Upload tokenizer
Browse files- special_tokens_map.json +5 -12
- tokenizer_config.json +3 -4
special_tokens_map.json
CHANGED
|
@@ -1,42 +1,35 @@
|
|
| 1 |
{
|
| 2 |
"additional_special_tokens": [
|
| 3 |
{
|
| 4 |
-
"content": "<
|
| 5 |
-
"lstrip": false,
|
| 6 |
-
"normalized": false,
|
| 7 |
-
"rstrip": false,
|
| 8 |
-
"single_word": false
|
| 9 |
-
},
|
| 10 |
-
{
|
| 11 |
-
"content": "<BLANK_SPACE>",
|
| 12 |
"lstrip": false,
|
| 13 |
"normalized": false,
|
| 14 |
"rstrip": false,
|
| 15 |
"single_word": false
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"content": "<
|
| 19 |
"lstrip": false,
|
| 20 |
"normalized": false,
|
| 21 |
"rstrip": false,
|
| 22 |
"single_word": false
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"content": "
|
| 26 |
"lstrip": false,
|
| 27 |
"normalized": false,
|
| 28 |
"rstrip": false,
|
| 29 |
"single_word": false
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"content": "<
|
| 33 |
"lstrip": false,
|
| 34 |
"normalized": false,
|
| 35 |
"rstrip": false,
|
| 36 |
"single_word": false
|
| 37 |
},
|
| 38 |
{
|
| 39 |
-
"content": "
|
| 40 |
"lstrip": false,
|
| 41 |
"normalized": false,
|
| 42 |
"rstrip": false,
|
|
|
|
| 1 |
{
|
| 2 |
"additional_special_tokens": [
|
| 3 |
{
|
| 4 |
+
"content": "<SURFACE>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"lstrip": false,
|
| 6 |
"normalized": false,
|
| 7 |
"rstrip": false,
|
| 8 |
"single_word": false
|
| 9 |
},
|
| 10 |
{
|
| 11 |
+
"content": "<COLUMN>",
|
| 12 |
"lstrip": false,
|
| 13 |
"normalized": false,
|
| 14 |
"rstrip": false,
|
| 15 |
"single_word": false
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"content": "<BLANK_SPACE>",
|
| 19 |
"lstrip": false,
|
| 20 |
"normalized": false,
|
| 21 |
"rstrip": false,
|
| 22 |
"single_word": false
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"content": "<RULING>",
|
| 26 |
"lstrip": false,
|
| 27 |
"normalized": false,
|
| 28 |
"rstrip": false,
|
| 29 |
"single_word": false
|
| 30 |
},
|
| 31 |
{
|
| 32 |
+
"content": "\n",
|
| 33 |
"lstrip": false,
|
| 34 |
"normalized": false,
|
| 35 |
"rstrip": false,
|
tokenizer_config.json
CHANGED
|
@@ -90,12 +90,11 @@
|
|
| 90 |
}
|
| 91 |
},
|
| 92 |
"additional_special_tokens": [
|
| 93 |
-
"<
|
|
|
|
| 94 |
"<BLANK_SPACE>",
|
| 95 |
-
"<
|
| 96 |
"\n",
|
| 97 |
-
"<COLUMN>",
|
| 98 |
-
"<SURFACE>",
|
| 99 |
"..."
|
| 100 |
],
|
| 101 |
"bos_token": "<s>",
|
|
|
|
| 90 |
}
|
| 91 |
},
|
| 92 |
"additional_special_tokens": [
|
| 93 |
+
"<SURFACE>",
|
| 94 |
+
"<COLUMN>",
|
| 95 |
"<BLANK_SPACE>",
|
| 96 |
+
"<RULING>",
|
| 97 |
"\n",
|
|
|
|
|
|
|
| 98 |
"..."
|
| 99 |
],
|
| 100 |
"bos_token": "<s>",
|