Upload tokenizer
Browse files- special_tokens_map.json +63 -9
- tokenizer_config.json +0 -1
special_tokens_map.json
CHANGED
|
@@ -13,15 +13,69 @@
|
|
| 13 |
"<|vision_pad|>",
|
| 14 |
"<|image_pad|>",
|
| 15 |
"<|video_pad|>",
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
],
|
| 26 |
"eos_token": {
|
| 27 |
"content": "<|im_end|>",
|
|
|
|
| 13 |
"<|vision_pad|>",
|
| 14 |
"<|image_pad|>",
|
| 15 |
"<|video_pad|>",
|
| 16 |
+
{
|
| 17 |
+
"content": "<img>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"content": "</img>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"content": "<IMG_CONTEXT>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"content": "<quad>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"content": "</quad>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"content": "<ref>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"content": "</ref>",
|
| 60 |
+
"lstrip": false,
|
| 61 |
+
"normalized": false,
|
| 62 |
+
"rstrip": false,
|
| 63 |
+
"single_word": false
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"content": "<box>",
|
| 67 |
+
"lstrip": false,
|
| 68 |
+
"normalized": false,
|
| 69 |
+
"rstrip": false,
|
| 70 |
+
"single_word": false
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"content": "</box>",
|
| 74 |
+
"lstrip": false,
|
| 75 |
+
"normalized": false,
|
| 76 |
+
"rstrip": false,
|
| 77 |
+
"single_word": false
|
| 78 |
+
}
|
| 79 |
],
|
| 80 |
"eos_token": {
|
| 81 |
"content": "<|im_end|>",
|
tokenizer_config.json
CHANGED
|
@@ -282,7 +282,6 @@
|
|
| 282 |
"extra_special_tokens": {},
|
| 283 |
"model_max_length": 8192,
|
| 284 |
"pad_token": "<|endoftext|>",
|
| 285 |
-
"processor_class": "InternVLProcessor",
|
| 286 |
"return_token_type_ids": false,
|
| 287 |
"split_special_tokens": false,
|
| 288 |
"tokenizer_class": "Qwen2Tokenizer",
|
|
|
|
| 282 |
"extra_special_tokens": {},
|
| 283 |
"model_max_length": 8192,
|
| 284 |
"pad_token": "<|endoftext|>",
|
|
|
|
| 285 |
"return_token_type_ids": false,
|
| 286 |
"split_special_tokens": false,
|
| 287 |
"tokenizer_class": "Qwen2Tokenizer",
|