Locutusque commited on
Commit
adb4d6d
·
verified ·
1 Parent(s): 56beedd

Upload tokenizer

Browse files
Files changed (2) hide show
  1. special_tokens_map.json +1 -7
  2. tokenizer_config.json +1 -8
special_tokens_map.json CHANGED
@@ -13,11 +13,5 @@
13
  "rstrip": false,
14
  "single_word": false
15
  },
16
- "pad_token": {
17
- "content": "<|end_of_text|>",
18
- "lstrip": false,
19
- "normalized": false,
20
- "rstrip": false,
21
- "single_word": false
22
- }
23
  }
 
13
  "rstrip": false,
14
  "single_word": false
15
  },
16
+ "pad_token": "<|end_of_text|>"
 
 
 
 
 
 
17
  }
tokenizer_config.json CHANGED
@@ -2053,18 +2053,11 @@
2053
  "clean_up_tokenization_spaces": true,
2054
  "eos_token": "<|end_of_text|>",
2055
  "extra_special_tokens": {},
2056
- "max_length": 1024,
2057
  "model_input_names": [
2058
  "input_ids",
2059
  "attention_mask"
2060
  ],
2061
  "model_max_length": 131072,
2062
- "pad_to_multiple_of": null,
2063
  "pad_token": "<|end_of_text|>",
2064
- "pad_token_type_id": 0,
2065
- "padding_side": "right",
2066
- "stride": 0,
2067
- "tokenizer_class": "PreTrainedTokenizerFast",
2068
- "truncation_side": "right",
2069
- "truncation_strategy": "longest_first"
2070
  }
 
2053
  "clean_up_tokenization_spaces": true,
2054
  "eos_token": "<|end_of_text|>",
2055
  "extra_special_tokens": {},
 
2056
  "model_input_names": [
2057
  "input_ids",
2058
  "attention_mask"
2059
  ],
2060
  "model_max_length": 131072,
 
2061
  "pad_token": "<|end_of_text|>",
2062
+ "tokenizer_class": "PreTrainedTokenizerFast"
 
 
 
 
 
2063
  }