yuw385 commited on
Commit
0764067
1 Parent(s): 2532719

Upload tokenizer

Browse files
special_tokens_map.json CHANGED
@@ -12,5 +12,6 @@
12
  ">>SUFFIX<<",
13
  ">>MIDDLE<<"
14
  ],
15
- "eos_token": "<|endoftext|>"
 
16
  }
 
12
  ">>SUFFIX<<",
13
  ">>MIDDLE<<"
14
  ],
15
+ "eos_token": "<|endoftext|>",
16
+ "pad_token": "<|endoftext|>"
17
  }
tokenizer.json CHANGED
@@ -1,7 +1,21 @@
1
  {
2
  "version": "1.0",
3
- "truncation": null,
4
- "padding": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  "added_tokens": [
6
  {
7
  "id": 0,
 
1
  {
2
  "version": "1.0",
3
+ "truncation": {
4
+ "direction": "Right",
5
+ "max_length": 2048,
6
+ "strategy": "LongestFirst",
7
+ "stride": 0
8
+ },
9
+ "padding": {
10
+ "strategy": {
11
+ "Fixed": 2048
12
+ },
13
+ "direction": "Right",
14
+ "pad_to_multiple_of": null,
15
+ "pad_id": 11,
16
+ "pad_type_id": 0,
17
+ "pad_token": "<|endoftext|>"
18
+ },
19
  "added_tokens": [
20
  {
21
  "id": 0,
tokenizer_config.json CHANGED
@@ -118,5 +118,6 @@
118
  "attention_mask"
119
  ],
120
  "model_max_length": 2048,
 
121
  "tokenizer_class": "PreTrainedTokenizerFast"
122
  }
 
118
  "attention_mask"
119
  ],
120
  "model_max_length": 2048,
121
+ "pad_token": "<|endoftext|>",
122
  "tokenizer_class": "PreTrainedTokenizerFast"
123
  }