Arabic
arabic
tokenizer
morphology
nlp
dialect
File size: 255 Bytes
3b90e9e
ea23987
 
 
 
 
 
 
 
 
3b90e9e
1
2
3
4
5
6
7
8
9
10
11
{
  "tokenizer_class": "LlamaTokenizer",
  "model_max_length": 32768,
  "clean_up_tokenization_spaces": false,
  "unk_token": "[UNK]",
  "bos_token": "<s>",
  "eos_token": "</s>",
  "pad_token": "[PAD]",
  "add_bos_token": true,
  "add_eos_token": false
}