porridgeraisin

4 ms·

61.
by porridgeraisin 1mo ago
Literally many previous instance used one. Right from alphaevolve onwards. I'm not some "LLM is just a next token predictor guy" (GP seems to have a thing against LLMs), but to use LLMs properly you genuinely do need a ground
62.
Pipelined SQL in ClickHouse 26.8(clickhouse.com)
1 points by porridgeraisin 1mo ago|0 comments
63.
by porridgeraisin 1mo ago
Yep. It's not that important unless you're really just relying on one shot responses.
64.
by porridgeraisin 2mo ago
Many of the other search APIs are better than in the inbuilt claude or codex ones. You can try this out yourself by e.g enabling exa in codex plugins.
65.
Sidekick's continual learning loop (2026) – Shopify(shopify.engineering)
1 points by porridgeraisin 2mo ago|1 comments
66.
GPT-6 Astra Achieves SOTA on ARC-AGI(twitter.com)
3 points by porridgeraisin 2mo ago|0 comments
67.
by porridgeraisin 2mo ago
HF is an american company.
68.
by porridgeraisin 2mo ago
https://archive.is/h6ysi
69.
Inside Google’s $200bn Wall Street finance machine for Anthropic(ft.com)
19 points by porridgeraisin 2mo ago|2 comments
70.
Welcome to Pintos(pkuflyingpig.gitbook.io)
2 points by porridgeraisin 2mo ago|0 comments
71.
Is There I/O After Death? What Happens to Io_uring When a Process Dies(blog.ydb.tech)
86 points by porridgeraisin 2mo ago|36 comments
72.
by porridgeraisin 2mo ago
agy is good for those cases where you are willing to put the effort into the harness specifically for a task or family of tasks. The full suite, with evals, monitoring, hooks, custom tools, custom verifiers, etc,. It is not good if you want
73.
by porridgeraisin 2mo ago
Yeah the speed in agy cli is amazing. Whole files get written and "py_compile"d in a single blink of the eye its crazy. In india, my telco gives me google ai pro for free. And agy with flash goes a long way.
74.
by porridgeraisin 2mo ago
Yep. It has access to much better route planning tools than the other models. The results are really good IME.
75.
Is this the simplest (and most surprising) sorting algorithm?(arxiv.org)
3 points by porridgeraisin 2mo ago|0 comments
76.
Lachy Groom backs Indian startup aiming to keep aircraft aloft for a year(techcrunch.com)
2 points by porridgeraisin 2mo ago|0 comments
77.
GoPro Is Being Bought for $285M and Pivoting to AI and Defense Optics(fstoppers.com)
7 points by porridgeraisin 2mo ago|1 comments
78.
by porridgeraisin 2mo ago
In my unpopular opinion, it wasn't transformers or attention, but pretraining on language data that kicked off the LLM Boom. Alec radford in his little jupyter notebook trained a very small non-transformer to predict simply the next-ch
79.
by porridgeraisin 2mo ago
@mods I am not sure the editorialized title is better! (It is a bit clickbaity, since this is a transformer that does not compete with LLMs at all outside the arc-agi-1 benchmark)
80.
by porridgeraisin 2mo ago
True, but I am not sure how much uptake it has in their enterprise accounts. Slowly they are all coming to only care about that.
81.
by porridgeraisin 2mo ago
With compute crunches and everything I am not sure it makes sense for anthropic to commit to haiku as an endpoint and thus a product. There is no telling they aren't using a similarly sized model behind their existing opus/fable e
82.
by porridgeraisin 2mo ago
It does not, if it gets the answer (or any information about them, even % of qns solved) and is able to adjust itself in response, then that is considered training on the test set and is wrong.
83.
by porridgeraisin 2mo ago
They have _not_ trained on the test set. On the private test set, the right way to evaluate this type of model, is giving i it the test question Q, which it will first train to AR predict first, and then it will inference using the just-u
84.
by porridgeraisin 2mo ago
Basically, you have a bunch of Q,A pairs in the training dataset. Here, it was trained to next-word predict the question itself, as well as next-word predict the answer given the question as prompt. This is bog-standard, no one's compl
85.
Io_uring Without Readahead(frn.sh)
121 points by porridgeraisin 2mo ago|45 comments
86.
I trained a small transformer in 1.5hrs and it beats many LLMs(mvakde.github.io)
671 points by porridgeraisin 2mo ago|168 comments
87.
Omarchy Quattro is the #3 Linux distro by active 1Password users(twitter.com)
6 points by porridgeraisin 2mo ago|1 comments
88.
by porridgeraisin 2mo ago
No, semi fabs do not take IOUs or other similar. They only do prepayments and non cancellable take or pay contracts. The money is here and real. There is a lot of SPV backed financial engineering upstream, but not here.
89.
by porridgeraisin 2mo ago
Just see the top of this thread. The theme of the comments is that the SK Hynix, Micron, etc CEOs are lying about this in order to collectively increase prices artificially.
90.
by porridgeraisin 2mo ago
Some of the backlogs apart from the "top level" one are not being viewed that positively from what I have heard. One of my football friends works at Vertiv's India office, soon transferring to Thane, they basically sell cooli

More ›