New hands-on course: implement KV Cache (faster inference) and LoRA (cheap fine-tuning) from scratch on top of @karpathy minGPT. You write the code, benchmark the speedups:
https://t.co/yteUHy4cLH
@ominousEureka Wow, attending NeurIPS is a huge milestone to be proud of :) Congrats!
O feel like sometimes we get so focused on chasing goals that we forget to celebrate what weโre doing.
Be proud of being part of something that cool and have fun!
I want to attend sooner or later ๐ค
Gonna post a bit too about lifestyle and looksmaxxing, since Iโve been working my ass off till I have become pretty decent (will keep improving - gua sha red vaseline etc).
Even though I sometimes wish Iโd stayed sub-3 so I wouldnโt be this distracted by love life as I am ๐
Ok fr this time: back on the TPOT grind.
Switched teams to a more LLM fine-tuning oriented role within the company, grabbed a Masterโs in Education for no reason (always have a backup plan), and got in decent shape (sub-40 10k + 100kg bench).
We ball again. ML updates soon.