Upload tokenizer
Browse files- special_tokens_map.json +2 -2
- tokenizer_config.json +3 -3
special_tokens_map.json
CHANGED
@@ -7,11 +7,11 @@
|
|
7 |
"single_word": false
|
8 |
},
|
9 |
"eos_token": {
|
10 |
-
"content": "<|
|
11 |
"lstrip": false,
|
12 |
"normalized": false,
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
-
"pad_token": "<|
|
17 |
}
|
|
|
7 |
"single_word": false
|
8 |
},
|
9 |
"eos_token": {
|
10 |
+
"content": "<|eot_id|>",
|
11 |
"lstrip": false,
|
12 |
"normalized": false,
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
+
"pad_token": "<|eot_id|>"
|
17 |
}
|
tokenizer_config.json
CHANGED
@@ -2050,15 +2050,15 @@
|
|
2050 |
}
|
2051 |
},
|
2052 |
"bos_token": "<|begin_of_text|>",
|
2053 |
-
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\nAssistant:' }}{% elif message['role'] == 'assistant' %}{{ content + '<|
|
2054 |
"clean_up_tokenization_spaces": true,
|
2055 |
-
"eos_token": "<|
|
2056 |
"model_input_names": [
|
2057 |
"input_ids",
|
2058 |
"attention_mask"
|
2059 |
],
|
2060 |
"model_max_length": 131072,
|
2061 |
-
"pad_token": "<|
|
2062 |
"padding_side": "left",
|
2063 |
"split_special_tokens": false,
|
2064 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
|
|
2050 |
}
|
2051 |
},
|
2052 |
"bos_token": "<|begin_of_text|>",
|
2053 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\nAssistant:' }}{% elif message['role'] == 'assistant' %}{{ content + '<|eot_id|>' + '\n' }}{% endif %}{% endfor %}",
|
2054 |
"clean_up_tokenization_spaces": true,
|
2055 |
+
"eos_token": "<|eot_id|>",
|
2056 |
"model_input_names": [
|
2057 |
"input_ids",
|
2058 |
"attention_mask"
|
2059 |
],
|
2060 |
"model_max_length": 131072,
|
2061 |
+
"pad_token": "<|eot_id|>",
|
2062 |
"padding_side": "left",
|
2063 |
"split_special_tokens": false,
|
2064 |
"tokenizer_class": "PreTrainedTokenizerFast"
|