Information from Documents
@docs2info
Read documents. Not too many. Mostly PDFs. Our parents read and wrote documents. Having the world's knowledge an internet connection away doesn't change this.
قد يعجبك
Is Vibe Coding to modern software development as the Sea People were to the Hittites, Egyptians, and Mycenaeans.
You probably know @clattner_llvm built the foundation of the software we all use today. Well now Chris is concerned: AI might create a whole cohort of coders that don't understand how to built software that lasts. Watch the full video to be sure you don't end up one of them:
I'm proud to partner with @samofort , @kevbtu, Ramine and the entire team at DFJ Growth to help scale into the next epoch of Modular's journey. Their summary of the path ahead says it better than we do: 👇
Today, we’re proud to back @Modular $250 million Series C financing and its mission to build AI’s unified compute layer. dfjgrowth.com/story/modular-…
There is a deep reason why LLMs need to be non-deterministic: it's because human language is non-deterministic. If it was, we could not use it for communicating our internal states, unobservable from the outside. LLMs of course don't have these intentions, so nondeterminism is…
I haven't enjoyed an AI talk as much as this one for a long time. A smart person doing practical work. Just chipping away at a problem intelligently and diligently. youtube.com/watch?v=pxPCKR…
youtube.com
YouTube
Grokking, Generalization Collapse, and Dynamics of Training Deep...
youtube.com
YouTube
Interpretability: Understanding how AI models think
in this work, we show that reasoning texts are really not transparent, in the sense that we (humans) are really bad in inferring how the steps (really) depend on each other for the model. yoo can test yourself here: do-you-understand-ai.com
Producing reasoning texts boosts the capabilities of AI models, but do we humans correctly understand these texts? Our latest research suggests that we do not. This highlights a new angle on the "Are they transparent?" debate: they might be, but we misinterpret them. 🧵
The greatest joy modern developers experience is putting their, or their employers' or customers' data into a proprietary online service. LLM foundation model, data store etc. linkedin.com/pulse/githubs-… They do this from laptops that are more powerful than 1980s super-computers.
Producing reasoning texts boosts the capabilities of AI models, but do we humans correctly understand these texts? Our latest research suggests that we do not. This highlights a new angle on the "Are they transparent?" debate: they might be, but we misinterpret them. 🧵
I still can't believe people took scree plots seriously, let alone reverentially. youtu.be/geUH6K8o9nw?si…
youtube.com
YouTube
UWEE Research Colloquium: November 8, 2016 - David Donoho, Stanford...
I vibe coded a Turing Machine accepting 0^n1^n for the Theory of Computing class I'm teaching this fall. inquisitive-lolly-1bb788.netlify.app Took me about 10 minutes on bolt.new. This is just too easy.
the OCR conjecture: some examples include artifacts such as OCRV ROOT, which indicate the training data may have been reading between the lines: OpenAI is scanning books (for some reason the model loves mentioning how many deaf people live in Malaysia)
> sign up for new AI coding product > it’s free, get an email asking setup a meeting to onboard My brother, this is the most expensive software imaginable.
Swedish Prime Minister is using AI models "quite often" at his job. He says he uses it get a "second opinion" and asks questions such as "what have others done?" At the moment he is not uploading any documents. IMO, when these models are capable of giving seemingly better…
This is how one of the big LLM vendors' chatbot told me to avoid big LLM vendor lockin. docs.google.com/document/d/1Jw… "Build an Abstraction Layer This is your single most powerful defense." This is the advice our parents gave about pre-AI vendor lockin years ago.
Vibe code is legacy code @karpathy coined vibe coding as a kind of AI-assisted coding where you "forget that the code even exists" We already have a phrase for code that nobody understands: legacy code Legacy code is universally despised, and for good reason. But why? You have…
The go file editing agent example is impressive. github.com/johndocs/agent… shows a convoluted way to write a javascript fizzbuzz.
A simple example of the variation between runs of a simple Claude 3.7 Sonnet code summarization from the Go agent sample. github.com/johndocs/agent… For some reason, the variation has reduced dramatically. With so many factors to logged, I missed variation.
A simple example of the variation between runs of a simple Claude 3.7 Sonnet code summarization from the Go agent sample. github.com/johndocs/agent… For some reason, the variation has reduced dramatically. With so many factors to logged, I missed variation.
There's so much variation between runs on this simple agent that I struggle to see how people are going to use libraries like this in commercial products.
There's so much variation between runs on this simple agent that I struggle to see how people are going to use libraries like this in commercial products.
Simple Agent code in Go github.com/johndocs/agent copied from ampcode.com/how-to-build-a…
Simple Agent code in Go github.com/johndocs/agent copied from ampcode.com/how-to-build-a…
United States الاتجاهات
- 1. Charlie Jones 1,175 posts
- 2. LaPorta 2,848 posts
- 3. #HardRockBet 3,663 posts
- 4. #Skol 1,040 posts
- 5. #OnePride 1,460 posts
- 6. Theo Johnson N/A
- 7. Cody Barton N/A
- 8. #HereWeGo 1,924 posts
- 9. Scott 111K posts
- 10. #AskFFT N/A
- 11. Haaland 46.6K posts
- 12. Highsmith N/A
- 13. Richard Hightower N/A
- 14. Darius Slay N/A
- 15. Alec Pierce N/A
- 16. Cherki 26.2K posts
- 17. Metellus N/A
- 18. Myles Price N/A
- 19. Bob Trumpy N/A
- 20. Full PPR 1,846 posts
قد يعجبك
-
Piotr Nawrot
@p_nawrot -
EU-LAC Digital Accelerator
@EU_LAC_Accelera -
Zekun Wang (ZenMoore) 🔥
@ZenMoore1 -
Sankeerth
@trippy_hustler -
Puyuan Peng
@PuyuanPeng -
Harshit Joshi
@harshitj__ -
Forrest
@m_forrest -
AcroCrypto
@AcroCrypto -
Yujia Qin
@TsingYoga -
Ansong Ni
@AnsongNi -
Reliable Technology Services
@ReliableTSI -
Kalyan KS
@kalyan_kpl -
Edward Hu
@edward_s_hu -
Boshi Wang
@BoshiWang2 -
Vineet
@SaiVineet2
Something went wrong.
Something went wrong.