Upload tokenizer_config.json with huggingface_hub
Browse files- tokenizer_config.json +26 -0
tokenizer_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {},
|
3 |
+
"additional_special_tokens": [
|
4 |
+
"<|startoftext|>",
|
5 |
+
"<|extra_0|>",
|
6 |
+
"<|extra_4|>",
|
7 |
+
"<|extra_5|>",
|
8 |
+
"<|eos|>"
|
9 |
+
],
|
10 |
+
"architectures": [
|
11 |
+
"GPT2LMHeadModel"
|
12 |
+
],
|
13 |
+
"auto_map": {
|
14 |
+
"AutoTokenizer": [
|
15 |
+
"tokenization_hy.HYTokenizer",
|
16 |
+
null
|
17 |
+
]
|
18 |
+
},
|
19 |
+
"clean_up_tokenization_spaces": false,
|
20 |
+
"eos_token": "<|eos|>",
|
21 |
+
"extra_special_tokens": {},
|
22 |
+
"model_max_length": 1048576,
|
23 |
+
"model_type": "gpt2",
|
24 |
+
"pad_token": "<|pad|>",
|
25 |
+
"tokenizer_class": "HYTokenizer"
|
26 |
+
}
|