Add chat template
#1
by emilss - opened
- tokenizer_config.json +23 -1
tokenizer_config.json
CHANGED
|
@@ -8002,13 +8002,35 @@
|
|
| 8002 |
"rstrip": false,
|
| 8003 |
"single_word": false,
|
| 8004 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8005 |
}
|
| 8006 |
},
|
| 8007 |
"bos_token": "<s>",
|
|
|
|
| 8008 |
"clean_up_tokenization_spaces": false,
|
| 8009 |
"eos_token": "</s>",
|
| 8010 |
"extra_special_tokens": {},
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8011 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 8012 |
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 8013 |
"unk_token": "<unk>"
|
| 8014 |
-
}
|
|
|
|
| 8002 |
"rstrip": false,
|
| 8003 |
"single_word": false,
|
| 8004 |
"special": true
|
| 8005 |
+
},
|
| 8006 |
+
"131072": {
|
| 8007 |
+
"content": "<|im_start|>",
|
| 8008 |
+
"lstrip": false,
|
| 8009 |
+
"normalized": false,
|
| 8010 |
+
"rstrip": false,
|
| 8011 |
+
"single_word": false,
|
| 8012 |
+
"special": false
|
| 8013 |
+
},
|
| 8014 |
+
"131073": {
|
| 8015 |
+
"content": "<|im_end|>",
|
| 8016 |
+
"lstrip": false,
|
| 8017 |
+
"normalized": false,
|
| 8018 |
+
"rstrip": false,
|
| 8019 |
+
"single_word": false,
|
| 8020 |
+
"special": false
|
| 8021 |
}
|
| 8022 |
},
|
| 8023 |
"bos_token": "<s>",
|
| 8024 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 8025 |
"clean_up_tokenization_spaces": false,
|
| 8026 |
"eos_token": "</s>",
|
| 8027 |
"extra_special_tokens": {},
|
| 8028 |
+
"model_input_names": [
|
| 8029 |
+
"input_ids",
|
| 8030 |
+
"attention_mask"
|
| 8031 |
+
],
|
| 8032 |
"model_max_length": 1000000000000000019884624838656,
|
| 8033 |
+
"pad_token": "<pad>",
|
| 8034 |
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 8035 |
"unk_token": "<unk>"
|
| 8036 |
+
}
|