3 ms·

My quick (subjective) impression is that GPT-4.5 is doing better at maintaining philosophical discussions compared to GPT-4.0 or Claude. When using a Socratic a

• hnuser123456 2y ago
  These super large models seem better at keeping track of unsaid nuance. I wonder if that can still be distilled into smaller models or if there is a minimum size for a minimum level of nuance even given infinite training.
• A_D_E_P_T 2y ago
  You can fix the people-pleasing mode thing by simply adding the words "be critical" to your prompt. As for 4.5... I've been playing around with it all day, and as far as I can tell it's objectively worse than o3-mini-high and Deepseek-R1. It's less imaginative, doesn't reason as well, doesn't code as well as o3-mini, doesn't write nearly as well as R1, its book and product recommendations are far more mainstream/normie, and all in all it's totally unimpressive. Frankly, I don't know why OpenAI released it in this form, to people who already have access to o3-mini, o1-Pro, and Deep Research -- all of which are better tools.
  • farts_mckensy 2y ago
    No, GPT 4 cannot consistently hold a position for longer than a few minutes in my experience.
  • vessenes 2y ago
    Hmm. I’m on the other side of this - this feels like what I imagined a scaled up gpt 4 would be: more nuanced and thoughtful. It did the best of any model at my “write an essay as if Hemingway went along with rfk jr when he left the bear in Central Park.” Actual prompt longer. This is a hard task because Hemingway’s prose is extremely multilayered, and his perspective and physical engagement are notable as well. I’d say 4.5 is by far the best at this of released models. It’s probably the only one that thought through both what skepticism and connection Hemingway might have had along for that day and the combination of alienation posing and privilege rfk had. I just retried deepseek on it: the language is good to very good. Theory of mind not as much. Edit: grok 3 is also pretty good. Maybe a bit too wordy still, and maybe a little less insightful.
    • phillipharris 2y ago
      What was your actual prompt? I just asked it for that Hemingway story and the result didn't impress me -- it had none of the social nuance you mentioned.
• farts_mckensy 2y ago
  People underestimate how valuable this is. If you can get an assistant that is capable of being the devil's advocate in any given scenario, it's much easier to game out scenarios.
  • torginus 2y ago
    Unfortunately, the ability to have more nuanced takes on Hegelian dialectic seem like slim pickings to people who have spent tens of billions to train this thing, and need it to justify NVIDIA's P/E ratio of over 100
• maeil 2y ago
  o3-mini is similar in that respect, so that doesn't seem particularly revolutionary nor worth the 100x pricing.
  • cowlby 2y ago
    It's nice that GPT-4.5 doesn't need the thinking time, but yes hard to justify cost.
• elif 2y ago
  Thrilled to see someone else using up all their tokens to dive deeply into ontological meaning. I think LLM are even better for this purpose than coding. After all, what is an LLM but an excessively elaborate symbolic order? My trick for this (works on all models) is to generate a dialogue with 2 distinct philosophical speakers going back and forth with each other, rather than my own ideas being part of the event loop. It's really exposed me to the ideas of philosophers who are less prolific, harder to read, obscure, overshadowed, etc. My prompt has the chosen figures transported via time machine from 1 year prior to their death to the present era, having months to become fully versed in all manner of modern life.
  • al_hag 2y ago
    That seems worth publishing.
    • elif 2y ago
      Yep I am collecting my debates and one day I want to organize them with AI face and voices and release them as YouTube videos.
      • botanical76 2y ago
        I would implore that you just write a blog post or two (or three) instead, but it is your choice of course.
  • farts_mckensy 2y ago
    Symbolic order? Are you a Lacanian? It's rare to see a Lacanian in the wild, especially on hacker news.
    • vessenes 2y ago
      Hello fellow Lacanian. I found 4.5 pretty good at doing some therapy in the style of Lacan in fact. Insightful and generally on point with the tiny amount of Lacan I could claim to understand.