Upload tokenizer
Browse files- tokenizer.json +18 -0
- tokenizer_config.json +17 -1
tokenizer.json
CHANGED
|
@@ -47,6 +47,24 @@
|
|
| 47 |
"rstrip": false,
|
| 48 |
"normalized": true,
|
| 49 |
"special": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
}
|
| 51 |
],
|
| 52 |
"normalizer": {
|
|
|
|
| 47 |
"rstrip": false,
|
| 48 |
"normalized": true,
|
| 49 |
"special": false
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"id": 32002,
|
| 53 |
+
"content": "<pad>",
|
| 54 |
+
"single_word": false,
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"normalized": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"id": 32003,
|
| 62 |
+
"content": "<eot_id>",
|
| 63 |
+
"single_word": false,
|
| 64 |
+
"lstrip": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"normalized": true,
|
| 67 |
+
"special": false
|
| 68 |
}
|
| 69 |
],
|
| 70 |
"normalizer": {
|
tokenizer_config.json
CHANGED
|
@@ -41,10 +41,26 @@
|
|
| 41 |
"rstrip": false,
|
| 42 |
"single_word": false,
|
| 43 |
"special": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
}
|
| 45 |
},
|
| 46 |
"bos_token": "<s>",
|
| 47 |
-
"chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<
|
| 48 |
"clean_up_tokenization_spaces": false,
|
| 49 |
"eos_token": "</s>",
|
| 50 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 41 |
"rstrip": false,
|
| 42 |
"single_word": false,
|
| 43 |
"special": false
|
| 44 |
+
},
|
| 45 |
+
"32002": {
|
| 46 |
+
"content": "<pad>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"32003": {
|
| 54 |
+
"content": "<eot_id>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": true,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": false
|
| 60 |
}
|
| 61 |
},
|
| 62 |
"bos_token": "<s>",
|
| 63 |
+
"chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<start_header_id>' + message['role'] + '<end_header_id>\n\n'+ message['content'] | trim + '<eot_id>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<start_header_id>assistant<end_header_id>\n\n' }}{% endif %}",
|
| 64 |
"clean_up_tokenization_spaces": false,
|
| 65 |
"eos_token": "</s>",
|
| 66 |
"model_max_length": 1000000000000000019884624838656,
|