m_ke

16 ms·

211.
by m_ke 3y ago
I'm not sure how this would work with their funding / ownership structure but a delaware c corp with such a high valuation and so much money raised would have to take a huge down round if it lost half the team and pissed off their
212.
by m_ke 3y ago
He is one of the most well connected people in tech and probably responsible for all of their partnerships and fundraising success. His departure in this fashion would make it really hard for them to raise as much money as they will need an
213.
by m_ke 3y ago
The last 10 years of deep learning progress are all thanks to Open Source and Open Research. 95% of LLM progress in the past year is also thanks to open models, which allowed the field to work on things like QLoRa, long context lenghts, RAG
214.
by m_ke 3y ago
I'm currently 60 minutes in, waiting on hold for American Airlines support, because I clicked on the cancel flight button on my trip details page hoping to get a warning page with details of my cancellation options, but instead it just
215.
by m_ke 3y ago
I wonder if Wojciech Zaremba will leave as well
216.
by m_ke 3y ago
A lot of engineers are interested in working on important things, not growing at all cost in hope of a large exit.
217.
by m_ke 3y ago
Sounds to me like it wasn't just Ilya, but the whole research and engineering team that was tired of his shit. See: https://www.reddit.com/user/Anxious_Bandicoot126
218.
by m_ke 3y ago
Their main value add was Sam's network, which is probably why they were able to raise so much money, close the deal with Microsoft and get a ton of YC companies to sign on early on. Now that they're so well established and intertw
219.
by m_ke 3y ago
ChatGPT was definitely not some visionary project led by Sam. They had a great LLM in GPT-3 that was hard to use because it wasn't instruction tuned, so the research team did InstructGPT and then took it even further and added RLHF to
220.
by m_ke 3y ago
I was just thinking about this too after using ChatGPT-4 a bunch this past week. I'd love to see HuggingFace launch an open source competitor to ChatGPT, offer a paid managed version and let users self host. I'd pay 3-4x more for
221.
by m_ke 3y ago
https://blog.mozilla.org/en/mozilla/announcing-the-initial-r...
222.
by m_ke 3y ago
Yeah that was my concern as well. I used to work for an AI company that ended up in the same predicament, they raised way too much money at a high valuation and in the process cut down the potential acquirer pool to 3-4 companies. They are
223.
by m_ke 3y ago
MS makes sense since they could integrate it with github and azure into an ML platform that they could sell to their enterprise consumer base. It would also be a great addition to teams, bing and word suite.
224.
by m_ke 3y ago
As someone who has dealt with countless breaking changes in keras and wasted days of my life attempting to upgrade, no thank you. My pytorch code from years ago still works with no issues, my old keras code would break all the time even in
225.
by m_ke 3y ago
Since people are suggesting alternatives, I'd like to shoutout skypilot: https://github.com/skypilot-org/skypilot EDIT: looks like this actually uses it under the hood: https://github.com/run-house
226.
by m_ke 3y ago
Judging by their glassdoor reviews they're not going to make it: https://www.glassdoor.com/Reviews/Stability-AI-Reviews-E7022...
227.
by m_ke 4y ago
I mentioned when the deal happened that he might bring it under X and use all of the losses he's about to take on this deal to avoid ever paying taxes again. https://news.ycombinator.com/item?id=33603794
228.
by m_ke 4y ago
Yeah sorry about that, I was just repeating something I heard on a podcast and it looks like they assumed that was top line revenue and took 10% of it. I’d edit the post but it’s too late to do so. https://news.ycombinator.com&#x
229.
by m_ke 4y ago
I couldn’t reply sooner because I have noprocast on but my numbers were a bit off, I was repeating what I heard on a tech podcast. According to the numbers they had to release it was a bit higher https://www.axios.com/2023&#
230.
by m_ke 4y ago
They raised a ton of money on a high valuation, spent 25mil to make 1 mil last year and are now scrambling to raise a new crowd sourced round because they don't want to get wiped out in a down round. This notes thing looks like an atte
231.
by m_ke 4y ago
I think this has a lot more to do with the US having a larger english speaking population under one legal framework. All of the capital goes here first since it has the most upside and makes it easier to expand to other english speaking cou
232.
by m_ke 4y ago
I never watched any Willink stuff before but gave a listen to his interview on Huberman's podcast and he had a great quote that goes something like "discipline beats motivation". Looking it up on youtube here's another c
233.
by m_ke 4y ago
ChatGPT is already better at copywriting than 90% of startup founders and marketing people at big cos. You'll soon be able to let it generate 1000s of different versions of marketing material to A/B test or personalize based on us
234.
by m_ke 4y ago
OpenAI is working with Microsoft so they definitely had access to the full Bing index and data from their other platforms like Github and Linkedin. They also paid for private datasets, from what I heard they might have gotten a copy of Quor
235.
by m_ke 4y ago
Ask chatgpt to implement some of the things you worked on the last few months. I was very skeptical too until I tried this. Here are some sample prompts that I tried and got full working code for: - "write pytorch code to train a trans
236.
by m_ke 4y ago
https://wersm.com/hugging-face-chatbot-is-the-best-artificia...
237.
by m_ke 4y ago
That's because they started out as a chat bot for teens. I wonder if they'll ever attempt to revive that product, backed by LLMs.
238.
The Emergence of General AI for Medicine (Lecture by Peter Lee on ChatGPT)(youtube.com)
1 points by m_ke 4y ago|0 comments
239.
by m_ke 4y ago
Yeah I share these concerns as well ( https://news.ycombinator.com/item?id=35305791 ). My hope is that some countries will see this as an opportunity to expand their safety nets and reduce the work burden on their citizens, w
240.
by m_ke 4y ago
GPT-4 might be close to the best we'll get on the general LLM model front for a while since they trained on a huge chunk of web text. Next real advances will probably be in tuning them for specific applications in medicine, law, accoun

More ›