Vocabulary Trimmed facebook/xlm-v-base: vocabtrimmer/xlm-v-base-trimmed-de

This model is a trimmed version of facebook/xlm-v-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.

facebook/xlm-v-base vocabtrimmer/xlm-v-base-trimmed-de
parameter_size_full 779,396,349 270,057,658
parameter_size_embedding 692,451,072 183,774,720
vocab_size 901,629 239,290
compression_rate_full 100.0 34.65
compression_rate_embedding 100.0 26.54

Following table shows the parameter used to trim vocabulary.

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