Upload tokenizer
Browse files- added_tokens.json +3 -0
- special_tokens_map.json +4 -1
- tokenizer.json +27 -0
added_tokens.json
CHANGED
|
@@ -1,6 +1,9 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
| 2 |
"<END_A>": 32003,
|
| 3 |
"<END_Q>": 32001,
|
|
|
|
| 4 |
"<START_A>": 32002,
|
| 5 |
"<START_Q>": 32000
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"<ASC>": 32006,
|
| 3 |
+
"<DIM>": 32004,
|
| 4 |
"<END_A>": 32003,
|
| 5 |
"<END_Q>": 32001,
|
| 6 |
+
"<MET>": 32005,
|
| 7 |
"<START_A>": 32002,
|
| 8 |
"<START_Q>": 32000
|
| 9 |
}
|
special_tokens_map.json
CHANGED
|
@@ -3,7 +3,10 @@
|
|
| 3 |
"<START_Q>",
|
| 4 |
"<END_Q>",
|
| 5 |
"<START_A>",
|
| 6 |
-
"<END_A>"
|
|
|
|
|
|
|
|
|
|
| 7 |
],
|
| 8 |
"bos_token": {
|
| 9 |
"content": "<s>",
|
|
|
|
| 3 |
"<START_Q>",
|
| 4 |
"<END_Q>",
|
| 5 |
"<START_A>",
|
| 6 |
+
"<END_A>",
|
| 7 |
+
"<DIM>",
|
| 8 |
+
"<MET>",
|
| 9 |
+
"<ASC>"
|
| 10 |
],
|
| 11 |
"bos_token": {
|
| 12 |
"content": "<s>",
|
tokenizer.json
CHANGED
|
@@ -70,6 +70,33 @@
|
|
| 70 |
"rstrip": false,
|
| 71 |
"normalized": false,
|
| 72 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
}
|
| 74 |
],
|
| 75 |
"normalizer": {
|
|
|
|
| 70 |
"rstrip": false,
|
| 71 |
"normalized": false,
|
| 72 |
"special": true
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"id": 32004,
|
| 76 |
+
"content": "<DIM>",
|
| 77 |
+
"single_word": false,
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"id": 32005,
|
| 85 |
+
"content": "<MET>",
|
| 86 |
+
"single_word": false,
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"normalized": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"id": 32006,
|
| 94 |
+
"content": "<ASC>",
|
| 95 |
+
"single_word": false,
|
| 96 |
+
"lstrip": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"normalized": false,
|
| 99 |
+
"special": true
|
| 100 |
}
|
| 101 |
],
|
| 102 |
"normalizer": {
|