Upload tokenizer
Browse files- added_tokens.json +2 -1
- special_tokens_map.json +7 -0
- tokenizer.json +9 -0
- tokenizer_config.json +10 -1
added_tokens.json
CHANGED
@@ -1,3 +1,4 @@
|
|
1 |
{
|
2 |
-
"<image>": 32000
|
|
|
3 |
}
|
|
|
1 |
{
|
2 |
+
"<image>": 32000,
|
3 |
+
"<pad>": 32001
|
4 |
}
|
special_tokens_map.json
CHANGED
@@ -13,6 +13,13 @@
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
"unk_token": {
|
17 |
"content": "<unk>",
|
18 |
"lstrip": false,
|
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "<pad>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
"unk_token": {
|
24 |
"content": "<unk>",
|
25 |
"lstrip": false,
|
tokenizer.json
CHANGED
@@ -38,6 +38,15 @@
|
|
38 |
"rstrip": false,
|
39 |
"normalized": false,
|
40 |
"special": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
41 |
}
|
42 |
],
|
43 |
"normalizer": {
|
|
|
38 |
"rstrip": false,
|
39 |
"normalized": false,
|
40 |
"special": false
|
41 |
+
},
|
42 |
+
{
|
43 |
+
"id": 32001,
|
44 |
+
"content": "<pad>",
|
45 |
+
"single_word": false,
|
46 |
+
"lstrip": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"special": true
|
50 |
}
|
51 |
],
|
52 |
"normalizer": {
|
tokenizer_config.json
CHANGED
@@ -32,6 +32,14 @@
|
|
32 |
"normalized": false,
|
33 |
"rstrip": false,
|
34 |
"single_word": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
"special": true
|
36 |
}
|
37 |
},
|
@@ -41,11 +49,12 @@
|
|
41 |
"eos_token": "</s>",
|
42 |
"legacy": true,
|
43 |
"model_max_length": 1000000000000000019884624838656,
|
44 |
-
"pad_token":
|
45 |
"processor_class": "LlavaProcessor",
|
46 |
"sp_model_kwargs": {},
|
47 |
"spaces_between_special_tokens": false,
|
48 |
"tokenizer_class": "LlamaTokenizer",
|
|
|
49 |
"unk_token": "<unk>",
|
50 |
"use_default_system_prompt": true
|
51 |
}
|
|
|
32 |
"normalized": false,
|
33 |
"rstrip": false,
|
34 |
"single_word": false,
|
35 |
+
"special": false
|
36 |
+
},
|
37 |
+
"32001": {
|
38 |
+
"content": "<pad>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
"special": true
|
44 |
}
|
45 |
},
|
|
|
49 |
"eos_token": "</s>",
|
50 |
"legacy": true,
|
51 |
"model_max_length": 1000000000000000019884624838656,
|
52 |
+
"pad_token": "<pad>",
|
53 |
"processor_class": "LlavaProcessor",
|
54 |
"sp_model_kwargs": {},
|
55 |
"spaces_between_special_tokens": false,
|
56 |
"tokenizer_class": "LlamaTokenizer",
|
57 |
+
"trust_remote_code": false,
|
58 |
"unk_token": "<unk>",
|
59 |
"use_default_system_prompt": true
|
60 |
}
|