feat: add tokenizer_config.json for AutoTokenizer support
Browse files- tokenizer_config.json +31 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"bos_token": "<|bos|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|eos|>",
|
| 7 |
+
"model_max_length": 1048576,
|
| 8 |
+
"pad_token": "<|pad|>",
|
| 9 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 10 |
+
"unk_token": "<|endoftext|>",
|
| 11 |
+
"extra_special_tokens": [
|
| 12 |
+
"<|endoftext|>",
|
| 13 |
+
"<|pad|>",
|
| 14 |
+
"<|bos|>",
|
| 15 |
+
"<|eos|>",
|
| 16 |
+
"<repo>",
|
| 17 |
+
"</repo>",
|
| 18 |
+
"<file>",
|
| 19 |
+
"</file>",
|
| 20 |
+
"<path>",
|
| 21 |
+
"</path>",
|
| 22 |
+
"<think>",
|
| 23 |
+
"</think>",
|
| 24 |
+
"<separator>",
|
| 25 |
+
"</separator>",
|
| 26 |
+
" ",
|
| 27 |
+
" ",
|
| 28 |
+
" ",
|
| 29 |
+
"\t"
|
| 30 |
+
]
|
| 31 |
+
}
|