Upload tokenizer_config.json
Browse files- tokenizer_config.json +21 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"[STREAMING_PAD]",
|
| 10 |
+
"[STREAMING_WORD]",
|
| 11 |
+
"[LANGUAGE_ZH]",
|
| 12 |
+
"[LANGUAGE_EN]"
|
| 13 |
+
],
|
| 14 |
+
"is_local": true,
|
| 15 |
+
"local_files_only": false,
|
| 16 |
+
"model_max_length": 131072,
|
| 17 |
+
"pad_token": "<|endoftext|>",
|
| 18 |
+
"split_special_tokens": false,
|
| 19 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 20 |
+
"unk_token": null
|
| 21 |
+
}
|