Update tokenizer files
#2
by
Xenova
HF staff
- opened
- tokenizer.json +10 -1
- tokenizer_config.json +1 -1
tokenizer.json
CHANGED
|
@@ -34,6 +34,10 @@
|
|
| 34 |
"normalizer": {
|
| 35 |
"type": "Sequence",
|
| 36 |
"normalizers": [
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
{
|
| 38 |
"type": "Replace",
|
| 39 |
"pattern": {
|
|
@@ -85,6 +89,12 @@
|
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"type": "Fuse"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
}
|
| 89 |
]
|
| 90 |
},
|
|
@@ -96,7 +106,6 @@
|
|
| 96 |
"end_of_word_suffix": null,
|
| 97 |
"fuse_unk": true,
|
| 98 |
"byte_fallback": true,
|
| 99 |
-
"ignore_merges": false,
|
| 100 |
"vocab": {
|
| 101 |
"<unk>": 0,
|
| 102 |
"<s>": 1,
|
|
|
|
| 34 |
"normalizer": {
|
| 35 |
"type": "Sequence",
|
| 36 |
"normalizers": [
|
| 37 |
+
{
|
| 38 |
+
"type": "Prepend",
|
| 39 |
+
"prepend": "▁"
|
| 40 |
+
},
|
| 41 |
{
|
| 42 |
"type": "Replace",
|
| 43 |
"pattern": {
|
|
|
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"type": "Fuse"
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"type": "Strip",
|
| 95 |
+
"content": " ",
|
| 96 |
+
"start": 1,
|
| 97 |
+
"stop": 0
|
| 98 |
}
|
| 99 |
]
|
| 100 |
},
|
|
|
|
| 106 |
"end_of_word_suffix": null,
|
| 107 |
"fuse_unk": true,
|
| 108 |
"byte_fallback": true,
|
|
|
|
| 109 |
"vocab": {
|
| 110 |
"<unk>": 0,
|
| 111 |
"<s>": 1,
|
tokenizer_config.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"add_bos_token": false,
|
| 3 |
"add_eos_token": false,
|
| 4 |
-
"add_prefix_space":
|
| 5 |
"added_tokens_decoder": {
|
| 6 |
"0": {
|
| 7 |
"content": "<unk>",
|
|
|
|
| 1 |
{
|
| 2 |
"add_bos_token": false,
|
| 3 |
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
"added_tokens_decoder": {
|
| 6 |
"0": {
|
| 7 |
"content": "<unk>",
|