Commit
·
3de4a0a
1
Parent(s):
9cf58e4
Upload tokenizer
Browse files- tokenizer_config.json +12 -3
tokenizer_config.json
CHANGED
|
@@ -2,6 +2,14 @@
|
|
| 2 |
"add_bos_token": true,
|
| 3 |
"add_eos_token": false,
|
| 4 |
"bos_token": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"__type": "AddedToken",
|
| 6 |
"content": "<s>",
|
| 7 |
"lstrip": false,
|
|
@@ -9,7 +17,6 @@
|
|
| 9 |
"rstrip": false,
|
| 10 |
"single_word": false
|
| 11 |
},
|
| 12 |
-
"clean_up_tokenization_spaces": false,
|
| 13 |
"eos_token": {
|
| 14 |
"__type": "AddedToken",
|
| 15 |
"content": "</s>",
|
|
@@ -19,8 +26,9 @@
|
|
| 19 |
"single_word": false
|
| 20 |
},
|
| 21 |
"legacy": null,
|
| 22 |
-
"model_max_length":
|
| 23 |
"pad_token": null,
|
|
|
|
| 24 |
"sp_model_kwargs": {},
|
| 25 |
"tokenizer_class": "LlamaTokenizer",
|
| 26 |
"unk_token": {
|
|
@@ -30,5 +38,6 @@
|
|
| 30 |
"normalized": true,
|
| 31 |
"rstrip": false,
|
| 32 |
"single_word": false
|
| 33 |
-
}
|
|
|
|
| 34 |
}
|
|
|
|
| 2 |
"add_bos_token": true,
|
| 3 |
"add_eos_token": false,
|
| 4 |
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": {
|
| 13 |
"__type": "AddedToken",
|
| 14 |
"content": "<s>",
|
| 15 |
"lstrip": false,
|
|
|
|
| 17 |
"rstrip": false,
|
| 18 |
"single_word": false
|
| 19 |
},
|
|
|
|
| 20 |
"eos_token": {
|
| 21 |
"__type": "AddedToken",
|
| 22 |
"content": "</s>",
|
|
|
|
| 26 |
"single_word": false
|
| 27 |
},
|
| 28 |
"legacy": null,
|
| 29 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 30 |
"pad_token": null,
|
| 31 |
+
"padding_side": "right",
|
| 32 |
"sp_model_kwargs": {},
|
| 33 |
"tokenizer_class": "LlamaTokenizer",
|
| 34 |
"unk_token": {
|
|
|
|
| 38 |
"normalized": true,
|
| 39 |
"rstrip": false,
|
| 40 |
"single_word": false
|
| 41 |
+
},
|
| 42 |
+
"use_fast": false
|
| 43 |
}
|