Dataf3l

@rachelnabors For bad logic try Monty python’s duck sketch maybe idk

I tried to find a gif for "wrong algo" but doesn't exist yet

I have a whole article on sustainable AI coming up, and it's just "use smaller models" and "you don't need and LLM for that." I thought this stuff was obvious, but there's a whole class of engineers out there that think Meta's investing in LLMs to make a better recommendations

@ml_collective @nsaphra I sincerely enjoyed your talk, it was inspirational and very educational, amazing work

Starting in 30 min at 10am PT! @nsaphra presents at DLCT: Sudden Drops in the Loss: Syntax Acquisition, Phase Transitions, and Simplicity Bias in MLMs Angelica Chen, Ravid Shwartz-Ziv, Kyunghyun Cho, Matthew L. Leavitt, Naomi Saphra arxiv.org/abs/2309.07311 Zoom 👇

@Keir_Starmer This tweet was featured youtube.com/watch?v=9Ch4a6… on this video, have a nice day friends.

It’s time to look forward — to move on from the stale old political fights and to find common sense, practical solutions that improve the lives of British people. Secure borders. Cheaper bills. More jobs. We will close a deal in the national interest.

@Wattenberger hello! I hope you are having a great day, I would appreciate your opinion on this youtube.com/watch?v=VXktVb… which is related to TEXT levels of abstraction and the FISHEYE document you so excelently created wattenberger.com/thoughts/fish-… which I loved!!!

@Jonifico i saw you in the HN Map, let's hangout IRL ?

@Romantisismica @iPolitologa @silenceway @pezick @Ndermus @cdorjuelav sumerced es en el 7 de agosto?

@iPolitologa @silenceway @pezick @dataf3l @Ndermus @cdorjuelav Claro que sí. Conocí un lugar increíble de comida colombiana. Se llama Sopas de la Tierrita, comida Santandereana, Paisa, Costeña. 10/10. (Venden chicharrón y chunchullo).

10/10 el vestido rosa, el collar, la bufanda y todos los regalos preciosos que me dieron en mi cumpleaños. Gracias a @silenceway @pezick @dataf3l @Ndermus Wen y @cdorjuelav. Y a todos por compartir conmigo. Amor profundo. 🙌

@levelsio @GroqInc are we sure we need the rapist ai? what about the ethics of the thing?

Nice to be on this list with 🧠 TherapistAI.com as one of the first @GroqInc customers If you're building anything that needs lightning fast LLM replies, try Groq

@DamonLinker condoleza rice was not "criminally bad" at her job, she is a criminal, period, you defending her actions, shows no moral compass. why did you sell out? for a bag of coin?

@CondoleezzaRice I saw a lot of news 10 years ago about condoleza rice being evil, one wonders, is there rule of law? is she supposed to be in prison? did people die because of her actions?

@morciatka i saw a post about a X based OSX ui and have some questions, can you email me at dataf8L@gmail.com ?

@_briancorrea no idea exactly how many people search for those tho, some research is required

@_briancorrea I have no insights, but maybe having a title be a question that may be on people's minds, and answering the question while showing the tool, like for instance "how to organize my files better" or "I can't find anything on downloads folder" or "my stuff is too disorganized"

@_briancorrea I reviewed your website, I think people want it, but they don't know they want it, it's kinda like Henry Ford and Horses, I think the problem may be "customer education" as to what is possible in my opinion, and the other problem is that MSFT/AAPL might write the feature yknow?

@_briancorrea we have the same downloads folder

@dickpountain Hi mr Pountain, I am looking for an article titled WRITE YOUR OWN COMPILER which featured in the SOFT magazine Issue 1, where your article on OCCAM's CURTAIN RAISER was published, do you know where can I find a PDF of the ISSUE 1 of SOFT by any chance? Thanks!

@tangentstorm I was reading this: tangentstorm.github.io/winfield-pasca… but I cannot the ISSUE 1 of "SOFT" anywhere on the internet, I tried archive.org etc, to no avail, do you know where can I find said magazine?

@hbouammar a LLM is a black box. if somebody publishes something useful, I don't see a reason not to read it. I read this particular paper: arxiv.org/pdf/2307.11760… and found it very interesting. perhaps if we were to follow your idea, this paper would not exist. something to consider.

With this hype around LLMs, I am seeing tons of well respected professors publishing half (if not quarter) - baked work just to keep up with the competition! I will advice them as they did advise me once during my PhD. This has to stop, focus on quality and not quantity… I feel for you, but you are better than that!! Take your time, flush out the work and then arxiv it! It is starting to really bother me that the quality of papers I am reading is steadily decreasing. Sorry, but a new prompt for GPT4 is not a paper, a let’s think step by step is not a scientific discovery, a let’s go backwards and rethink prompt is not science, one experiment on one data set with seed picking is not worthy reporting. We need to grow up again and redo science! Additionally, we all need to understand this is just one good empirical result, now we need to make it useful… and oh yea will you also cut this terminator nonsense as well .. thanks! #ai #MachineLearning

@0xbnomial "Application error: a client-side exception has occurred (see the browser console for more information)." looks like the site is down, maybe analytics problem? idk

@NorthstarBrain @OpenAI I am interested in the guide, (not sure how to DM)

@OpenAI Wild news! At least my custom made QR code is a bit more pretty. Feel free to DM me for the guide. P.S. don't scan it :)

Code Interpreter will be available to all ChatGPT Plus users over the next week. It lets ChatGPT run code, optionally with access to files you've uploaded. You can ask ChatGPT to analyze data, create charts, edit files, perform math, etc. Plus users can opt in via settings.

@eric__ssn604323 yo check this out dude youtu.be/nc9kq9ZkNV8?t=… they mention you specifically, let me know when I delete this tweet I do it.