Vocabulary Trimmed google/mt5-base: vocabtrimmer/mt5-base-trimmed-it-105000

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-it-105000
parameter_size_full 582,401,280 359,510,784
parameter_size_embedding 384,172,032 161,281,536
vocab_size 250,112 105,001
compression_rate_full 100.0 61.73
compression_rate_embedding 100.0 41.98

Following table shows the parameter used to trim vocabulary.

language dataset dataset_column dataset_name dataset_split target_vocab_size min_frequency
it vocabtrimmer/mc4_validation text it validation 105000 2