Upload 3 files
Browse files- tokenizer.json +8 -8
- tokenizer_config.json +11 -0
tokenizer.json
CHANGED
|
@@ -5,7 +5,7 @@
|
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
| 7 |
"id": 0,
|
| 8 |
-
"content": "<
|
| 9 |
"single_word": false,
|
| 10 |
"lstrip": false,
|
| 11 |
"rstrip": false,
|
|
@@ -14,7 +14,7 @@
|
|
| 14 |
},
|
| 15 |
{
|
| 16 |
"id": 1,
|
| 17 |
-
"content": "<
|
| 18 |
"single_word": false,
|
| 19 |
"lstrip": false,
|
| 20 |
"rstrip": false,
|
|
@@ -23,7 +23,7 @@
|
|
| 23 |
},
|
| 24 |
{
|
| 25 |
"id": 2,
|
| 26 |
-
"content": "<pad>",
|
| 27 |
"single_word": false,
|
| 28 |
"lstrip": false,
|
| 29 |
"rstrip": false,
|
|
@@ -32,7 +32,7 @@
|
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"id": 99900,
|
| 35 |
-
"content": "<
|
| 36 |
"single_word": false,
|
| 37 |
"lstrip": false,
|
| 38 |
"rstrip": false,
|
|
@@ -41,7 +41,7 @@
|
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"id": 99901,
|
| 44 |
-
"content": "<
|
| 45 |
"single_word": false,
|
| 46 |
"lstrip": false,
|
| 47 |
"rstrip": false,
|
|
@@ -1006,9 +1006,9 @@
|
|
| 1006 |
"byte_fallback": false,
|
| 1007 |
"ignore_merges": false,
|
| 1008 |
"vocab": {
|
| 1009 |
-
"<
|
| 1010 |
-
"<
|
| 1011 |
-
"<pad>": 2,
|
| 1012 |
"!": 3,
|
| 1013 |
"\"": 4,
|
| 1014 |
"#": 5,
|
|
|
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
| 7 |
"id": 0,
|
| 8 |
+
"content": "<|begin▁of▁sentence|>",
|
| 9 |
"single_word": false,
|
| 10 |
"lstrip": false,
|
| 11 |
"rstrip": false,
|
|
|
|
| 14 |
},
|
| 15 |
{
|
| 16 |
"id": 1,
|
| 17 |
+
"content": "<|end▁of▁sentence|>",
|
| 18 |
"single_word": false,
|
| 19 |
"lstrip": false,
|
| 20 |
"rstrip": false,
|
|
|
|
| 23 |
},
|
| 24 |
{
|
| 25 |
"id": 2,
|
| 26 |
+
"content": "<|▁pad▁|>",
|
| 27 |
"single_word": false,
|
| 28 |
"lstrip": false,
|
| 29 |
"rstrip": false,
|
|
|
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"id": 99900,
|
| 35 |
+
"content": "<|im▁start|>",
|
| 36 |
"single_word": false,
|
| 37 |
"lstrip": false,
|
| 38 |
"rstrip": false,
|
|
|
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"id": 99901,
|
| 44 |
+
"content": "<|im▁end|>",
|
| 45 |
"single_word": false,
|
| 46 |
"lstrip": false,
|
| 47 |
"rstrip": false,
|
|
|
|
| 1006 |
"byte_fallback": false,
|
| 1007 |
"ignore_merges": false,
|
| 1008 |
"vocab": {
|
| 1009 |
+
"<|begin▁of▁sentence|>": 0,
|
| 1010 |
+
"<|end▁of▁sentence|>": 1,
|
| 1011 |
+
"<|▁pad▁|>": 2,
|
| 1012 |
"!": 3,
|
| 1013 |
"\"": 4,
|
| 1014 |
"#": 5,
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"special_tokens": {
|
| 3 |
+
"bos": "<|begin▁of▁sentence|>",
|
| 4 |
+
"eos": "<|end▁of▁sentence|>",
|
| 5 |
+
"pad": "<|▁pad▁|>",
|
| 6 |
+
"im_start": "<|im▁start|>",
|
| 7 |
+
"im_end": "<|im▁end|>"
|
| 8 |
+
},
|
| 9 |
+
"chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}<|im▁start|>system\n{{ message['content'] }}<|im▁end|>\n{% elif message['role'] == 'user' %}<|im▁start|>user\n{{ message['content'] }}<|im▁end|>\n{% elif message['role'] == 'assistant' %}<|im▁start|>assistant\n{{ message['content'] }}<|im▁end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im▁start|>assistant\n{% endif %}"
|
| 10 |
+
|
| 11 |
+
}
|