Jack Morris
@jxmnop
research @engramlab // language models, information theory, science of AI
Joined October 2016
1.1K Following 53.7K Followers
If you think about it, it's pretty embarrassing that frontier labs still pretrain models. Don't they know there's a theory for this? Given how big the model is, and how long you're going to train, you can predict the loss exactly. No need to actually train. They spend billions on this!
Show more
