Staff Engineer @ Google + Recommendation Systems Professor @ Columbia University

New York, USA
Joined November 2017
Wtf? What is this unknown limit @OpenAIDevs
1
1
254
Replying to @rohindhar
Definitely getting interesting but doesn’t seem insane yet
67
Replying to @karpathy
Why 25% and not 20% (1/5th in training are 0)?
2
822
Built a ChatGPT plugin for personalized finance AI
4
5
490
GPT4 scored 86 percentile on march madness bracket, better than 8/10 of your friends/colleagues
1
4
360
Replying to @dylanhendricks
If you ask it to use wolfram plugin it can give you the hash. Which commits it by having it add the answer to the context
1
6
526
Cool to use Wolfram + GPT4 for recipes
150
Replying to @OpenAI
@OpenAI plug-ins seem to work much better if you only do 1 instead of 3 plug-ins active. Are Instacart/Expedia paying for each time their plug-in is used? Or some monthly fee? Or only when an actual order is placed? Is this going to be the main monetization strategy?
39
Replying to @OpenAI
@OpenAI #Support #GPT4 Plugins have a bunch of bugs. OpenTable, Wolfram and Zapier work consistently. + some new bugs if you switch chats before getting a full response back. Looking forward to writing my own plugins
94
Replying to @finbarrtimbers
just tried (screenshot) which works for small set of numbers, but fails once you get above a couple thousand numbers
12
#GPT4 goes 12/16 in #MarchMadness, 98th percentile after day 1 xitter.cf/pythonprimes/status/16…
Had Sydney/#ChatGPT/#gpt4 create a #MarchMadness bracket: @ZagMBB you are the AI's favorite, good luck. Included are reasons why it thinks Gonzaga will beat @UA_Athletics. CC @MarchMadnessMBB play.ncaa.com/mens-bracket-c…
158
Sometimes cooperative, sometimes not, as well as different formats. And yes I did the entire bracket manually
97