# Vocabulary Trimmed [google/mt5-base](https://huggingface.co/google/mt5-base): `vocabtrimmer/mt5-base-trimmed-ja-30000` 
This model is a trimmed version of [google/mt5-base](https://huggingface.co/google/mt5-base) by [`vocabtrimmer`](https://github.com/asahi417/lm-vocab-trimmer), 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-30000   |
|:---------------------------|:------------------|:-----------------------------------------|
| parameter_size_full        | 582,401,280       | 244,310,784                              |
| parameter_size_embedding   | 384,172,032       | 46,081,536                               |
| vocab_size                 | 250,112           | 30,001                                   |
| compression_rate_full      | 100.0             | 41.95                                    |
| compression_rate_embedding | 100.0             | 12.0                                     |


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      |               30000 |               2 |