4 ms·

I also think they might never become reliable.

by kubb 1y ago

diggan 1y ago
But what does that mean? If you tell the LLM "Say just 'hi' without any extra words or explanations", do you not get "hi" back from it?
- TeMPOraL 1y ago
  That's literally the wrong way to use LLMs though. LLMs think in tokens, the less they emit the dumber they are, so asking them to be concise, or to give the answer before explanation, is extremely counterproductive.
  - diggan 1y ago
    I was trying to make a point regarding "reliability", not a point about how to prompt or how to use them for work.
    - TeMPOraL 1y ago
      This is relevant. Your example may be simple enough, but for anything more complex, letting the model have its space to think/compute is critical to reliability - if you starve it for compute, you'll get more errors/hallucinations.
      - diggan 1y ago
        Yeah I mean I agree with you, but I'm still not sure how it's relevant. I'd also urge people to have unit tests they treat as production code, and proper system prompts, and X and Y, but it's really beyond the original point of "LLMs aren't reliable" which is the context in this sub-tree.
- kubb 1y ago
  Sometimes I get "Hi!", sometimes "Hey!".
  - diggan 1y ago
    Which model? Just tried a bunch of ChatGPT, OpenAI's API, Claude, Anthropic's API and DeepSeek's API with both chat and reasonee, every single one replied with a single "hi".
    - throwdbaaway 1y ago
      o3-mini-2025-01-31 with high reasoning effort replied with "Hi" after 448 reasoning tokens. gpt-4.5-preview-2025-02-27 replied with "Hi!"
      - diggan 1y ago
        > o3-mini-2025-01-31 with high reasoning effort replied with "Hi" after 448 reasoning tokens. I got "hi", as expected. What is the full system prompt + user message you're using? https://i.imgur.com/Y923KXB.png https://i.imgur.com/Y923KXB.png > gpt-4.5-preview-2025-02-27 Same "hi": https://i.imgur.com/VxiIrIy.png https://i.imgur.com/VxiIrIy.png
        throwdbaaway 1y ago
        Ah right, my bad. Somehow I thought the prompt was only: Say just 'hi' while the "without any extra words or explanations" part was for the readers of your comment. Perhaps kubb also made a similar mistake. I used empty system prompt.
flir 1y ago
There is a bar below which they are reliable. "Write a Python script that adds three numbers together". Is that bar going up? I think it probably is, although not as fast/far as some believe. I also think that "unreliable" can still be "useful".

I also think they might never become reliable.

by kubb 1y ago

I also think they might never become reliable.

But what does that mean? If you tell the LLM "Say just 'hi' without any extra words or explanations", do you not get "hi" back from it?

TeMPOraL 1y ago
That's literally the wrong way to use LLMs though. LLMs think in tokens, the less they emit the dumber they are, so asking them to be concise, or to give the answer before explanation, is extremely counterproductive.
- diggan 1y ago
  I was trying to make a point regarding "reliability", not a point about how to prompt or how to use them for work.
  - TeMPOraL 1y ago
    This is relevant. Your example may be simple enough, but for anything more complex, letting the model have its space to think/compute is critical to reliability - if you starve it for compute, you'll get more errors/hallucinations.
    - diggan 1y ago
      Yeah I mean I agree with you, but I'm still not sure how it's relevant. I'd also urge people to have unit tests they treat as production code, and proper system prompts, and X and Y, but it's really beyond the original point of "LLMs aren't reliable" which is the context in this sub-tree.
kubb 1y ago
Sometimes I get "Hi!", sometimes "Hey!".
- diggan 1y ago
  Which model? Just tried a bunch of ChatGPT, OpenAI's API, Claude, Anthropic's API and DeepSeek's API with both chat and reasonee, every single one replied with a single "hi".
  - throwdbaaway 1y ago
    o3-mini-2025-01-31 with high reasoning effort replied with "Hi" after 448 reasoning tokens. gpt-4.5-preview-2025-02-27 replied with "Hi!"
    - diggan 1y ago
      > o3-mini-2025-01-31 with high reasoning effort replied with "Hi" after 448 reasoning tokens. I got "hi", as expected. What is the full system prompt + user message you're using? https://i.imgur.com/Y923KXB.png https://i.imgur.com/Y923KXB.png > gpt-4.5-preview-2025-02-27 Same "hi": https://i.imgur.com/VxiIrIy.png https://i.imgur.com/VxiIrIy.png
      - throwdbaaway 1y ago
        Ah right, my bad. Somehow I thought the prompt was only: Say just 'hi' while the "without any extra words or explanations" part was for the readers of your comment. Perhaps kubb also made a similar mistake. I used empty system prompt.

flir 1y ago

There is a bar below which they are reliable. "Write a Python script that adds three numbers together". Is that bar going up? I think it probably is, although not as fast/far as some believe. I also think that "unreliable" can still be "useful".