Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatGPT loses users for first time, shaking faith in AI revolution

washingtonpost.com

Post reply on HN
Print view View original

231–240 of 305 posts

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#231

Post by kazinator »

ksaj wrote: post #226
ragequitta wrote: post #214

Earlier quoted context omitted.

>It doesn't parse, it doesn't explain, it does not grok. It guesses at best and the blood sucking robot-horse is not telling the truth. In my experience with coding (I've only done javascript and python myself) you have to tell it to explain and grok. It takes on the role you give it. Even just saying something like "you are a professional unreal developer specializing in C++, I am your apprentice writing code to (x)…

I do think this is why I sometimes get amazing results, and other times I have to go over a snippet of code so often I just give up and do it myself. It's a matter of how the question was asked in the first place. Knowing that, it makes sense that your prompt should be as specific as possible if you want the results to be as specific as possible. The best results I got was feeding it Lisp code that I wanted translate…

GPT4 works best when you assume that you're the professional dev with decades of experience, whereas GPT4 is a bright and broadly-informed co-op student lacking in experience in getting stuff working. You have to have a solution in mind, and coach it with specifics. And recognize the tipping point where it takes you more keystrokes of English to say what should be done, than keystrokes in Vim to do it yourself.

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#232

Post by gremlinsinc »

ygouzerh wrote: post #33

If ChatGPT will provided more free-tier for GPT4, it's pretty sure that the numbers will go up. At least from a software engineer point of view, the code provided by GPT3.5 is no match against GPT4

just use phone.com it's gpt4 with search and some magic sauce that sometimes runs 3 or 4 retrospections to get a better result.

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#234

Post by barrysteve »

ragequitta wrote: post #214
barrysteve wrote: post #193

Earlier quoted context omitted.

Oh come on. I fed Unreal c++ engine code to ChatGPT4 and it couldn't understand inheritence in Slate classes and therefore kept offering me the same broken solution for a parameter with the wrong type. The Unreal engine code is documented and publicly avaiable for OpenAI to ingest and it still gets the basics wrong. I wasted hours trying to get it to explain to me what I didn't know, if it doesn't understand the inte…

>It doesn't parse, it doesn't explain, it does not grok. It guesses at best and the blood sucking robot-horse is not telling the truth. In my experience with coding (I've only done javascript and python myself) you have to tell it to explain and grok. It takes on the role you give it. Even just saying something like "you are a professional unreal developer specializing in C++, I am your apprentice writing code to (x)…

I did prompt engineer, using the 'you are an expert, desribe to student with examples' in many different variations.

In my testing prompts did not unlock an ability in GPT to grok the structure of code.

Empirical testing of LLM's is going to prove and map out it's weaknesses.

It is wise to infer from intution and examples what it can handle, leave the empirical map of it's capabilities to the academics, for the provable conclusions.

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#235

Post by vorticalbox »

sebastiennight wrote: post #201
vorticalbox wrote: post #88

Earlier quoted context omitted.

Sometimes it's very good I had a list of strings in this format "foo:bar:baz" and needed to convert this into an dict {"foo":{"bar":"baz"} It wrote me the function and I got on with the rest of my work. Other times it's just completely wrong or doesn't get what I'm after is might be a "me" problem with my prompting

It's 11PM and the unclosed curly bracket here is going to prevent me from sleeping. Did you mean ``` {"foo":{"bar":"baz"}} ``` ? By the way, have you had success at asking GPT to generate unit tests that include edge cases? I've been wondering about that.

I did mean that I apologize for the missed }.

I haven't had much luck slot of the time it just adds a comment saying "rest of tests here" which is annoying.

I have had good success with it generating things like JSDOC/doc strings and sending git diff and asking it to create a "developer friendly summary for use in a pull request description"

This has definitely increased my work output

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#236

Post by aftbit »

People are so short sighted. If we predict that inflation will be 6.8%, then it comes in at 6.7%, the market goes straight up, even though the numbers are basically the same. If a new exciting company is growing like crazy, but then has a few months of negative growth, we assume it's dead? Chill out people - real life takes time to resolve. You can't live on the month-to-month business cycle.

