metadata

language:
  - sl
license: cc-by-sa-4.0

[legacy] t5-sl-small

This is the first version of the t5-sl-small model, which has since been replaced by an updated model (cjvt/t5-sl-small). The architecture of the two models is the same, but the legacy version was trained for about 6 times less (i.e. the model has seen 6 times less data during the training).

This version remains here due to reproducibility reasons.

Corpora

The following corpora were used for training the model:

Gigafida 2.0
Kas 1.0
Janes 1.0 (only Janes-news, Janes-forum, Janes-blog, Janes-wiki subcorpora)
Slovenian parliamentary corpus siParl 2.0
slWaC