init
Browse files- tokenizer.json +0 -0
- tokenizer_config.json +0 -6
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"add_prefix_space": false,
|
| 3 |
"added_tokens_decoder": {
|
| 4 |
"0": {
|
| 5 |
"content": "[MASK]",
|
|
@@ -228,12 +227,7 @@
|
|
| 228 |
},
|
| 229 |
"clean_up_tokenization_spaces": true,
|
| 230 |
"cls_token": "[CLS]",
|
| 231 |
-
"extra_special_tokens": {},
|
| 232 |
"mask_token": "[MASK]",
|
| 233 |
-
"model_input_names": [
|
| 234 |
-
"input_ids",
|
| 235 |
-
"attention_mask"
|
| 236 |
-
],
|
| 237 |
"model_max_length": 128000,
|
| 238 |
"pad_token": "[PAD]",
|
| 239 |
"sep_token": "[SEP]",
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"added_tokens_decoder": {
|
| 3 |
"0": {
|
| 4 |
"content": "[MASK]",
|
|
|
|
| 227 |
},
|
| 228 |
"clean_up_tokenization_spaces": true,
|
| 229 |
"cls_token": "[CLS]",
|
|
|
|
| 230 |
"mask_token": "[MASK]",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 231 |
"model_max_length": 128000,
|
| 232 |
"pad_token": "[PAD]",
|
| 233 |
"sep_token": "[SEP]",
|