Javier de la Rosa
Javier de la Rosa
Interesting. Does this work with the Blocks API as well?
OK, I finally had the chance to try this but now the lines are not breaking at the length of their container. Any idea why? 
It's been over a year and now HF supports GPG-signed commits. Is there any update or extra thought on this? Is handling separately HF snapshot download and Gensim load still...
Thanks for finding the key line, @liaeh! As I see it then, we only have two options: 1. We re-label our datasets if not IOB-style to start each label with...
Did you find any good solution to the misalignments?
I see. Are you using any other solution that provides satisfactory results?
Any chance of getting this merged?
Can I ask what kind of TPU are you using for the test, @borisdayma? I do have available a v4-32 that I'd like to use to do continue pre-training on...
Any luck training the 30B on a single TPU v3-8 so far? Does it even fit? The 7B needs 84GB of VRAM, so I would expect the 30B to need...