321–330 of 412 posts
I have a small physics-based problem I pose to LLMs. It's tricky for humans as well, and all LLMs I've tried (GPT o3, Claude 3.7, Gemini 2.5 Pro) fail to answer correctly. If I ask them to explain their answer, they do get it eventually, but none get it right the first time. Qwen3 with max thinking got it even more wrong than the rest, for what it's worth.
Earlier quoted context omitted.
I don't really want it added to the training set, but eh. Here you go: > Assume I have a 3D printer that's currently printing, and I pause the print. What expends more energy, keeping the hotend at some temperature above room temperature and heating it up the rest of the way when I want to use it, or turning it completely off and then heat it all the way when I need it? Is there an amount of time beyond which the ans…
What kind of answer do you expect? It all depends on the hotend shape and material, temperature differences, how fast air moves in the room, humidity of the air, etc.
Earlier quoted context omitted.
Except that the plural of anecdotes is definitely not data, because without controlling for confounding variables and sampling biases, you will get garbage.
Garbage data is still data, and data (garbage or not) is still more valuable than a single anecdote. Insights can only be distilled from data, by first applying those controls you mentioned.
At a managers meeting "user stories" about poor support but all the KPIs looked good from the call center so Jeff dials in the number from the meeting speaker phone, gets put on hold, IVR spin cycle, hold again, etc .... His take away was basically "if the data and anecdotes don't match always default to the customer stories".
Earlier quoted context omitted.
What type of content is removed from US counterparts? Porn, creation of chemical weapons? But not on historical events?
Differ from engine to engine: Googles latest for example put in a few minorities when asking it to create images of nazis. Bing used to be able to create images of a Norwegian birthday party in the 90ies (every single kid was white) but they disappeared a few months ago. Or you can try to ask them about the grooming scandal in UK. I haven't tried but I have an idea. It is not as hilariously bad as I expected, for exa…
Earlier quoted context omitted.
My first try (omitting chain of thought for brevity): When you remove the cup and the mirror, you will see tails. Here's the breakdown: Setup: The coin is inside an upside-down cup on a glass table. The cup blocks direct view of the coin from above and below (assuming the cup's base is opaque). Mirror Observation: A mirror is slid under the glass table, reflecting the underside of the coin (the side touching the tabl…
Manus just answered this correctly for me in one crack: Me: I put a coin in a cup and slam it upside-down on a glass table. I can't see the coin because the cup is over it. I slide a mirror under the table and see heads. What will I see if I take the cup (and the mirror) away? Manus: This is a classic riddle! Let's think it through. You placed a coin in a cup, turned it upside down on a glass table, and then slid a m…
And why did they hide the generalist benchmarks like MMLU-pro & TruthfulQA?
I wish we had proper public benchmarks that are up to date. LMarena was proven useless by the Llama4 scandal, and LiveBench is unrealistic and misses too many models.
Earlier quoted context omitted.
Absolutely. All models ar terrible with Objective-C and Swift, compared to let's say JS/HTML/Python. However, I've realized that Claude Code is extremely useful for generating somewhat simple landing pages for some of my projects. It spits out static html+js which is easy to host, with somewhat good looking design. The code isn't the best and to some extent isn't maintainable by a human at all, but it gets the job do…
Building a basic static html landing page is ridiculously easy though. What js is even needed? If it's just an html file and maybe a stylesheet of course it's easy to host. You can apply 20 lines of css and have a decent looking page. These aren't hard problems.
It used take me an hr or two to get it all done up properly. Now it’s literal seconds. It’s a handy tool.
Earlier quoted context omitted.
I hadn't heard that, but it would certainly explain why the model made a mess of this task. Tried it again like this, using a regular prompt rather than a system prompt (with the https://github.com/simonw/llm-hacker-news plugin for the hn: prefix): llm -f hn:43825900 \ 'Summarize the themes of the opinions expressed here. For each theme, output a markdown header. Include direct "quotations" (with author attribution)…
Seems to truncate the input to only 2048 input tokens
The num_predict setting controls output size.
Earlier quoted context omitted.
> The MoE version with 3b active parameters ~34 tok/s on a Radeon RX 7900 XTX under today's Debian 13.
And vmem use?
ollama 0.6.6 invoked with:
# server
OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 ollama serve
# client
ollama run --verbose qwen3:30b-a3b
/set parameter num_ctx 32768