AmharicRoBERTa / tokenizer_config.json
israel's picture
Upload 7 files
2e7fa7c
raw
history blame
382 Bytes
{
"cls_token": "[CLS]",
"do_lower_case": true,
"mask_token": "[MASK]",
"name_or_path": "israel/AmhWordPieceTokenizer",
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": "israel/AmhWordPieceTokenizer/special_tokens_map.json",
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "BertTokenizer",
"unk_token": "[UNK]"
}