Publishing XXS XS S models
#5
by
moussaKam
- opened
Hi there,
Thanks for the great work !
Any plans to publish the smaller versions of the models mentioned in the paper?
All the best
hello ! Yes of course ! I was convinced they were up already sorry !
Will push today !
Actually they have been up for 3 months:
croissantllm/croissant_small_models
Sorry for the late response... Note they only have been trained on about 100B tokens.
manu
changed discussion status to
closed