@haldaume3 @yoavgo @karpathy seems like that should make it much harder to learn long-distance functions, e.g. “forget” gate.
@haldaume3 @yoavgo @karpathy seems like that shoul…
Posted
in
by
Tags:
@haldaume3 @yoavgo @karpathy seems like that should make it much harder to learn long-distance functions, e.g. “forget” gate.
Posted
in
by
Tags: