Revert "test: remove tokenizer_config.json"
Browse filesThis reverts commit 2934fb7134ced003ecb34ea8246c61bf92850481.
- tokenizer_config.json +14 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"unk_token": "<unk>",
|
| 5 |
+
"sep_token": "</s>",
|
| 6 |
+
"cls_token": "<s>",
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"mask_token": "<mask>",
|
| 9 |
+
"additional_special_tokens": [
|
| 10 |
+
"<s>NOTUSED",
|
| 11 |
+
"</s>NOTUSED",
|
| 12 |
+
"▁"
|
| 13 |
+
]
|
| 14 |
+
}
|