Fix tokenizer class for direct AutoTokenizer load
Browse files- tokenizer_config.json +18 -18
tokenizer_config.json
CHANGED
|
@@ -1,18 +1,18 @@
|
|
| 1 |
-
{
|
| 2 |
-
"add_prefix_space": true,
|
| 3 |
-
"backend": "tokenizers",
|
| 4 |
-
"bos_token": "<bos>",
|
| 5 |
-
"clean_up_tokenization_spaces": false,
|
| 6 |
-
"eos_token": "<eos>",
|
| 7 |
-
"extra_special_tokens": [
|
| 8 |
-
"<|im_start|>",
|
| 9 |
-
"<|im_end|>"
|
| 10 |
-
],
|
| 11 |
-
"is_local": false,
|
| 12 |
-
"local_files_only": false,
|
| 13 |
-
"model_max_length": 1000000000,
|
| 14 |
-
"pad_token": "<eos>",
|
| 15 |
-
"tokenizer_class": "
|
| 16 |
-
"unk_token": "<unk>",
|
| 17 |
-
"vocab_size": 4096
|
| 18 |
-
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<bos>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<eos>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|im_start|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": false,
|
| 12 |
+
"local_files_only": false,
|
| 13 |
+
"model_max_length": 1000000000,
|
| 14 |
+
"pad_token": "<eos>",
|
| 15 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 16 |
+
"unk_token": "<unk>",
|
| 17 |
+
"vocab_size": 4096
|
| 18 |
+
}
|