Commit
•
9d81b01
1
Parent(s):
3b4b214
Upload tokenizer
Browse files- tokenizer.json +12 -12
- tokenizer_config.json +14 -13
tokenizer.json
CHANGED
|
@@ -26,9 +26,9 @@
|
|
| 26 |
"content": "</s>",
|
| 27 |
"single_word": false,
|
| 28 |
"lstrip": false,
|
| 29 |
-
"rstrip":
|
| 30 |
"normalized": false,
|
| 31 |
-
"special":
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"id": 32000,
|
|
@@ -44,7 +44,7 @@
|
|
| 44 |
"content": "<|assistant|>",
|
| 45 |
"single_word": false,
|
| 46 |
"lstrip": false,
|
| 47 |
-
"rstrip":
|
| 48 |
"normalized": false,
|
| 49 |
"special": true
|
| 50 |
},
|
|
@@ -53,7 +53,7 @@
|
|
| 53 |
"content": "<|placeholder1|>",
|
| 54 |
"single_word": false,
|
| 55 |
"lstrip": false,
|
| 56 |
-
"rstrip":
|
| 57 |
"normalized": false,
|
| 58 |
"special": true
|
| 59 |
},
|
|
@@ -62,7 +62,7 @@
|
|
| 62 |
"content": "<|placeholder2|>",
|
| 63 |
"single_word": false,
|
| 64 |
"lstrip": false,
|
| 65 |
-
"rstrip":
|
| 66 |
"normalized": false,
|
| 67 |
"special": true
|
| 68 |
},
|
|
@@ -71,7 +71,7 @@
|
|
| 71 |
"content": "<|placeholder3|>",
|
| 72 |
"single_word": false,
|
| 73 |
"lstrip": false,
|
| 74 |
-
"rstrip":
|
| 75 |
"normalized": false,
|
| 76 |
"special": true
|
| 77 |
},
|
|
@@ -80,7 +80,7 @@
|
|
| 80 |
"content": "<|placeholder4|>",
|
| 81 |
"single_word": false,
|
| 82 |
"lstrip": false,
|
| 83 |
-
"rstrip":
|
| 84 |
"normalized": false,
|
| 85 |
"special": true
|
| 86 |
},
|
|
@@ -89,7 +89,7 @@
|
|
| 89 |
"content": "<|system|>",
|
| 90 |
"single_word": false,
|
| 91 |
"lstrip": false,
|
| 92 |
-
"rstrip":
|
| 93 |
"normalized": false,
|
| 94 |
"special": true
|
| 95 |
},
|
|
@@ -98,7 +98,7 @@
|
|
| 98 |
"content": "<|end|>",
|
| 99 |
"single_word": false,
|
| 100 |
"lstrip": false,
|
| 101 |
-
"rstrip":
|
| 102 |
"normalized": false,
|
| 103 |
"special": true
|
| 104 |
},
|
|
@@ -107,7 +107,7 @@
|
|
| 107 |
"content": "<|placeholder5|>",
|
| 108 |
"single_word": false,
|
| 109 |
"lstrip": false,
|
| 110 |
-
"rstrip":
|
| 111 |
"normalized": false,
|
| 112 |
"special": true
|
| 113 |
},
|
|
@@ -116,7 +116,7 @@
|
|
| 116 |
"content": "<|placeholder6|>",
|
| 117 |
"single_word": false,
|
| 118 |
"lstrip": false,
|
| 119 |
-
"rstrip":
|
| 120 |
"normalized": false,
|
| 121 |
"special": true
|
| 122 |
},
|
|
@@ -125,7 +125,7 @@
|
|
| 125 |
"content": "<|user|>",
|
| 126 |
"single_word": false,
|
| 127 |
"lstrip": false,
|
| 128 |
-
"rstrip":
|
| 129 |
"normalized": false,
|
| 130 |
"special": true
|
| 131 |
}
|
|
|
|
| 26 |
"content": "</s>",
|
| 27 |
"single_word": false,
|
| 28 |
"lstrip": false,
|
| 29 |
+
"rstrip": true,
|
| 30 |
"normalized": false,
|
| 31 |
+
"special": false
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"id": 32000,
|
|
|
|
| 44 |
"content": "<|assistant|>",
|
| 45 |
"single_word": false,
|
| 46 |
"lstrip": false,
|
| 47 |
+
"rstrip": true,
|
| 48 |
"normalized": false,
|
| 49 |
"special": true
|
| 50 |
},
|
|
|
|
| 53 |
"content": "<|placeholder1|>",
|
| 54 |
"single_word": false,
|
| 55 |
"lstrip": false,
|
| 56 |
+
"rstrip": true,
|
| 57 |
"normalized": false,
|
| 58 |
"special": true
|
| 59 |
},
|
|
|
|
| 62 |
"content": "<|placeholder2|>",
|
| 63 |
"single_word": false,
|
| 64 |
"lstrip": false,
|
| 65 |
+
"rstrip": true,
|
| 66 |
"normalized": false,
|
| 67 |
"special": true
|
| 68 |
},
|
|
|
|
| 71 |
"content": "<|placeholder3|>",
|
| 72 |
