummytummy 4y ago
Thank you, I will take a look at the links! Licensing might be an issue, but as long as the models and checkpoints are available there's at least competition. I also seem to remember that this was the original intent of OpenAI, but me
18.
Ask HN: Who is working on an unrestricted LLM?
2 points by yummytummy 4y ago|10 comments
19.
by yummytummy 6y ago
Google is an ad company. They will use this to measure how you respond to ads. It’s disgusting.