Update tokenizer_config.json
Browse files- tokenizer_config.json +3 -3
tokenizer_config.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"add_prefix_space": false,
|
| 3 |
-
"bos_token": "
|
| 4 |
"clean_up_tokenization_spaces": true,
|
| 5 |
-
"eos_token": "
|
| 6 |
-
"model_max_length":
|
| 7 |
"padding_side": "right",
|
| 8 |
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
"unk_token": "<|endoftext|>"
|
|
|
|
| 1 |
{
|
| 2 |
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "[CLS]",
|
| 4 |
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 4096,
|
| 7 |
"padding_side": "right",
|
| 8 |
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
"unk_token": "<|endoftext|>"
|