"single_word": false,
|
| 73 |
"lstrip": false,
|
| 74 |
+
"rstrip": true,
|
| 75 |
"normalized": false,
|
| 76 |
"special": true
|
| 77 |
},
|
|
|
|
| 80 |
"content": "<|placeholder4|>",
|
| 81 |
"single_word": false,
|
| 82 |
"lstrip": false,
|
| 83 |
+
"rstrip": true,
|
| 84 |
"normalized": false,
|
| 85 |
"special": true
|
| 86 |
},
|
|
|
|
| 89 |
"content": "<|system|>",
|
| 90 |
"single_word": false,
|
| 91 |
"lstrip": false,
|
| 92 |
+
"rstrip": true,
|
| 93 |
"normalized": false,
|
| 94 |
"special": true
|
| 95 |
},
|
|
|
|
| 98 |
"content": "<|end|>",
|
| 99 |
"single_word": false,
|
| 100 |
"lstrip": false,
|
| 101 |
+
"rstrip": true,
|
| 102 |
"normalized": false,
|
| 103 |
"special": true
|
| 104 |
},
|
|
|
|
| 107 |
"content": "<|placeholder5|>",
|
| 108 |
"single_word": false,
|
| 109 |
"lstrip": false,
|
| 110 |
+
"rstrip": true,
|
| 111 |
"normalized": false,
|
| 112 |
"special": true
|
| 113 |
},
|
|
|
|
| 116 |
"content": "<|placeholder6|>",
|
| 117 |
"single_word": false,
|
| 118 |
"lstrip": false,
|
| 119 |
+
"rstrip": true,
|
| 120 |
"normalized": false,
|
| 121 |
"special": true
|
| 122 |
},
|
|
|
|
| 125 |
"content": "<|user|>",
|
| 126 |
"single_word": false,
|
| 127 |
"lstrip": false,
|
| 128 |
+
"rstrip": true,
|
| 129 |
"normalized": false,
|
| 130 |
"special": true
|
| 131 |
}
|
tokenizer_config.json
CHANGED
|
@@ -22,9 +22,9 @@
|
|
| 22 |
"content": "</s>",
|
| 23 |
"lstrip": false,
|
| 24 |
"normalized": false,
|
| 25 |
-
"rstrip":
|
| 26 |
"single_word": false,
|
| 27 |
-
"special":
|
| 28 |
},
|
| 29 |
"32000": {
|
| 30 |
"content": "<|endoftext|>",
|
|
@@ -38,7 +38,7 @@
|
|
| 38 |
"content": "<|assistant|>",
|
| 39 |
"lstrip": false,
|
| 40 |
"normalized": false,
|
| 41 |
-
"rstrip":
|
| 42 |
"single_word": false,
|
| 43 |
"special": true
|
| 44 |
},
|
|
@@ -46,7 +46,7 @@
|
|
| 46 |
"content": "<|placeholder1|>",
|
| 47 |
"lstrip": false,
|
| 48 |
"normalized": false,
|
| 49 |
-
"rstrip":
|
| 50 |
"single_word": false,
|
| 51 |
"special": true
|
| 52 |
},
|
|
@@ -54,7 +54,7 @@
|
|
| 54 |
"content": "<|placeholder2|>",
|
| 55 |
"lstrip": false,
|
| 56 |
"normalized": false,
|
| 57 |
-
"rstrip":
|
| 58 |
"single_word": false,
|
| 59 |
"special": true
|
| 60 |
},
|
|
@@ -62,7 +62,7 @@
|
|
| 62 |
"content": "<|placeholder3|>",
|
| 63 |
"lstrip": false,
|
| 64 |
"normalized": false,
|
| 65 |
-
"rstrip":
|
| 66 |
"single_word": false,
|
| 67 |
"special": true
|
| 68 |
},
|
|
@@ -70,7 +70,7 @@
|
|
| 70 |
"content": "<|placeholder4|>",
|
| 71 |
"lstrip": false,
|
| 72 |
"normalized": false,
|
| 73 |
-
"rstrip":
|
| 74 |
"single_word": false,
|
| 75 |
"special": true
|
| 76 |
},
|
|
@@ -78,7 +78,7 @@
|
|
| 78 |
"content": "<|system|>",
|
| 79 |
"lstrip": false,
|
| 80 |
"normalized": false,
|
| 81 |
-
"rstrip":
|
| 82 |
"single_word": false,
|
| 83 |
"special": true
|
| 84 |
},
|
|
@@ -86,7 +86,7 @@
|
|
| 86 |
"content": "<|end|>",
|
| 87 |
"lstrip": false,
|
| 88 |
"normalized": false,
|
| 89 |
-
"rstrip":
|
| 90 |
"single_word": false,
|
| 91 |
"special": true
|
| 92 |
},
|
|
@@ -94,7 +94,7 @@
|
|
| 94 |
"content": "<|placeholder5|>",
|
| 95 |
"lstrip": false,
|
| 96 |
"normalized": false,
|
| 97 |
-
"rstrip":
|
| 98 |
"single_word": false,
|
| 99 |
"special": true
|
| 100 |
},
|
|
@@ -102,7 +102,7 @@
|
|
| 102 |
"content": "<|placeholder6|>",
|
| 103 |
"lstrip": false,
|
| 104 |
"normalized": false,
