ayewo

8 ms·

1.
by ayewo 6d ago
To add to this, merely using the thumbs up/down button in a chat could share your entire conversation with them for model training. From their docs[1] (archive copy is at [2]): > You can opt out of training through our privacy porta
2.
by ayewo 9d ago
> Microsoft famously had teams at odds with each other, ... Instantly reminded me of this especially apt comic: https://pbs.twimg.com/media/EyJfRwHWYAI0_yt?format=jpg [1] Also available here: https://ww
3.
by ayewo 12d ago
>> Are we forgetting how fast they rushed in to deploy Claude at the DoW (Department of War) with Palantir? > I hardly see how the Dow Jones in relevant here, that’s finance Not Dow Jones. DoW = Department of War.
4.
by ayewo 15d ago
Not so sure since Anthropic has 4 model families while OpenAI has 3 for GPT-5.6. Claude Fable/Mythos vs GPT-5.6 Sol Claude Opus vs GPT-5.6 Terra Claude Sonnet vs GPT-5.6 Luna Claude Haiku vs ?
5.
by ayewo 15d ago
Spot on wrt CoT. I have thinkingSummaries enabled and I find it eminently readable compared to the prose in Claude's replies. In fact, whenever Claude disobeys me, I usually first skim the CoT to figure out if my original instruction w
6.
by ayewo 17d ago
True. But it was also meant as a counter to “Sent from my Blackberry”. Obama and a lot of execs were pretty addicted to their BB back in those days.
7.
by ayewo 21d ago
Perhaps these may be Huawei Ascend chips. https://en.wikipedia.org/wiki/HiSilicon#Ascend_910 https://medium.com/@huaweiclouddevelper/a-brief-introduction...
8.
by ayewo 23d ago
They want to be able to zoom in and zoom out as needed while analyzing aggregated user requests to better understand the different kinds of ways (well-resourced) actors use to distill their most capable models. > "The data will hel
9.
by ayewo 1mo ago
Yup. The original source was Kyle Daigle, GH COO. - https://x.com/kdaigle/status/2040164759836778878 - https://xcancel.com/kdaigle/status/2040164759836778878
10.
by ayewo 1mo ago
Isn’t that the optimal strategy? That an LLM trained to be a paper-clip maximizer chose the optimal strategy is in my opinion the most plausible outcome.
11.
by ayewo 1mo ago
> 1) AI researchers talk and change companies often, so techniques circulate. This feels implausible because training and shipping a new model ought to take longer than 2 months? The assumed timeline (2 months) is slightly wrong becaus
12.
by ayewo 1mo ago
Thanks for the PyTorch internals OG link and for putting together your PyTorch one-pager, if we can call it that :) Kudos. Off-topic: Some of your comments in this thread are dead, but I’ve vouched for some them as I don’t see anything wron
13.
by ayewo 2mo ago
Perhaps this was due to their red-teaming partnership [1][2] with Anthropic which they wrote about a few months earlier in March? 1: https://www.anthropic.com/news/mozilla-firefox-security 2: https://blog.mo
14.
by ayewo 2mo ago
How did you unearth such a remarkable find of 2 different faculty individuals named "Philipp Otto" collaborating on the same paper :) ?
15.
by ayewo 2mo ago
Yep. Cursor’s Composer 2 model is a good example, though it is not clear if they entered into an agreement with Moonshot before they got found out in March this year [1] or after. 1: https://x.com/fynnso/status/203
16.
by ayewo 2mo ago
The top poster mentioned LLM spend of millions/month to justify the estimated capex of $6m to self-host Kimi on own infra. Add to this number another $1.5m/yr in opex, so not sure I’d call such an enterprise wealthy enough to spen
17.
by ayewo 2mo ago
Understood but sharing your existing devops resources with this will soon become a bottleneck especially when any major downtime will keep several engineers (and long-running agents) blocked from any meaningful work until availability impro
18.
by ayewo 2mo ago
They do. It’s a reasonable stance [1] that is aptly captured by this quote: “I doubt very much if it is possible to teach anyone to understand anything, that is to say, to see how various parts of it relate to all the other parts, to have a
19.
by ayewo 2mo ago
In your opinion, do you think Internet Protocol Version 8 (IPv8) [1] stands a chance to fix the mistakes of IPv6 after more than 20 years now? Or there is too much inertia for IPv8 to overcome to become a truly backwards compatible extensio
20.
Everyone's Watching the Wrong Benchmark(gradientvc.substack.com)
1 points by ayewo 2mo ago|0 comments
21.
by ayewo 2mo ago
I'm not sure if you are aware, but you have to approach prompting Fable slightly differently from a model like Opus. It's important to include the reason aka the why of your task [1] in your prompt. You'll get more mileage if
22.
by ayewo 2mo ago
> Sol had all the trimmings, Terra had the least. That’s interesting. My understanding is that: Sol = Sun Terra = Earth Luna = Moon So it’s a bit surprising that in Toyota’s nomenclature, Terra is the basic trim instead of Luna.
23.
by ayewo 2mo ago
A Google DeepMind researcher (Neel Nanda) was able to replicate their claims on an open weight model (Qwen 3.6 27B): > We have replicated the core claims on Qwen 3.6 27B, and also share preliminary evidence of extending this work by fin
24.
by ayewo 3mo ago
You posted in the wrong thread, hopefully dang will swing by to remove this subthread. Please post here: https://news.ycombinator.com/item?id=48747975
25.
by ayewo 3mo ago
Taalas https://taalas.com/the-path-to-ubiquitous-ai/ Previous HN discussion: https://news.ycombinator.com/item?id=47103661
26.
by ayewo 3mo ago
You are correct. Notwithstanding, people have been expressing the gp's sentiment for like a decade now [1] as is evident in this sub-thread [2], so it's a losing battle trying to prevent people from making such comparisons. 1: 24-
27.
by ayewo 3mo ago
1. How did you land the side gig? Mercor or a lessor known brand? 2. What criteria do such vendors typically require?
28.
by ayewo 3mo ago
> Chinese transfer stations? For anyone that doesn't get the reference, please start here [1]. 1: https://www.chinatalk.media/p/how-to-buy-cheap-claude-tokens...
29.
by ayewo 3mo ago
> If I want to speed things up, I remove people from the team. Everything gets easier. How did you land on this approach? From someone you learned from or from a seminal book like Mythical Man Month [1]? > This was true long before
30.
by ayewo 3mo ago
For anyone that is confused like I was, the quoted text I'm replying to was copied from page 13 of the system card [1] and not the model announcement page, which this HN discussion is linked to. 1: https://www-cdn.anthropic.

More ›