Upload tokenizer
Browse files- added_tokens.json +5 -4
- special_tokens_map.json +2 -2
- tokenizer_config.json +21 -5
added_tokens.json
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"</
|
| 3 |
-
"<
|
| 4 |
-
"<
|
| 5 |
-
"<
|
|
|
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"</EOS>": 304,
|
| 3 |
+
"</s>": 306,
|
| 4 |
+
"<mask>": 308,
|
| 5 |
+
"<s>": 307,
|
| 6 |
+
"<unk>": 305
|
| 7 |
}
|
special_tokens_map.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"bos_token": {
|
| 3 |
-
"content": "<
|
| 4 |
"lstrip": false,
|
| 5 |
"normalized": true,
|
| 6 |
"rstrip": false,
|
|
@@ -14,7 +14,7 @@
|
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
"eos_token": {
|
| 17 |
-
"content": "</
|
| 18 |
"lstrip": false,
|
| 19 |
"normalized": true,
|
| 20 |
"rstrip": false,
|
|
|
|
| 1 |
{
|
| 2 |
"bos_token": {
|
| 3 |
+
"content": "<BOS>",
|
| 4 |
"lstrip": false,
|
| 5 |
"normalized": true,
|
| 6 |
"rstrip": false,
|
|
|
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
"eos_token": {
|
| 17 |
+
"content": "</EOS>",
|
| 18 |
"lstrip": false,
|
| 19 |
"normalized": true,
|
| 20 |
"rstrip": false,
|
tokenizer_config.json
CHANGED
|
@@ -9,8 +9,16 @@
|
|
| 9 |
"single_word": false,
|
| 10 |
"special": true
|
| 11 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
"304": {
|
| 13 |
-
"content": "<
|
| 14 |
"lstrip": false,
|
| 15 |
"normalized": true,
|
| 16 |
"rstrip": false,
|
|
@@ -18,7 +26,7 @@
|
|
| 18 |
"special": true
|
| 19 |
},
|
| 20 |
"305": {
|
| 21 |
-
"content": "<
|
| 22 |
"lstrip": false,
|
| 23 |
"normalized": true,
|
| 24 |
"rstrip": false,
|
|
@@ -26,7 +34,7 @@
|
|
| 26 |
"special": true
|
| 27 |
},
|
| 28 |
"306": {
|
| 29 |
-
"content": "<
|
| 30 |
"lstrip": false,
|
| 31 |
"normalized": true,
|
| 32 |
"rstrip": false,
|
|
@@ -34,6 +42,14 @@
|
|
| 34 |
"special": true
|
| 35 |
},
|
| 36 |
"307": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"content": "<mask>",
|
| 38 |
"lstrip": true,
|
| 39 |
"normalized": true,
|
|
@@ -42,10 +58,10 @@
|
|
| 42 |
"special": true
|
| 43 |
}
|
| 44 |
},
|
| 45 |
-
"bos_token": "<
|
| 46 |
"clean_up_tokenization_spaces": true,
|
| 47 |
"cls_token": "<s>",
|
| 48 |
-
"eos_token": "</
|
| 49 |
"errors": "replace",
|
| 50 |
"mask_token": "<mask>",
|
| 51 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 9 |
"single_word": false,
|
| 10 |
"special": true
|
| 11 |
},
|
| 12 |
+
"2": {
|
| 13 |
+
"content": "<BOS>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
"304": {
|
| 21 |
+
"content": "</EOS>",
|
| 22 |
"lstrip": false,
|
| 23 |
"normalized": true,
|
| 24 |
"rstrip": false,
|
|
|
|
| 26 |
"special": true
|
| 27 |
},
|
| 28 |
"305": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
"lstrip": false,
|
| 31 |
"normalized": true,
|
| 32 |
"rstrip": false,
|
|
|
|
| 34 |
"special": true
|
| 35 |
},
|
| 36 |
"306": {
|
| 37 |
+
"content": "</s>",
|
| 38 |
"lstrip": false,
|
| 39 |
"normalized": true,
|
| 40 |
"rstrip": false,
|
|
|
|
| 42 |
"special": true
|
| 43 |
},
|
| 44 |
"307": {
|
| 45 |
+
"content": "<s>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"308": {
|
| 53 |
"content": "<mask>",
|
| 54 |
"lstrip": true,
|
| 55 |
"normalized": true,
|
|
|
|
| 58 |
"special": true
|
| 59 |
}
|
| 60 |
},
|
| 61 |
+
"bos_token": "<BOS>",
|
| 62 |
"clean_up_tokenization_spaces": true,
|
| 63 |
"cls_token": "<s>",
|
| 64 |
+
"eos_token": "</EOS>",
|
| 65 |
"errors": "replace",
|
| 66 |
"mask_token": "<mask>",
|
| 67 |
"model_max_length": 1000000000000000019884624838656,
|