Update tokenizer_config.json
Browse filesUse correct bos token that the model was trained on
- tokenizer_config.json +2 -2
tokenizer_config.json
CHANGED
@@ -2051,9 +2051,9 @@
|
|
2051 |
"special": true
|
2052 |
}
|
2053 |
},
|
2054 |
-
"bos_token": "
|
2055 |
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
2056 |
-
"clean_up_tokenization_spaces":
|
2057 |
"eos_token": "<|im_end|>",
|
2058 |
"legacy": false,
|
2059 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
2051 |
"special": true
|
2052 |
}
|
2053 |
},
|
2054 |
+
"bos_token": "<|im_start|>",
|
2055 |
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
2056 |
+
"clean_up_tokenization_spaces": true,
|
2057 |
"eos_token": "<|im_end|>",
|
2058 |
"legacy": false,
|
2059 |
"model_max_length": 1000000000000000019884624838656,
|