elliotthwang
commited on
Commit
·
3de9a1f
1
Parent(s):
34203d3
Upload tokenizer
Browse files- tokenizer.json +2 -11
- tokenizer_config.json +6 -3
tokenizer.json
CHANGED
@@ -9,7 +9,7 @@
|
|
9 |
"single_word": false,
|
10 |
"lstrip": false,
|
11 |
"rstrip": false,
|
12 |
-
"normalized":
|
13 |
"special": true
|
14 |
},
|
15 |
{
|
@@ -27,17 +27,8 @@
|
|
27 |
"single_word": false,
|
28 |
"lstrip": false,
|
29 |
"rstrip": false,
|
30 |
-
"normalized":
|
31 |
"special": true
|
32 |
-
},
|
33 |
-
{
|
34 |
-
"id": 32000,
|
35 |
-
"content": "<pad>",
|
36 |
-
"single_word": false,
|
37 |
-
"lstrip": false,
|
38 |
-
"rstrip": false,
|
39 |
-
"normalized": true,
|
40 |
-
"special": false
|
41 |
}
|
42 |
],
|
43 |
"normalizer": {
|
|
|
9 |
"single_word": false,
|
10 |
"lstrip": false,
|
11 |
"rstrip": false,
|
12 |
+
"normalized": true,
|
13 |
"special": true
|
14 |
},
|
15 |
{
|
|
|
27 |
"single_word": false,
|
28 |
"lstrip": false,
|
29 |
"rstrip": false,
|
30 |
+
"normalized": false,
|
31 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
}
|
33 |
],
|
34 |
"normalizer": {
|
tokenizer_config.json
CHANGED
@@ -16,10 +16,12 @@
|
|
16 |
"rstrip": false,
|
17 |
"single_word": false
|
18 |
},
|
19 |
-
"legacy":
|
20 |
-
"model_max_length":
|
21 |
"pad_token": null,
|
|
|
22 |
"sp_model_kwargs": {},
|
|
|
23 |
"tokenizer_class": "LlamaTokenizer",
|
24 |
"unk_token": {
|
25 |
"__type": "AddedToken",
|
@@ -28,5 +30,6 @@
|
|
28 |
"normalized": true,
|
29 |
"rstrip": false,
|
30 |
"single_word": false
|
31 |
-
}
|
|
|
32 |
}
|
|
|
16 |
"rstrip": false,
|
17 |
"single_word": false
|
18 |
},
|
19 |
+
"legacy": null,
|
20 |
+
"model_max_length": 4096,
|
21 |
"pad_token": null,
|
22 |
+
"padding_side": "right",
|
23 |
"sp_model_kwargs": {},
|
24 |
+
"spaces_between_special_tokens": false,
|
25 |
"tokenizer_class": "LlamaTokenizer",
|
26 |
"unk_token": {
|
27 |
"__type": "AddedToken",
|
|
|
30 |
"normalized": true,
|
31 |
"rstrip": false,
|
32 |
"single_word": false
|
33 |
+
},
|
34 |
+
"use_default_system_prompt": true
|
35 |
}
|