Commit
•
b84d66a
1
Parent(s):
6a8f443
Update chat template (#2)
Browse files- Update chat template (3ffc11ecd4fb0fee657c59c3f090fec046b13928)
Co-authored-by: Eduardo Garcia <eduagarcia@users.noreply.huggingface.co>
- tokenizer_config.json +59 -2
tokenizer_config.json
CHANGED
|
@@ -25,16 +25,73 @@
|
|
| 25 |
"rstrip": false,
|
| 26 |
"single_word": false,
|
| 27 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
}
|
| 29 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"auto_map": {
|
| 31 |
"AutoTokenizer": [
|
| 32 |
-
"
|
| 33 |
null
|
| 34 |
]
|
| 35 |
},
|
| 36 |
"bos_token": "<s>",
|
| 37 |
-
"clean_up_tokenization_spaces":
|
|
|
|
| 38 |
"eos_token": "</s>",
|
| 39 |
"legacy": true,
|
| 40 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 25 |
"rstrip": false,
|
| 26 |
"single_word": false,
|
| 27 |
"special": true
|
| 28 |
+
},
|
| 29 |
+
"92543": {
|
| 30 |
+
"content": "<|im_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"92542": {
|
| 38 |
+
"content": "<|im_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"92541": {
|
| 46 |
+
"content": "<|action_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"92540": {
|
| 54 |
+
"content": "<|action_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"92539": {
|
| 62 |
+
"content": "<|interpreter|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"92538": {
|
| 70 |
+
"content": "<|plugin|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
}
|
| 77 |
},
|
| 78 |
+
"additional_special_tokens": [
|
| 79 |
+
"<|im_start|>",
|
| 80 |
+
"<|im_end|>",
|
| 81 |
+
"<|action_start|>",
|
| 82 |
+
"<|action_end|>",
|
| 83 |
+
"<|interpreter|>",
|
| 84 |
+
"<|plugin|>"
|
| 85 |
+
],
|
| 86 |
"auto_map": {
|
| 87 |
"AutoTokenizer": [
|
| 88 |
+
"internlm/internlm2-chat-7b--tokenization_internlm2.InternLM2Tokenizer",
|
| 89 |
null
|
| 90 |
]
|
| 91 |
},
|
| 92 |
"bos_token": "<s>",
|
| 93 |
+
"clean_up_tokenization_spaces": false,
|
| 94 |
+
"chat_template": "{{ bos_token }}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 95 |
"eos_token": "</s>",
|
| 96 |
"legacy": true,
|
| 97 |
"model_max_length": 1000000000000000019884624838656,
|