Commit History
tokenizer: fix number of extra ids
ad60d7b
stefan-it
commited on
readme: add some interesting details about umT5
0685c7d
stefan-it
commited on
readme: mention conversion script
a67d2ab
stefan-it
commited on
readme: add initial version
19d7d84
stefan-it
commited on
model: add latest PyTorch version
f8d09e1
stefan-it
commited on
tokenizer: bump number of extra ids
82caf5f
stefan-it
commited on
tokenizer: add config
e166915
stefan-it
commited on
tokenizer: add special tokens map (imported from mT5)
347c39d
stefan-it
commited on
model: fix copy paste bug in conversion script!!
6f56448
stefan-it
commited on
readme: currently do not maintain PyTorch weights (but they are coming back!)
5e438c9
stefan-it
commited on
model: update latest version
cf6ff25
stefan-it
commited on
config: adjust architecture
97679e8
stefan-it
commited on
config: prepare for upcoming umT5 integration
dad4406
stefan-it
commited on
model: revert transpose change
533e5ef
stefan-it
commited on
model: transpose changes
3849401
stefan-it
commited on
config: adjust vocab size (slightly larger than mT5)
9e5b4bd
stefan-it
commited on
model: add converted PyTorch model
c93fb7e
stefan-it
commited on
vocab: add original spm model
488bf51
stefan-it
commited on
model: add converted model
bc9fe7c
stefan-it
commited on
initial commit
cba65cb
stefan-it
commited on