emasters

joined 1 year ago
 

Long Sequence Modeling with XGen: A 7B LLM Trained on 8K Input Sequence Length -- https://blog.salesforceairesearch.com/xgen/

[–] emasters@sh.itjust.works 1 points 1 year ago

Started with Slackware back in 1993. First issue was convincing my boss I needed a couple dozen 3-1/2 inch floppies. Next was compiling the kernel with support for my network and video cards. Good times!

These days it's pretty much Ubuntu everywhere and all the time from our cloud systems to the deep learning workstation I built last month.

I don't miss compiling my own kernels.