Vocabulary Trimmed google/mt5-base: vocabtrimmer/mt5-base-trimmed-ja-45000

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-ja-45000
parameter_size_full	582,401,280	267,350,784
parameter_size_embedding	384,172,032	69,121,536
vocab_size	250,112	45,001
compression_rate_full	100.0	45.9
compression_rate_embedding	100.0	17.99

Following table shows the parameter used to trim vocabulary.

language	dataset	dataset_column	dataset_name	dataset_split	target_vocab_size	min_frequency
ja	vocabtrimmer/mc4_validation	text	ja	validation	45000	2