Commit
•
144eeb1
1
Parent(s):
7b7af6b
Upload tokenizer
Browse files- special_tokens_map.json +3 -5
- spiece.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +10 -13
special_tokens_map.json
CHANGED
|
@@ -1,7 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"sep_token": "[SEP]",
|
| 6 |
-
"unk_token": "[UNK]"
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
|
|
|
|
|
|
| 5 |
}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49e3442998423498b1c036276787f7d065fa673d76b9dafb83bd5fa19d17a6c5
|
| 3 |
+
size 742072
|
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
CHANGED
|
@@ -1,15 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"name_or_path": "AlanRobotics/
|
| 7 |
-
"
|
| 8 |
-
"
|
| 9 |
-
"
|
| 10 |
-
"
|
| 11 |
-
"
|
| 12 |
-
"tokenize_chinese_chars": true,
|
| 13 |
-
"tokenizer_class": "BertTokenizer",
|
| 14 |
-
"unk_token": "[UNK]"
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"additional_special_tokens": null,
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"extra_ids": 0,
|
| 5 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 6 |
+
"name_or_path": "AlanRobotics/ruT5-base",
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"sp_model_kwargs": {},
|
| 9 |
+
"special_tokens_map_file": "/Users/alanbarsag/.cache/huggingface/hub/models--AlanRobotics--ruT5-base/snapshots/06d1ee6609acbe036c7699aa744d6aca561e7a85/special_tokens_map.json",
|
| 10 |
+
"tokenizer_class": "T5Tokenizer",
|
| 11 |
+
"unk_token": "<unk>"
|
|
|
|
|
|
|
|
|
|
| 12 |
}
|