We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
should see entropy decrease with ascending pause token, or not, who k… …nows
handle all pause lengths of 0
can save a little compute
allow for generating random uniform pause lengths across sequence len… …gth, and also allow for passing in custom pause lengths
allow for turning off pausing, both during training and inference
add the previous pause token integrator
first complete the basic idea