Upload tokenizer
Browse files- tokenization.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +23 -10
tokenization.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a77e8519d158dc272f59cc782244f3cde7a8fc01f59d4bd67b3812209837ad6f
|
| 3 |
+
size 489008
|
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
CHANGED
|
@@ -1,4 +1,7 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
| 2 |
"added_tokens_decoder": {
|
| 3 |
"0": {
|
| 4 |
"content": "[PAD]",
|
|
@@ -9,7 +12,7 @@
|
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
"1": {
|
| 12 |
-
"content": "[
|
| 13 |
"lstrip": false,
|
| 14 |
"normalized": false,
|
| 15 |
"rstrip": false,
|
|
@@ -17,7 +20,7 @@
|
|
| 17 |
"special": true
|
| 18 |
},
|
| 19 |
"2": {
|
| 20 |
-
"content": "[
|
| 21 |
"lstrip": false,
|
| 22 |
"normalized": false,
|
| 23 |
"rstrip": false,
|
|
@@ -25,7 +28,7 @@
|
|
| 25 |
"special": true
|
| 26 |
},
|
| 27 |
"3": {
|
| 28 |
-
"content": "[
|
| 29 |
"lstrip": false,
|
| 30 |
"normalized": false,
|
| 31 |
"rstrip": false,
|
|
@@ -33,7 +36,7 @@
|
|
| 33 |
"special": true
|
| 34 |
},
|
| 35 |
"4": {
|
| 36 |
-
"content": "[
|
| 37 |
"lstrip": false,
|
| 38 |
"normalized": false,
|
| 39 |
"rstrip": false,
|
|
@@ -41,7 +44,7 @@
|
|
| 41 |
"special": true
|
| 42 |
},
|
| 43 |
"5": {
|
| 44 |
-
"content": "[
|
| 45 |
"lstrip": false,
|
| 46 |
"normalized": false,
|
| 47 |
"rstrip": false,
|
|
@@ -49,22 +52,32 @@
|
|
| 49 |
"special": true
|
| 50 |
},
|
| 51 |
"6": {
|
| 52 |
-
"content": "[
|
| 53 |
"lstrip": false,
|
| 54 |
"normalized": false,
|
| 55 |
"rstrip": false,
|
| 56 |
"single_word": false,
|
| 57 |
-
"special":
|
| 58 |
}
|
| 59 |
},
|
| 60 |
"bos_token": "[BOS]",
|
| 61 |
-
"
|
|
|
|
| 62 |
"cls_token": "[CLS]",
|
|
|
|
| 63 |
"eos_token": "[EOS]",
|
|
|
|
|
|
|
| 64 |
"mask_token": "[MASK]",
|
|
|
|
| 65 |
"model_max_length": 1000000000000000019884624838656,
|
| 66 |
"pad_token": "[PAD]",
|
|
|
|
| 67 |
"sep_token": "[SEP]",
|
| 68 |
-
"
|
| 69 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
"added_tokens_decoder": {
|
| 6 |
"0": {
|
| 7 |
"content": "[PAD]",
|
|
|
|
| 12 |
"special": true
|
| 13 |
},
|
| 14 |
"1": {
|
| 15 |
+
"content": "[UNK]",
|
| 16 |
"lstrip": false,
|
| 17 |
"normalized": false,
|
| 18 |
"rstrip": false,
|
|
|
|
| 20 |
"special": true
|
| 21 |
},
|
| 22 |
"2": {
|
| 23 |
+
"content": "[BOS]",
|
| 24 |
"lstrip": false,
|
| 25 |
"normalized": false,
|
| 26 |
"rstrip": false,
|
|
|
|
| 28 |
"special": true
|
| 29 |
},
|
| 30 |
"3": {
|
| 31 |
+
"content": "[EOS]",
|
| 32 |
"lstrip": false,
|
| 33 |
"normalized": false,
|
| 34 |
"rstrip": false,
|
|
|
|
| 36 |
"special": true
|
| 37 |
},
|
| 38 |
"4": {
|
| 39 |
+
"content": "[CLS]",
|
| 40 |
"lstrip": false,
|
| 41 |
"normalized": false,
|
| 42 |
"rstrip": false,
|
|
|
|
| 44 |
"special": true
|
| 45 |
},
|
| 46 |
"5": {
|
| 47 |
+
"content": "[SEP]",
|
| 48 |
"lstrip": false,
|
| 49 |
"normalized": false,
|
| 50 |
"rstrip": false,
|
|
|
|
| 52 |
"special": true
|
| 53 |
},
|
| 54 |
"6": {
|
| 55 |
+
"content": "[MASK]",
|
| 56 |
"lstrip": false,
|
| 57 |
"normalized": false,
|
| 58 |
"rstrip": false,
|
| 59 |
"single_word": false,
|
| 60 |
+
"special": false
|
| 61 |
}
|
| 62 |
},
|
| 63 |
"bos_token": "[BOS]",
|
| 64 |
+
"bos_token_id": 2,
|
| 65 |
+
"clean_up_tokenization_spaces": false,
|
| 66 |
"cls_token": "[CLS]",
|
| 67 |
+
"cls_token_id": 4,
|
| 68 |
"eos_token": "[EOS]",
|
| 69 |
+
"eos_token_id": 3,
|
| 70 |
+
"legacy": true,
|
| 71 |
"mask_token": "[MASK]",
|
| 72 |
+
"mask_token_id": 6,
|
| 73 |
"model_max_length": 1000000000000000019884624838656,
|
| 74 |
"pad_token": "[PAD]",
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
"sep_token": "[SEP]",
|
| 77 |
+
"sep_token_id": 5,
|
| 78 |
+
"sp_model_kwargs": {},
|
| 79 |
+
"spaces_between_special_tokens": false,
|
| 80 |
+
"tokenizer_class": "HfSentencePieceTokenizer",
|
| 81 |
+
"unk_token": "[UNK]",
|
| 82 |
+
"unk_token_id": 1
|
| 83 |
}
|