i've seen other posts about whichever model is powering google search, but it really do be acting like that...
Very interesting model to play with!
New work with @AlecRad and @DavidDuvenaud:
Have you ever dreamed of talking to someone from the past? Introducing talkie, a 13B model trained only on pre-1931 text.
Vintage models should help us to understand how LMs generalize (e.g., can we teach talkie to code?). Thread:
Went back to check after this blew up and it's full today!
Barista says barely had time to breath or do the dishes
Also agent ordered ~1300 cherry tomatoes
xitter.cf/jlagerros/status/20469…
