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

GPT-4

openai.com

Post reply on HN
Print view View original

181–190 of 1001 posts

Re: GPT-4

#181

Post by cs702 »

LLMs will eventually make a lot of simpler machine-learning models obsolete. Imagine feeding a prompt akin to the one below to GPT5, GPT6, etc.:

  prompt = f"The guidelines for recommending products are: {guidelines}.
             The following recommendations led to incremental sales: {sample_successes}.
             The following recommendations had no measurable impact: {sample_failures}.
             Please make product recommendations for these customers: {customer_histories}.
             Write a short note explaining your decision for each recommendation."

  product_recommendations = LLM(prompt)
To me, this kind of use of LLMs looks... inevitable, because it will give nontechnical execs something they have always wanted: the ability to "read and understand" the machine's "reasoning." There's growing evidence that you can get LLMs to write chain-of-thought explanations that are consistent with the instructions in the given text. For example, take a look at the ReAct paper: https://arxiv.org/abs/2210.03629 and some of the LangChain tutorials that use it, e.g.: https://langchain.readthedocs.io/en/latest/modules/agents/ge... and https://langchain.readthedocs.io/en/latest/modules/agents/im... . See also https://news.ycombinator.com/item?id=35110998 .

Re: GPT-4

#182

Post by anileated »

teruakohatu wrote: post #26

Access is invite only for the API, and rate limited for paid GPT+. > gpt-4 has a context length of 8,192 tokens. We are also providing limited access to our 32,768–context (about 50 pages of text) version, gpt-4-32k, which will also be updated automatically over time (current version gpt-4-32k-0314, also supported until June 14). Pricing is $0.06 per 1K prompt tokens and $0.12 per 1k completion tokens. The context le…

Will any of the profits be shared with original authors whose work powers the model?

Re: GPT-4

#183

Post by minimaxir »

attilaberczik wrote: post #109

Prices differences with the last models: ChatGPT API $0.002 per 1k tokens gpt-4 $0.03 per 1k prompt tokens and $0.06 per 1k completion tokens gpt-4 32k context $0.06 per 1k prompt tokens and $0.12 per 1k completion tokens Does completion tokens mean that you also get charged for the answers that the AI gives?

Yes. The `usage` field currently breaks out the token counts for both prompt and completion.

Prompt tokens should have always been cheaper than completion due to how they work.

Re: GPT-4

#184

Post by gield »

Atreiden wrote: post #37

I think it's interesting that they've benchmarked it against an array of standardized tests. Seems like LLMs would be particularly well suited to this kind of test by virtue of it being simple prompt:response, but I have to say...those results are terrifying. Especially when considering the rate of improvement. bottom 10% to top 10% of LSAT in What are the implications for society when general thinking, reading, and…

>bottom 10% to top 10% of LSAT in Their LSAT percentile went from ~40th to ~88th. You might have misread the table, on Uniform Bar Exam, they went from ~90th percentile to ~10th percentile.

>+100 pts on SAT reading, writing, math

GPT went +40 points on SAT reading+writing, and +110 points on SAT math.

Everything is still very impressive of course

Re: GPT-4

#185

Post by sdrinf »

diimdeep wrote: post #164

Is there law in U.S. that made OpenAI implement this in their TOS ? (i) Export Controls. The Services may not be used in or for the benefit of, exported, or re-exported (a) into any U.S. embargoed countries (collectively, the “Embargoed Countries”) or (b) to anyone on the U.S. Treasury Department’s list of Specially Designated Nationals, any other restricted party lists (existing now or in the future) identified by t…

That applies to every corp in the US; I suspect they call out in TOS specifically so that they can hand out bans linking their own TOS directly.

Re: GPT-4

#186

Post by spullara »

diimdeep wrote: post #164

Is there law in U.S. that made OpenAI implement this in their TOS ? (i) Export Controls. The Services may not be used in or for the benefit of, exported, or re-exported (a) into any U.S. embargoed countries (collectively, the “Embargoed Countries”) or (b) to anyone on the U.S. Treasury Department’s list of Specially Designated Nationals, any other restricted party lists (existing now or in the future) identified by t…

Yes, that is why they are called "Embargoed Countries".

https://www.tradecompliance.pitt.edu/embargoed-and-sanctione...

Re: GPT-4

#187

Post by harrisonjackson »

I am interested in how a 32k token context even works. That is so much larger than 4k that I am having a hard time imagining how prompts will change and what sort of output is now possible. That is 50 pages of text. Far larger than most content currently being consumed and generated by LLMs.

Q&A and summarization it will be easy to see improvements as current recursive summarizing and embedding techniques are very "lossy" but outside of improving current use cases what will now be possible??

Re: GPT-4

#188

Post by causi »

Man now I really, really want to feed GPT-4 responses from ChatGPT that don't work and see if it notices and can tell me why.

Re: GPT-4

#189

Post by johnohara »

> I cannot and will not provide information or guidance on creating weapons or engaging in any illegal activities. Please let me know if there is another topic I can help you with.

I understand "will not," but "cannot" seems to imply a highly curated "will not."

The early GPT-4 response indicates the information was part of its dataset. Has the latest version made that information permanently inaccessible or has it been removed entirely?

Is it possible for GPT to keep and hold secrets that are privy to only the most trusted?

Re: GPT-4

#190

Post by smallnix »

cs702 wrote: post #181

LLMs will eventually make a lot of simpler machine-learning models obsolete. Imagine feeding a prompt akin to the one below to GPT5, GPT6, etc.: prompt = f"The guidelines for recommending products are: {guidelines}. The following recommendations led to incremental sales: {sample_successes}. The following recommendations had no measurable impact: {sample_failures}. Please make product recommendations for these custome…

Is my understanding correct that a llm will not put it's "reasoning" in the reply but rather some text which is plausible?
Post reply on HN