Tokenizer config
#5
by Compactbot - opened
- tokenizer_config.json +9 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "[[ctrl:endoftext]]",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "[[ctrl:endoftext]]",
|
| 6 |
+
"model_max_length": 512,
|
| 7 |
+
"pad_token": "[[ctrl:endoftext]]",
|
| 8 |
+
"tokenizer_class": "TokenizersBackend"
|
| 9 |
+
}
|