ricopags

7 ms·

31.
by ricopags 3y ago
We're onboarding to Cloudflare MagicWan and want to use them for logging, which they do to 's3-compliant' buckets.... on Google or Amazon. I was pretty surprised at the lack of dogfooding, wondered if it's an oversight,
32.
by ricopags 3y ago
Why do you use quotes? My reason would be NVDA sticker price.
33.
by ricopags 3y ago
Are you not going to say more? The site is pretty generic and it might help your case to elucidate clearly what you're after in a partner.
34.
by ricopags 3y ago
This is such exciting news! Huge thanks to you for your continued work in making sense of AI. I wonder if the recent Bitnet 1.58 paper [the use of ternary bits in lieu of fp/int] might be an advancement that could further reduce the co
35.
by ricopags 3y ago
The idea of applying "AI" as it stands to emergency services ought to freeze you with fear. If it was going to be implemented, it would probably be by a company that can spring for a non-builder-branded (and functional) website.
36.
by ricopags 3y ago
The important and overlooked distinction is that the choice of underlying model in the product ChatGPT is not the same as calling gpt4 via the api. Sending a prompt into one vs the other, the API sends through the model and back out, the pr
37.
by ricopags 3y ago
The model's weights were released in a torrent, but this is the much anticipated paper detailing some of the work.
38.
by ricopags 3y ago
I am no brainiac, and it isn't super clear from your post what you're describing, but here is some info that might help you better convey your question: This is a neural net built by conjoining Phi-2 [the best available small LLM,
39.
by ricopags 3y ago
Archive link [0] I'm confused about the timeline of jumping from Adept to Essential, and ignorant as to the implications. Adept still hasn't released a product and its $350m Series B was only in March, shortly after Vaswani and Pa
40.
by ricopags 3y ago
Just to +1 mmoskal, while Guidance does somewhat work with OpenAI APIs, AFAIK it was first designed around having direct access to the logits, thus is superior when using local models.
41.
by ricopags 3y ago
Ha, yay! So now: Do you think it might be useful/possible to incorporate as a plugin [or otherwise] into your LLM app?
42.
by ricopags 3y ago
My friend runs a bookkeeping service for small to mid-size businesses[0] and they usually recommend going through the state website. Oregon's site[1] looks pretty straightforward. If you want stuff wrapped around the LLC beyond its sca
43.
by ricopags 3y ago
Congrats on the launch! JFYI, app. and docs.twenty.com are both showing cloudflare origin errors at present. The signup from the main site asked me to message on WhatsApp, which I don't use, so I backed out from signing up.
44.
by ricopags 3y ago
Perhaps something as simple as stating it was first built around OpenAI models and later expanded to local via plugins? I've been meaning to ask you, have you seen/used MS Guidance[0] 'language' at all? I don't know
45.
by ricopags 3y ago
Cellebrite[0] is the company in question. [0] https://signal.org/blog/cellebrite-vulnerabilities/
46.
by ricopags 3y ago
It's a short, quick read. It's honestly shocking it could be written in 1909. True imagination. I think a nice balancing title [equally short] would be Marc Steigler's Gentle Seduction[0] to which I was first introduced in a
47.
by ricopags 3y ago
E.M. Forster's The Machine Stops[0] is probably the most prescient and early tale I've seen foretelling the outcome of Verilio's Pure War and Postman's Amusing Ourselves to Death. It's worth a read. [0] https:/
48.
by ricopags 3y ago
I achieved what proficiency I now possess by chatting with people on IRC, newsgroups, forums, and IM platforms in the 90s[0]. I took a typing class in high school on clicketyclack typewriters and was buzzing around 80-90wpm much to the chag
49.
by ricopags 3y ago
Not a class, but a whole school? Wow. For how long did you attend?
50.
by ricopags 3y ago
Congratulations on the launch preview! This is exciting, as I've been waiting for this announcement since Jim Keller's callout on Lex Fridman podcast two years ago[0]. I'm curious about the teasing around open source. Obvious
51.
by ricopags 3y ago
I'm not sure about pre-tokenized but there are dataset marketplaces[0][1][2] and for training datasets there are many open resources. I don't think the utility of any individuals scant tokens will be given much value to the way th
52.
by ricopags 3y ago
Pretty sure this is the approach bing is taking right now. It'll start to output 'naughty' responses and then they'll be deleted and it'll tell you off.
53.
by ricopags 4y ago
Curious, will your software see a release any time soon? Just to surface it, your website's "news" section is quite outdated.
54.
by ricopags 4y ago
I think they went the GPT-4 route and ^z their release so they could productize[0]. [0] https://www.nvidia.com/en-us/gpu-cloud/picasso/
55.
by ricopags 4y ago
Came here to point this out, though not as pithily :D Really, really bad mark on whoever is in charge of their web marketing. Images should never look that bad, not even in support, but definitely not in marketing. edit: so this post is mor
56.
by ricopags 4y ago
This is known. The HF in RLHF upon which InstructGPT is fine-tuned is human feedback. With so many people offering their take on this new technology, perhaps it's important to at least read a little before writing about it.
57.
by ricopags 4y ago
In a literal sense in answer to your question: nothing. Language models have been used by Meta and others to enhance the protein fold work of Alphafold2. Conjoined with diffusion networks, they're able to generate de novo proteins[0].
58.
by ricopags 4y ago
did I or do I?
59.
by ricopags 4y ago
any plans to share the code? I've been thinking of rebooting a crappy club for jerks and this looks like it could be a great jumping off point if you'd be willing to share
60.
by ricopags 4y ago
might want to delete the comment, or at least the edit. you should really know better

More ›