The iMac mouse looks amazing though! Ergonomics, not too sure about those.

Not all LLM based applications are a user facing free form chat.

If you take an LLM that makes 10 tool calls in a row for an evaluation, any reduction in unpredictable drift is welcome. Same applies to running your prompt through DSPy Optimizer. [0] Countless other examples. Basically any situation where you are in control of the prompt, the token level input to the LLM, so there's no fuzziness.

In this case, if you would've eliminated token level fuzziness and can yourself guarantee that you're not introducing it from your own end, you can basically map out a much more reliable tree or graph structure of your system's behavior.

[0]: https://dspy.ai/#2-optimizers-tune-the-prompts-and-weights-o...

> If you take an LLM that makes 10 tool calls in a row for an evaluation, any reduction in unpredictable drift is welcome

why use an ambiguous natural language for a specific technical task? i get that its a cool trick but surely they can come up with another input method by now?

Could you provide a link to that system prompt? Becuase I'm confused. I typed in "Are you smart?" and got this back:

> That’s a thoughtful question! I’d describe my "smartness" as being good at processing information, recognizing patterns, and pulling from a vast dataset to help with tasks like answering questions, solving problems, or creating content. However, I’m not "smart" in the human sense—I don’t have consciousness, emotions, or independent critical thinking. I rely entirely on my training data and algorithms.

> Think of me as a tool that can assist with creativity, analysis, or learning, but I lack the depth of human intuition, lived experience, or true understanding. If you’re curious, test me with a question or challenge — I’ll do my best! (smiley emoji)

> Could you provide a link to that system prompt?

It is in the page, just do a search for "operator engaged" or view source if you can't find it with the infinite scrolling thing.

Ah, the site's bugged on Safari and wouldn't scroll. Worked on Chrome. Tried to look for it on the actual chat page, and wasn't in the source there.

Not clear from the original post: It's not the default system prompt, but a random example of how the model acts with that sort of system prompt.

Don't forget the millions of years of pre-training! ;)

You could argue that the suggested system is actually much simpler than the one we currently have for the sites that are "free", aka funded with ads.

The lengths Meta and the like go to in order to maximize clickthroughs...

Most of those look like context issues to me. Repo map (using Tree-sitter, etc) and documentation would already do wonders. Feeding 32-64kTok of context directly into a model like Gemini Pro 2.5 is something that more people should try out in situations like this. Or even 128kTok+.

Are you using a tool other than ChatGPT? If so, check the full prompt that's being sent. It can sometimes kneecap the model.

Tools having slightly unsuitable built in prompts/context sometimes lead to the models saying weird stuff out of the blue, instead of it actually being a 'baked in' behavior of the model itself. Seen this happen for both Gemini 2.5 Pro and o3.

Source?

It's not a model of text, though. It's a model of multiple types of data. Pretty much all modern models are multimodal.

A lot of people make small games just for fun, like any other creative hobby. Similar to: Making music, writing, drawing, 3d modeling, etc.

Actually, you can put all of those together and use them in a game. And the best part is that there's no target market, no KPIs to hit, so you can do anything!