ukuina

26 ms·

511.
by ukuina 3y ago
Unusual, but not a security risk, I hope? Thankfully, only the first name of the recipients is listed.
512.
by ukuina 3y ago
This was extremely difficult to read. The takeaway seems to be: Push enough non-factual arguments in a coherent whole, and your position becomes perfectly defensible?
513.
by ukuina 3y ago
PrivateGPT is one of the better-known examples, but most people are not aware that GPT4 Assistants handle RAG natively now: https://platform.openai.com/docs/assistants/overview
514.
by ukuina 3y ago
Microsoft Guidance will do this.
515.
by ukuina 3y ago
I enjoyed the breadth of this discussion.
516.
by ukuina 3y ago
Imagine what he would have thought of AI-driven programming.
517.
by ukuina 3y ago
Auto-tracking this would be amazing. I insert timestamps very frequently.
518.
by ukuina 3y ago
Unlike crypto, you can try AI for yourself to see the benefits.
519.
by ukuina 3y ago
They also have an official "ChatGPT Classic" that turns off plugins including Bing.
520.
by ukuina 3y ago
If you contextualize your request (e.g., "this is hypothetical"), GPT4 answers quite a few medical questions.
521.
Google Plans Ad Sales Restructuring as Automation Booms(theinformation.com)
2 points by ukuina 3y ago|1 comments
522.
by ukuina 3y ago
And role-playing games have shown us the way to have a curated overarching story while still enabling player agency. Many of those principles will apply to Generative Cinema. The current expectation that movie studios or the Hollywood mach
523.
by ukuina 3y ago
The q2 should fit.
524.
by ukuina 3y ago
And FindAnyFile ( https://apps.tempel.org/FindAnyFile/index.php ), disclosure: happy user with no affiliation.
525.
by ukuina 3y ago
Yes, every processor will have an AI core or two.
526.
by ukuina 3y ago
I built HackYourNews.com (free) to tackle exactly this. Here's how this story showed up there: Dehyped title: Microsoft considers nuclear energy including small modular reactors to power data centers and AI ambitions due to energy dema
527.
by ukuina 3y ago
This is very cool! Thank you for working on it and sharing it with us.
528.
by ukuina 3y ago
https://nitter.net/saurabhambry/status/1734243246140735493
529.
by ukuina 3y ago
The $385k is base salary, not total comp, right?
530.
by ukuina 3y ago
People forget the released version is v0.1 If the commerically-served model has improved capability and is exclusive to Mistral's service, there is a possible moat there.
531.
by ukuina 3y ago
Possibly a huge leap forward in open-source model capability. GPT4's prowess supposedly comes from strong dataset + RLHF + MoE (Mixture of Experts). Mixtral brings MoE to an already-powerful model.
532.
by ukuina 3y ago
LlamaFile typically outperforms LM Studio and even Ollama.
533.
by ukuina 3y ago
A certain minimum level of jank always makes demos more believable. Watching Brockman wade through Discord during the napkin-to-website demo immediately made the whole thing convincing. AI is in the "hold it together with hope and duct
534.
by ukuina 3y ago
> You will never find out the origin of the Traveler. Any hopes of this went out the door alongside Joseph Staten's departure.
535.
by ukuina 3y ago
That'll take it full circle.
536.
by ukuina 3y ago
Guess I was too early submitting this!
537.
by ukuina 3y ago
Exactly. It's inefficient for LLMs to recompute state based on previous output, but that's what we have so far, and it works, surprisingly.
538.
by ukuina 3y ago
But the GPT4 that is accessible via ChatGPT and the API is *not* the base model. It is the instruct-tuned, post-RLHF variant. And while GPT4 does not have a "world model", can we please move on from the stochastic parrot argument?
539.
by ukuina 3y ago
Looks like their CoLab works (and they've provided open access to a cloud-hosted NexusRaven-V2), but the links to HuggingFace and GitHub are not yet live. https://colab.research.google.com/drive/19JYixRPPlanmW5q49W
540.
NexusRaven-V2: Surpassing GPT-4 for Zero-Shot Function Calling(nexusflow.ai)
2 points by ukuina 3y ago|3 comments

More ›