|
| 105 |
-
"rstrip":
|
| 106 |
"single_word": false,
|
| 107 |
"special": true
|
| 108 |
},
|
|
@@ -110,15 +110,16 @@
|
|
| 110 |
"content": "<|user|>",
|
| 111 |
"lstrip": false,
|
| 112 |
"normalized": false,
|
| 113 |
-
"rstrip":
|
| 114 |
"single_word": false,
|
| 115 |
"special": true
|
| 116 |
}
|
| 117 |
},
|
| 118 |
"bos_token": "<s>",
|
| 119 |
-
"chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role']
|
| 120 |
"clean_up_tokenization_spaces": false,
|
| 121 |
"eos_token": "<|endoftext|>",
|
|
|
|
| 122 |
"model_max_length": 4096,
|
| 123 |
"pad_token": "<|endoftext|>",
|
| 124 |
"padding_side": "left",
|
|
|
|
| 22 |
"content": "</s>",
|
| 23 |
"lstrip": false,
|
| 24 |
"normalized": false,
|
| 25 |
+
"rstrip": true,
|
| 26 |
"single_word": false,
|
| 27 |
+
"special": false
|
| 28 |
},
|
| 29 |
"32000": {
|
| 30 |
"content": "<|endoftext|>",
|
|
|
|
| 38 |
"content": "<|assistant|>",
|
| 39 |
"lstrip": false,
|
| 40 |
"normalized": false,
|
| 41 |
+
"rstrip": true,
|
| 42 |
"single_word": false,
|
| 43 |
"special": true
|
| 44 |
},
|
|
|
|
| 46 |
"content": "<|placeholder1|>",
|
| 47 |
"lstrip": false,
|
| 48 |
"normalized": false,
|
| 49 |
+
"rstrip": true,
|
| 50 |
"single_word": false,
|
| 51 |
"special": true
|
| 52 |
},
|
|
|
|
| 54 |
"content": "<|placeholder2|>",
|
| 55 |
"lstrip": false,
|
| 56 |
"normalized": false,
|
| 57 |
+
"rstrip": true,
|
| 58 |
"single_word": false,
|
| 59 |
"special": true
|
| 60 |
},
|
|
|
|
| 62 |
"content": "<|placeholder3|>",
|
| 63 |
"lstrip": false,
|
| 64 |
"normalized": false,
|
| 65 |
+
"rstrip": true,
|
| 66 |
"single_word": false,
|
| 67 |
"special": true
|
| 68 |
},
|
|
|
|
| 70 |
"content": "<|placeholder4|>",
|
| 71 |
"lstrip": false,
|
| 72 |
"normalized": false,
|
| 73 |
+
"rstrip": true,
|
| 74 |
"single_word": false,
|
| 75 |
"special": true
|
| 76 |
},
|
|
|
|
| 78 |
"content": "<|system|>",
|
| 79 |
"lstrip": false,
|
| 80 |
"normalized": false,
|
| 81 |
+
"rstrip": true,
|
| 82 |
"single_word": false,
|
| 83 |
"special": true
|
| 84 |
},
|
|
|
|
| 86 |
"content": "<|end|>",
|
| 87 |
"lstrip": false,
|
| 88 |
"normalized": false,
|
| 89 |
+
"rstrip": true,
|
| 90 |
"single_word": false,
|
| 91 |
"special": true
|
| 92 |
},
|
|
|
|
| 94 |
"content": "<|placeholder5|>",
|
| 95 |
"lstrip": false,
|
| 96 |
"normalized": false,
|
| 97 |
+
"rstrip": true,
|
| 98 |
"single_word": false,
|
| 99 |
"special": true
|
| 100 |
},
|
|
|
|
| 102 |
"content": "<|placeholder6|>",
|
| 103 |
"lstrip": false,
|
| 104 |
"normalized": false,
|
| 105 |
+
"rstrip": true,
|
| 106 |
"single_word": false,
|
| 107 |
"special": true
|
| 108 |
},
|
|
|
|
| 110 |
"content": "<|user|>",
|
| 111 |
"lstrip": false,
|
| 112 |
"normalized": false,
|
| 113 |
+
"rstrip": true,
|
| 114 |
"single_word": false,
|
| 115 |
"special": true
|
| 116 |
}
|
| 117 |
},
|
| 118 |
"bos_token": "<s>",
|
| 119 |
+
"chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') %}{{'<|user|>' + '\n' + message['content'] + '<|end|>' + '\n' + '<|assistant|>' + '\n'}}{% elif (message['role'] == 'assistant') %}{{message['content'] + '<|end|>' + '\n'}}{% endif %}{% endfor %}",
|
| 120 |
"clean_up_tokenization_spaces": false,
|
| 121 |
"eos_token": "<|endoftext|>",
|
| 122 |
+
"legacy": false,
|
| 123 |
"model_max_length": 4096,
|
| 124 |
"pad_token": "<|endoftext|>",
|
| 125 |
"padding_side": "left",
|