Vocabulary Trimmed google/mt5-base: vocabtrimmer/mt5-base-trimmed-fr

This model is a trimmed version of google/mt5-base by vocabtrimmer, a tool for trimming vocabulary of language models to compress the model size. Following table shows a summary of the trimming process.

google/mt5-base vocabtrimmer/mt5-base-trimmed-fr
parameter_size_full 582,401,280 399,577,344
parameter_size_embedding 384,172,032 201,348,096
vocab_size 250,112 131,086
compression_rate_full 100.0 68.61
compression_rate_embedding 100.0 52.41

Following table shows the parameter used to trim vocabulary.

language dataset dataset_column dataset_name dataset_split target_vocab_size min_frequency
fr vocabtrimmer/mc4_validation text fr validation 2