Vocabulary Trimmed facebook/mbart-large-cc25: vocabtrimmer/mbart-large-cc25-trimmed-ru
This model is a trimmed version of facebook/mbart-large-cc25 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/mbart-large-cc25 | vocabtrimmer/mbart-large-cc25-trimmed-ru | |
|---|---|---|
| parameter_size_full | 610,851,840 | 456,877,056 |
| parameter_size_embedding | 512,055,296 | 204,105,728 |
| vocab_size | 250,027 | 99,661 |
| compression_rate_full | 100.0 | 74.79 |
| compression_rate_embedding | 100.0 | 39.86 |
Following table shows the parameter used to trim vocabulary.
| language | dataset | dataset_column | dataset_name | dataset_split | target_vocab_size | min_frequency |
|---|---|---|---|---|---|---|
| ru | vocabtrimmer/mc4_validation | text | ru | validation | 2 |
