Upload tokenizer_config.json with huggingface_hub
Browse files- tokenizer_config.json +16 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "sentencepiece",
|
| 3 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 4 |
+
"vocab_size": 8000,
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"bos_token": "<s>",
|
| 7 |
+
"eos_token": "</s>",
|
| 8 |
+
"unk_token": "<unk>",
|
| 9 |
+
"pad_token": "<pad>",
|
| 10 |
+
"sp_model_kwargs": {},
|
| 11 |
+
"add_bos_token": false,
|
| 12 |
+
"add_eos_token": false,
|
| 13 |
+
"clean_up_tokenization_spaces": true,
|
| 14 |
+
"legacy": true,
|
| 15 |
+
"name_or_path": "khopilot/khmer-tokenizer-v7"
|
| 16 |
+
}
|