Commit
•
b30df56
1
Parent(s):
e4eb1e4
Upload tokenizer
Browse files- added_tokens.json +3 -2
- special_tokens_map.json +7 -0
- vocab.json +0 -2
added_tokens.json
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"</s>":
|
| 3 |
-
"<s>":
|
|
|
|
| 4 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"</s>": 28,
|
| 3 |
+
"<s>": 27,
|
| 4 |
+
"[PAD]": 29
|
| 5 |
}
|
special_tokens_map.json
CHANGED
|
@@ -13,6 +13,13 @@
|
|
| 13 |
"normalized": true,
|
| 14 |
"rstrip": false,
|
| 15 |
"single_word": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
}
|
| 17 |
],
|
| 18 |
"bos_token": "<s>",
|
|
|
|
| 13 |
"normalized": true,
|
| 14 |
"rstrip": false,
|
| 15 |
"single_word": false
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"content": "[PAD]",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
}
|
| 24 |
],
|
| 25 |
"bos_token": "<s>",
|
vocab.json
CHANGED
|
@@ -1,6 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"[PAD]": 28,
|
| 3 |
-
"[UNK]": 27,
|
| 4 |
"a": 1,
|
| 5 |
"b": 2,
|
| 6 |
"c": 3,
|
|
|
|
| 1 |
{
|
|
|
|
|
|
|
| 2 |
"a": 1,
|
| 3 |
"b": 2,
|
| 4 |
"c": 3,
|