The aim is to compress the mT5-large model to leave only the Ukrainian language and some basic English.

Reproduced the similar result (but with another language) from this medium article.

Results:

  • 1.2B params -> 779M params (37%)
  • 250K tokens -> 8900 tokens
  • 4.6GB size model -> 2.9GB size model
Downloads last month
16
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.