adt

11 ms·

121.
by adt 3y ago
https://lifearchitect.ai/models-table/
122.
by adt 3y ago
"With automation replacing so many blue-collar jobs, and pretty soon white-collar jobs, we may need to consider bigger changes. We should start talking about that now. Things like a shorter work week, or a universal basic income, or gu
123.
by adt 3y ago
September 2023 was a massive month for LLM releases. I see 14 big highlights including the LeoLM family, and we haven't even finished the month yet... Models table: https://lifearchitect.ai/models-table/ Models ti
124.
by adt 3y ago
More comprehensive list: https://lifearchitect.ai/models-table/
125.
by adt 3y ago
Added: https://lifearchitect.ai/models-table/ https://lifearchitect.ai/timeline/
126.
by adt 3y ago
Bruce Schneier? How the mighty have fallen. In the US, 16 people were murdered today (by another human). Worldwide, one person is murdered every second, or 1,440 murders per day. Perspective...
127.
by adt 3y ago
Might be old (2020), as OpenAI models are now also covered by Quora for Poe.com and others: https://blogs.microsoft.com/blog/2020/09/22/microsoft-teams-...
128.
by adt 3y ago
Related table of models: https://lifearchitect.ai/models-table/
129.
by adt 3y ago
These guys were pioneers, bringing out the first LMS pre-popular-web (Aug/2002). The founder lives right here in sunny Perth (also the birthplace of the founder of Canva)!
130.
by adt 3y ago
Exactly. The False Promise of Imitating Proprietary LLMs (UC Berkeley, 25/May/2023) https://arxiv.org/abs/2305.15717
131.
by adt 3y ago
From the GPT-4 paper (pp20): 2.13 Overreliance ...despite GPT-4’s capabilities, it maintains a tendency to make up facts, to double-down on incorrect information, and to perform tasks incorrectly. Further, it often exhibits these tendencies
132.
by adt 3y ago
>"While 7B models were shown to be able to be run on phones before, QLORA is the first method that would enable the finetuning of such models. We estimate that with an iPhone 12 Plus, QLORA can finetune 3 million tokens per night wh
133.
by adt 3y ago
This thread is amazing, and has shades of the original iPod and iPhone launch [1,2] commentary. I can't believe that people would go on permanent written record with this stuff, but humans gotta human! 1. iPod launch thread on macrumor
134.
by adt 3y ago
It's very BOFH... https://www.theregister.com/2002/11/21/the_killer_bofh_bot/
135.
by adt 3y ago
Here's my estimate: https://lifearchitect.ai/bard#dataset
136.
by adt 3y ago
It is definitely PaLM 2: Versions Resource ID Release date Release stage Description text-bison@001 2023-05-10 Public Preview Quality improvements and restage -001 as the first stable base model release https://console.cloud.goog
137.
by adt 3y ago
Here's my comprehensive version (123 LLMs to PaLM 2 @ 10/May/2023), without the 'ranking' part: https://lifearchitect.ai/models-table/
138.
by adt 3y ago
PaLM 2 paper: https://ai.google/static/documents/palm2techreport.pdf
139.
by adt 3y ago
Excellent. Some of the theory here goes back to Oct/2021 and beyond [1]. The riffusion.com [2] guys made this practical. Also, my video showing high-level overview and examples [3]. 1. SpecVQGAN: https://github.com/v-ia
140.
by adt 3y ago
If they don't already own the data (Alphabet/YouTube, Meta/FB, etc.) they do scrape it, or use ready-made datasets. Comprehensive analysis paper (the 'what' not the 'how'): https://lifearchitect
141.
by adt 4y ago
The Memo: https://lifearchitect.ai/memo/
142.
by adt 4y ago
DeepMind (apparently there's some friction between those two!)
143.
by adt 4y ago
This is close, table of LLMs as released, and I try and add repos for the 'open' models: https://lifearchitect.ai/models-table/
144.
by adt 4y ago
Under-rated comment...
145.
by adt 4y ago
+ LA https://techcrunch.com/2023/02/27/waymo-to-test-driverless-r...
146.
by adt 4y ago
A list of the first 50 apps/organizations powered by OpenAI's GPT-4. More: https://lifearchitect.ai/gpt-4/
147.
by adt 4y ago
Also worth noting that CS India hired ChatGPT as their CEO; the first company to do so. https://www.indiatimes.com/technology/news/organisation-appo...
148.
by adt 4y ago
OpenAI are using it internally a lot; they used it as a co-author on the GPT-4 paper, and 'with great impact on functions like support, sales, content moderation, and programming.' Here's 25+ more of the first GPT-4 enterpris
149.
by adt 4y ago
Great! I always cite my sources, and you'll find a link to your page as usual. I wanted to point you towards OpenAI's FIM 6.9B as well. Trained on 100B tokens (Chinchilla-aligned), it was announced just before GPT-4 allegedly star
150.
by adt 4y ago
Nope

More ›