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#237

Post by callalex »

no_butterscotch wrote: post #45

I was using ChatGPT this week for help with a back injury I suddenly sustained. Because I don't have insurance I was looking for options as to what I should do and what it'd cost and where to go etc. I did get a suggestion that I followed up on for an in person visit. But my big beef with it was that it kept adding disclaimers: "I am not a doctor", "My current knowledge is up to September 2021, insurance and hospital…

Apparently the message aren’t frequent enough, as you clearly did not heed their message.

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#238

Post by horsawlarway »

frognumber wrote: post #175
horsawlarway wrote: post #150

Earlier quoted context omitted.

2 x 3090 (renewed) ~1800 128gb ram ~400 reasonable processor/mobo/psu ~600 2Tb m2 drive ~94 In hindsight - I don't know that the second GPU was worth the spend. The c++ tooling is doing a very good job right now at spreading work between GPU vram and main ram and still being fast enough. Even ~4/5 tokens a second is fast enough to not feel like you're waiting. I'd suggest skipping the second card and dropping the pri…

Are you using the second GPU at all? My experience is only a few systems will share load across GPUs. I didn't bother with dual GPUs for that reason. 4-5 tokens per second is slower than my system. I'm getting in the teens. I'm a little surprised since yours is newer, faster, and has way more RAM.

Yes. I am definitely using both GPUs, I can run the quant 4 65b models entirely in VRAM (they use about 40GB).

If I push everything into VRAM - I get 12.2 tokens on average running quant 4 llama 65b.

If I run a smaller model I get considerably faster generation. Ex: llama 7b runs at 52 tokens/sec, but it's small enough I don't need the second GPU.

Ex - here's my nvidia-smi output while 65b is running

https://imgur.com/a/JnaieKg

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#239

Post by Art9681 »

onlypositive wrote: post #218
frognumber wrote: post #168

Earlier quoted context omitted.

My experience with open source models places them as a little bit worse than GPT 3, and nowhere close to GPT3.5. That said: - For many uses, it doesn't matter. For many of the ways I use it, I don't care. For basic use (e.g. clean up an email for me), it's basically the same. For things like complex reasoning, algorithms, or foreign languages, the hosted service is critical. - GPT3-grade models have more soul. OpenAI…

> OpenAI trained GPT3.5 and 4 to never do anything offensive, and that has a lot of negative side effects, well-documented in research. I'm absolutely disgusted by OpenAI for this "do no offense" approach. How can people so smart be so damn uneducated? Then again, this industry has disgusted me for a long time so it's not really a surprise.

Would you write a check to fund a business that could potentially self-destruct via lawsuits alone? In the end, the best model will not be owned by a mega corp like MicrOpenAI. It may be the most popular, but it will be the equivalent of the sanitized version of history students learn in school. The best model will have no problem telling you, very factually, that the hallways of Versailles used to smell like sh--.

Re: ChatGPT loses users for first time, shaking faith in AI revolution

#240

Post by treeman79 »

konha wrote: post #108
treeman79 wrote: post #97

Earlier quoted context omitted.

Mentioned ChatGPT to my teenager a few months ago. Huge percentage of the teenagers at her school are using it to do their assignments. Later I her to read a few chapters of her book for summer reading. When I quizzed her it, she got it right. Found out later she used ChatGPT to give her a summary of the first chapters. She did not enjoy having first weeks of summer without a phone. School is having to change all ass…

That sounds incredibly backwards to me. How about accepting new realities and changing homework assignments instead?

It’s also acceptable for a kid to lean to read a book. She’s behind on reading and comprehension. Getting the summary isn’t helping to solve that.
Post reply on HN