HackerTrans
TopNewTrendsCommentsPastAskShowJobs

yonom

no profile record

Submissions

Assistant-UI is an open-source AI chat component for React

assistant-ui.com
5 points·by yonom·2 anni fa·0 comments

Pleading with OpenAI Developers to not retire GPT-3.5-turbo-0613 on June 13th

community.openai.com
4 points·by yonom·2 anni fa·1 comments

I made a graph of GPT-3's and GPT-4's price history

twitter.com
13 points·by yonom·2 anni fa·1 comments

Popover API

developer.mozilla.org
385 points·by yonom·2 anni fa·158 comments

Increase Your Luck

blog.simonfarshid.com
4 points·by yonom·2 anni fa·0 comments

Ask HN: Where can I find good legal documents?

319 points·by yonom·2 anni fa·79 comments

Eval("quire".replace(/^/,"re"))(moduleName)

github.com
3 points·by yonom·2 anni fa·2 comments

Save Time with Search-First Navigation

blog.simonfarshid.com
1 points·by yonom·3 anni fa·0 comments

Open Core Business Model

opencoreventures.notion.site
3 points·by yonom·3 anni fa·1 comments

The day my ping took countermeasures

blog.cloudflare.com
1 points·by yonom·3 anni fa·0 comments

Trusted Computing – How does a streaming site protect its contents?

blog.simonfarshid.com
1 points·by yonom·3 anni fa·0 comments

What Is Trusted Computing?

blog.simonfarshid.com
10 points·by yonom·3 anni fa·3 comments

Adapting Illustrations on Dark Mode

blog.simonfarshid.com
11 points·by yonom·3 anni fa·2 comments

Native JSON Output from GPT-4

yonom.substack.com
604 points·by yonom·3 anni fa·248 comments

comments

yonom
·anno scorso·discuss
This is super useful! Have you considered generating docs based on customer support in Slack or Discord conversations?
yonom
·3 anni fa·discuss
@Obscurity4340 Interested in bouncing off ideas for therapy? Email is in my about box :)
yonom
·3 anni fa·discuss
Yes, obviously! I began a blog post on the issues with Trusted Computing but ended up first writing up on the way Trusted Computing works, since most explanations I could find would bury the underlying motivations behind technical jargon.
yonom
·3 anni fa·discuss
I don't think the name "Transformer" is meant in the sense of "transforming between file formats".

My intuition is that LLMs tend to be good at things human brains are good at (e.g. reasoning), and bad at things human brains are bad at (e.g. math, writing pptx binary files from scratch, ...).

Eventually, we might get LLMs that can open PowerPoint and quickly design the whole presentation using a virtual mouse and keyboard but we're not there yet.
yonom
·3 anni fa·discuss
This is cool! Are you using one-shot learning under the hood with a user provided example?
yonom
·3 anni fa·discuss
- if the only reason you're using v4 over v3.5 is to generate JSON, you can now use this API and downgrade for faster and cheaper API calls. - malicious user input may break your json (by asking GPT to include comments around the JSON, as another user suggested); this may or may not be an issue (e. g. if one user can influence other users' experience)
yonom
·3 anni fa·discuss
I would recommend creating a simplified JSON schema for the slides (say, presentation is an array of slides, each slide has a title, body, optional image, optional diagram, each diagram is one of pie, table, ... Then use a library to generate the pptx file from the content generated.
yonom
·3 anni fa·discuss
I believe functions do count in some way toward the token usage; but it seems to be in a more efficient way than pasting raw JSON schemas into the prompt. Nevertheless, the token usage seems to be far lower than previous alternatives, which is awesome!