Very important read - https://t.co/dgAPyMBHkz
What do you want to write on your tombstone? I am sure that's not going to be your net worth or $$s.
So, are you working towards that?
@AjeyGore Reading this on a recommendation at the moment https://t.co/MGYLR4wQoU. Can highly recommend. In a gist, the body knows how to fix itself given the right biophysical inputs, which are more important than biochemical inputs.
My whole career sits on top of C/C++ , OS, Networking concepts.
Whoever says our college curriculum is bad hasn’t either studied anything or hasn’t worked on core engineering roles where these concepts are heavily used in everything you do.
It's a great question. I roughly think of finetuning as analogous to expertise in people:
- Describe a task in words ~= zero-shot prompting
- Give examples of solving task ~= few-shot prompting
- Allow person to practice task ~= finetuning
With this analogy in mind, it's awesome that we have models that can reach high levels of accuracy across many tasks with prompting alone, but I also expect that reaching top tier performance will include finetuning, especially in applications with concrete well-defined tasks where it is possible to collect a lot of data and "practice" on it.
Rough picture to have in mind maybe. Small models are incapable of in-context learning and will benefit very little from prompt engineering, but depending on the difficulty of the task it may be possible to still finetune them into decent experts.
Big caveat all of this is still very new.
Early in my career, a lawyer cost me millions of dollars.
Not in legal fees.
Not by doing a bad job
But by doing their job too well.
Here's a cautionary tale about why it's essential to manage your lawyers – closely.
/🧵
*We've Gone International!* 🇨🇳🇷🇺🇪🇸
Builders are selling their developments to foreign buyers, who have no idea about the #claddingscandal, so our site is now translated. Help spread the word & hurt them in their pocket!
https://t.co/jriDw6EC5a
#EndOurCladdingScandal#BuyerBeware
@ponnappa So true! I clearly remember converting fancy plastic or glass jars into something of use - pen holders, flower pots, decoration items.. We lived this Marathi saying - "kachrya madhun kala" which literally means "art from waste" :)
Amazon launched in July 1995, and every Xmas was a near death experience for the first 7 years. I joined in ‘99 and got to experience this first hand. Starting in late Nov, all corporate employees were shipped to fulfillment centers to pack boxes for 6 weeks. Here’s what I saw:
@Suhail I have been shipping, packaging and re-shipping only ideas and emotions across people in a very nascent stage development. #nocode till you know why and what to code. BTW: Love your #inspiring#motivation.