ipsum2

19 ms·

331.
by ipsum2 2y ago
It got cancelled/postponed.
332.
by ipsum2 2y ago
There's no good open source agent models at the moment unfortunately.
333.
by ipsum2 2y ago
Yes, that's a great point that I missed. From anecdotal evidence, it seems more people are using supercomputers for ML use cases, that would have been traditionally done by HPC. (eg training models for weather forecasts)
334.
by ipsum2 2y ago
Nvidia GPUs are also used for inference on Google products. It just depends on availability.
335.
by ipsum2 2y ago
AMD GPUs handle all inference for Llama3 at Meta btw.
336.
by ipsum2 2y ago
As someone who worked in the ML infra space: Google, Meta, XAI, Oracle, Microsoft, Amazon have clusters that perform better than the highest performing cluster on Top500. They don't submit because there's no reason to, and some wa
337.
by ipsum2 2y ago
The long context model has not been open sourced.
338.
by ipsum2 2y ago
People read text on the OLED screen of the iPad Pro all the time.
339.
by ipsum2 2y ago
There's a really obvious way to test whether the strawberry issue is tokenization - replace each letter with a number, then ask chatGPT to count the number of 3s. Count the number of 3s, only output a single number: 6 5 3 2 8 7 1 3 3 9
340.
by ipsum2 2y ago
People are doing this all the time with Claude 3.5.
341.
by ipsum2 2y ago
What you're saying is an explanation what I said, but I agree with you ;)
342.
by ipsum2 2y ago
The more obvious alternative is that CoT is making up for the deficiencies in tokenization, which I believe is the case.
343.
by ipsum2 2y ago
You can mask out the tokens at the end, so its technically autoregressive.
344.
by ipsum2 2y ago
You're right, I was referring to the former but got mixed up with the different Zeros.
345.
by ipsum2 2y ago
This article is very out of date and irrelevant now. The Chi-fi market has changed drastically in the last 4 years. You can get very neutral sounding IEMs from China now, V-shaped, U-shaped, sparkly, heavy bass, etc. One of the best-soundin
346.
by ipsum2 2y ago
Original: https://www.seevua.com/invention/s-0023 (in Japanese, browser translation does a good job)
347.
by ipsum2 2y ago
The title of the submission was modified. It you read the article it says: Audio Decomposition [Blind Source Seperation]
348.
by ipsum2 2y ago
I must be dumb, but none of the YouTube video demos are demonstrating source separation? Edit: to clarify, source separation in audio research means separating out the audio into separate clips.
349.
by ipsum2 2y ago
It's free advertising for the company. Perplexity is a bad product from my own experience, but their marketing is top notch.
350.
by ipsum2 2y ago
This is just some lighthearted trolling, and its hilarious that some are taking the bait. No one believes Perplexity will replace anyone anytime soon, based on how much hallucination it does currently.
351.
by ipsum2 2y ago
Nvidia buying Intel would be doing Intel a great service.
352.
by ipsum2 2y ago
Nvidia would've acquired the culture of Intel, so it wouldn't be the company it was today if that happened.
353.
by ipsum2 2y ago
I don't think its a Go-specific issue, its useful syntactic sugar. I would love an updated JSX for Python.
354.
by ipsum2 2y ago
> AI elevator background "music". What are some examples? I haven't encountered this.
355.
by ipsum2 2y ago
I don't know about you, but o1-preview/o1-mini has been able to solve many moderately challenging programming tasks that would've taken me 30 mins to an hour. No other models earlier could've done that.
356.
by ipsum2 2y ago
What is a "scissor concept"? Google returns no results.
357.
by ipsum2 2y ago
Sure. You're wrong and a cynic. The model does not connect to Google servers (unless you decide to use their colab, which is optional), you can use it offline without contributing anything back in data or code.
358.
by ipsum2 2y ago
> starting an entirely new project were The project was started in 2012, long before yt-dlp (but not before youtube-dl)
359.
by ipsum2 2y ago
I've also considered SQLite but decided against it. Using a custom compression means that you can't do a full text search while compressed on the file like you would in btrfs or zfs.
360.
by ipsum2 2y ago
People think of chatGPT now when they hear AI, but for on-device its more like speech recognition, image keyword search, and keyboard next word prediction.

More ›