lg-en / tokenizer_config.json
Conrad747's picture
End of training
2ea5d5d
raw
history blame
346 Bytes
{
"eos_token": "</s>",
"model_max_length": 512,
"name_or_path": "AI-Lab-Makerere/lg_en",
"pad_token": "<pad>",
"separate_vocabs": false,
"source_lang": "mul",
"sp_model_kwargs": {},
"special_tokens_map_file": null,
"target_lang": "eng",
"tokenizer_class": "MarianTokenizer",
"tokenizer_file": null,
"unk_token": "<unk>"
}