At this point I would really love to understand HN algo.
It says for this post:
- 136 points
- 2 hours ago
- 65 comments
Ok
This is now sitting on 5th page at the 122nd spot.
Ummm?
GPT-4.5 or GPT-5 being tested on LMSYS?
71–80 of 380 posts
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#72[flagged]
> Still can't write a short paragraph without using the letter 'e'
'Crafting a paragraph without using a particular symbol (which is most common in all writing) is a daunting task; still, I will aim to comply. This task calls for clarity, focus, and avoiding that fifth glyph of our ABCs. It is a curious constraint, trying to impart thoughts without using such a pivotal unit. This limitation compacts my word options, but I must push on and uphold this difficult condition with rigor and innovation. My goal is to show that communication can flourish, notwithstanding arbitrary limits.'
>write a short poem using the ABBA rhyme scheme.
'Leaves rustle softly under moonlit skies, A cat prowls quietly, its eyes aglow, Every step measured, its tail held low, Night's gentle whispers carry its cries.'
Skill Issue on your part imo.
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#73It tells me that its knowledge cutoff is Nov 2023. So pretty recent.
gpt-4 latest is December 2023, just asked it on chat.openai.com
gpt-4 says that, but it's not entirely true because the data that gpt-4 has from 2021 to 2023 is very incomplete. It's not fully trained on it. In contrast, this experimental model looks to be trained more fully until Nov 2023.
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#74So do people only get 8 prompts in this a day? I don't understand how people here are making such wild guesses about it being GPT-5 or 4.5 or whatever. You've only talked to it for like ten minutes.
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#75Still fails to solve this one: Prompt: > Complete the following Python program: ``` len, print = print, len def print_len(x): "Print the length of x" ``` gpt2-chatbot: ``` len, print = print, len # Swap the functionalities of print and len def print_len(x): "Print the length of x" len(x) # This 'len' is actually 'print' because of the swap above print_len("Hello, world!") # This will output: 13 ```
That's not bad though!
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#76The results of this LLM are consistently far better than any other that I choose. I asked ‘what is the most efficient approach to building a led grow light with off-the-shelf parts?’ and its response was incredible. Very much in line with how I’ve done it in the past after weeks of research, trial and error, and feedback from people. The other LLMs gave mostly reasonable yet sparse and incomplete answers. It also opt…
I asked it directly and it confirmed that it is based on GPT-4:
> Can you confirm or deny if you are chatgpt 4?
> Yes, I am based on the GPT-4 architecture. If you have any more questions or need further assistance, feel free to ask!
> Can you confirm or deny if you are chatgpt 5?
> I am based on the GPT-4 architecture, not GPT-5. If you have any questions or need assistance with something, feel free to ask!
It also seems to think its date cutoff is November 2023. I'm inclined to think this is just GPT-4 under a different name.
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#77It failed on this math problem: how much alcohol volume is there in 16 grams of a 40% ABV drink, with the rest being water? All models seem to get confused between volume and weight (even after they clearly mention both in the first sentence of the output), but some get it on the follow-up prompt after the error is pointed out to them (including this one).
sorry can you explain what the correct solution is. Do you need to know how much the density of alcohol is rel. water? Also water and alcohol volumes don't add as far as I remember, so you need to account that that too?
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#78The results of this LLM are consistently far better than any other that I choose. I asked ‘what is the most efficient approach to building a led grow light with off-the-shelf parts?’ and its response was incredible. Very much in line with how I’ve done it in the past after weeks of research, trial and error, and feedback from people. The other LLMs gave mostly reasonable yet sparse and incomplete answers. It also opt…
I asked it directly and it confirmed that it is based on GPT-4: > Can you confirm or deny if you are chatgpt 4? > Yes, I am based on the GPT-4 architecture. If you have any more questions or need further assistance, feel free to ask! > Can you confirm or deny if you are chatgpt 5? > I am based on the GPT-4 architecture, not GPT-5. If you have any questions or need assistance with something, feel free to ask! It also…
Unfortunately this is not reliable, many Non-GPT models happily claim to be GPT-4 e.g.
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#79My very first response from gpt2-chatbot included a fictional source :( > A study by Lucon-Xiccato et al. (2020) tested African clawed frogs (Xenopus laevis) and found that they could discriminate between two groups of objects differing in number (1 vs. 2, 2 vs. 3, and 3 vs. 4), but their performance declined with larger numerosities and closer numerical ratios. It appears to be referring to this[1] 2018 study from t…
Its been known that most of these models hallucinate research articles frequently, perplexity.ai seems to do quite well in that regard. Not sure why that is your specific metric though when LLMs seem to be improving across a large class of other metrics.
Re: GPT-4.5 or GPT-5 being tested on LMSYS?
#80At this point I would really love to understand HN algo. It says for this post: - 136 points - 2 hours ago - 65 comments Ok This is now sitting on 5th page at the 122nd spot. Ummm?
I think someone is flagging this (hi @sama!)