Commit
•
c763e3e
1
Parent(s):
40ce229
Upload tokenizer (#2)
Browse files- Upload tokenizer (eee8f7e68307d3a6dd5a9b81d6be2ea6cd834a7b)
- special_tokens_map.json +11 -0
- tokenizer.json +0 -0
- tokenizer_config.json +7 -0
special_tokens_map.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|endoftext|>",
|
| 4 |
+
"<|sep|>",
|
| 5 |
+
"<|acc|>",
|
| 6 |
+
"<|tel|>",
|
| 7 |
+
"<|rrn|>"
|
| 8 |
+
],
|
| 9 |
+
"eos_token": "<|endoftext|>",
|
| 10 |
+
"pad_token": "<|endoftext|>"
|
| 11 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 5 |
+
"pad_token": "<|endoftext|>",
|
| 6 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 7 |
+
}
|