Upload tokenizer
Browse files- .gitattributes +1 -0
- tokenizer.json +3 -0
- tokenizer_config.json +25 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
nos_modernBERT_lr5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
nos_modernBERT_lr5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94979129c5972653ae3bc43005b6dd35055a12546d418cd0613590e208f57a5b
|
| 3 |
+
size 37007954
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"extra_special_tokens": [
|
| 8 |
+
"<|translation|>"
|
| 9 |
+
],
|
| 10 |
+
"is_local": true,
|
| 11 |
+
"legacy": true,
|
| 12 |
+
"mask_token": "<mask>",
|
| 13 |
+
"model_input_names": [
|
| 14 |
+
"input_ids",
|
| 15 |
+
"attention_mask"
|
| 16 |
+
],
|
| 17 |
+
"model_max_length": 1024,
|
| 18 |
+
"pad_token": "<pad>",
|
| 19 |
+
"padding_side": "right",
|
| 20 |
+
"sp_model_kwargs": {},
|
| 21 |
+
"spaces_between_special_tokens": false,
|
| 22 |
+
"tokenizer_class": "TokenizersBackend",
|
| 23 |
+
"unk_token": "<unk>",
|
| 24 |
+
"use_default_system_prompt": false
|
| 25 |
+
}
|