Slight readme update on the CC100 dataset part

Refined (v5.3) update

Changed the citation

Fixed a typo

Updated the citation section

Clarified a section of the readme

Updated the readme

Fix with the correct model files

Slight readme update

Updated info on CC100

Updated v5.1's training loss

Refined (v5.1) update

Updated the training loss

Readme update for the v5 release

Major (v5) training update

Fixed an erroneous Kazakh section

Slight update to the readme

Slightly clarified a section in the readme

Fixed a typo

Re-added removed model files

Readme update

Pre-v5 update for the tokeniser (training date pushed to the 25th)

Including changes for the upcoming inclusion of validation metrics

Fixed a typo

Updated the readme

Fixed a typo in the readme

Readme and tokeniser update

Changelog update

Added tokeniser-related info to the readme

Fixed a typo

Removed unnecessary imports

Removed NFD and StripAccents from the tokeniser training process

Tokeniser update

Updated the readme with new loss values

Addition of a new tokeniser (pre-v5)

Fourth iteration with 1.9 million training records

Fixed a typo in the readme

Added more info to the "fine-tuning instructions" section

Minor update to DalaT5 v4 info

Minor update to the readme

Slight update to the readme

Pre-v4 readme and support files update

Slight formatting update to the readme

Readme update

Major update with 1.6 million training records

Updated the readme

Updated the readme

Updated the readme

Updated the readme and get_data.sh, and added a requirements file

Updated the readme