Bogus article. They didn't write anything to the drive, they faked writes to run up the drive's write counter to test whether the controller would declare the drive worn out despite the flash not being touched.
Yes, even the video's thumbnail explains this, I'm not sure how the article author managed to get it so badly wrong. (Although I'm not sure why it would need a windows bug, surely it's more of a drive firmware bug?)
I live in Poland and speak Polish but tend to have my devices set to English because of how low quality UI translations tend to be. The endless confusion this causes to software is amusing. Worst/best of all is Google Maps on Android, in which random parts of the interface, including voice directions, can independently be in Polish or English, and which is which changes on its own on a daily basis.
It's the latter. Some people abuse the system by highlighting the full length of every broadcast, turning their highlights section into a complete archive of their streams, which is not something Twitch ever wanted to offer.
I don't think it's fair to say twitch "never wanted to offer" when not long ago, that behavior was the base functionality. You could rewatch everyone's entire streams forever. There was a DMCA scare at some point when streamers were getting in trouble for their old streams having music and many took down all of their history, but before then you'd see years worth of streams for people
I did that, but not as a way to abuse the system. I used to export all my streams to YouTube directly from Twitch without downloading it first. I would just trim the beginning of the stream and sometimes split in more than one video if I had multiple content in one stream. I have hundreds of videos starting from 2018. I just thought this was ok and now I'm going through the effort of exporting them individually to a youtube account. I wish they had offered at least a way to export or download them in batch.
>Asking ChatGPT
That alone makes me want to discard your entire argument, but the answer is pretty simple: cost
Hosting huge amounts of video data throughout multiple data centers is just plain expensive.
Some ads isn't gonna fix that, especially since twitch already has a frustrating amount of ads everywhere.
Every time anyone posts anything from ChatGPT, it gets immediately and massively downvoted. It’s clear the community doesn’t want this; please don’t post it!
Very true. The most recent "leak" was of a document that is available on archive.org and has been for years[1]. But the bottoms of the pages have "NATO RESTRICTED" printed, so it was declined, and news sites once again exploded with "War Thunder player leaks secret military documents on forum!!"
The requirements of orbital launch are unyielding. If you make a car 50% heavier, it will have worse mileage and handling, but it will still get you where you need to go. If you make a spacecraft 50% heavier, it will never reach orbit.
A good rule of thumb is that models can be quantized to 6 to 8 bits per weight without significantly degrading quality. This is convenient for the math: 70GB plus some overhead for the attention matrices (ongoing requests). This overhead depends on workload and context lengths, but you should expect about 30% more. So, around 100GB for a server under load.
Funnily enough, of all that I've tried, the model by the best at writing porn has been not one of ones uncensored and tuned exactly for that purpose, but stock Command R - whose landing page lists such exciting uses as "suggest example press releases" and "assign a category to a document".
Usually "uncensored" models have been made by instruction tuning a model from scratch (i.e. starting from a pretrained-only model) on a dataset which doesn't contain refusals, so it's hard to compare directly to a "censored" model - it's a whole different thing, not an "uncensored" version of one.
More recently a technique called "orthogonal activation steering" aka "abliteration" has emerged which claims to edit refusals out of a model without affecting it otherwise. But I don't know how well that works, it's only been around for a few weeks.
I've seen some of the "abliterated" models flat-out refuse to write novels, other times they just choose to skip certain plot elements. Non-commercial LLMs seem to be hit or miss... (Is that a good thing? I don't know, I just screw around with them in my spare time)
I'll try command-r though, it wasn't on my list to try because it didn't suggest what it was good at.
