cl42

15 ms·

241.
by cl42 3y ago
You state there are benefits to alcohol. I specifically state there aren't based on the biological research I've read. That is directly addressing your point. EDIT: I don't see a 'reply' button beside those replying
242.
by cl42 3y ago
My understanding is that the latest biological research is pretty clear: no amount of alcohol is better than none.
243.
by cl42 3y ago
Related: Andrew Huberman's 2-hour blog post on the negative effects of alcohol[1] convinced me to stop drinking. Once you stop drinking, it becomes so easy to see just how much alcohol pervades popular culture. People drink constantly
244.
by cl42 3y ago
I think there's something to be said about trying to build a Twitter-like, open source protocol.
245.
by cl42 3y ago
I definitely got the inspiration there, though since then have seen quite a few business use cases for this as well... e.g., sales training, coaching, etc.
246.
Roleplaying with LLMs(phasellm.com)
8 points by cl42 3y ago|2 comments
247.
Winterhelm: Notion-like interface for LLM agents [video](youtube.com)
4 points by cl42 3y ago|0 comments
248.
by cl42 3y ago
SEEKING FREELANCER | Django Dev w/ frontend experience I run phasellm.com, an open source project to help manage and evaluate LLMs. We're going to build a hosted solution. I build everything in Python and Django and need someone t
249.
by cl42 3y ago
One of the things I regularly think about is the "Cafe of Broken Dreams". I've found myself in dive bars and coffee shops around the world that make me think, "Oh, wow, this is like the Cafe of Broken Dreams" when i
250.
by cl42 3y ago
I did, but couldn't get into it. First person just doesn't cut it for me.
251.
by cl42 3y ago
YES! So: Arcanum, Disco Elysium, Fallout 1/2... The superfecta.
252.
by cl42 3y ago
LOVE Disco Elysium. Different for sure, but I agree... That was a lot of fun too.
253.
by cl42 3y ago
Fallout 1 and 2 were the best RPG experiences I've ever had. Incredible... I've been chasing that high for decades now.
254.
by cl42 3y ago
Have you tried using GPT-4 via Azure? My understanding is that it's faster and more reliable.
255.
by cl42 3y ago
Those hacks are literally how a large language model using a transformer architecture to predict the next token in a sequence works. They take advantage of how a function choosing a token with maximal probability of appearing works.
256.
by cl42 3y ago
I love these multi-step prompt 'hacks'. They very much take advantage of the fact that this is still 'just' a model predicting the next token. Asking a model to write an email as if it were written by Hemingway requires
257.
by cl42 3y ago
AVs have also struggled with regulatory muddle, which is partly my point. Self-driving is very possible in many situations and if there was a "Manhattan Project" for self-driving to be up and running by 2025 I think we could do it
258.
by cl42 3y ago
The first steam engines were also written off as being less powerful than a horse. The first electric motors were written off as being less powerful than steam engines. So it goes. I think both of these views can be true at the same time: C
259.
How ResearchLLM works: architecture, data flows, and code(phasellm.com)
3 points by cl42 3y ago|0 comments
260.
by cl42 3y ago
I was going to ask the same thing -- maybe someone more astronomically inclined has an answer to this... If one were to take a phot of our solar system, would our asteroid belt really be dense enough to show up in any wavelengths?
261.
by cl42 3y ago
I've been playing with this a bit. Sharing some examples that aren't from their repo, in case people are curious. Simple objects are surprisingly good. Examples of a viking helmet below: -- https://phasellm.com/sta
262.
by cl42 3y ago
Fantastic questions! Re: working/not working at times -- this is still an issue. It's why I'm building PhaseLLM more broadly ( https://github.com/wgryc/phasellm ) -- need a robust pipeline that can also &q
263.
by cl42 3y ago
Oooh, if that’s your concern then give me a few months to launch a product. :-)
264.
by cl42 3y ago
The broader package I’m working on (PhaseLLM) is specifically focused on devtools for observability and robustness of LLM-powered products. I agree with you that there are lots issues with LLMs and taking them to production. I’m hoping prod
265.
by cl42 3y ago
Totally appreciate the feedback and I agree with you that a well structured data set can be trivially analyzed. Heck, at that point you can use drag and drop stats packages too. The data set I used for the demo has strings for income categ
266.
by cl42 3y ago
Nope! I'm not trying to suggest the LLM legitimately understands my query from a conceptual perspective via that question. What I am doing in that prompt is ensuring the LLM can follow the instructions. I specifically ask it to write &
267.
by cl42 3y ago
I'll use myself as an example. I love Pandas + Scikit Learn, but am by no means an expert. Every time I want to build a logistic regression, I have to go back to the docs to review the API. When I was doing my Master's degree in &
268.
by cl42 3y ago
Thanks! Fixing shortly.
269.
by cl42 3y ago
100% agree with you. There are two things driving my work with this demo: (1) A lot of researchers are bad at writing code, but they can audit it. This is true for sociologists, psychologists, etc. so I'm hoping something like this can
270.
by cl42 3y ago
I don't have GPT-4 API access so can't comment on that yet. The code quality via chat has been comparable. One of my goals in the coming weeks (when I get access) is to formally run tests across the models.

More ›