danielhanchen
commited on
Commit
•
66944c1
1
Parent(s):
cdd1524
Upload tokenizer
Browse files- special_tokens_map.json +7 -1
- tokenizer.json +0 -9
- tokenizer_config.json +1 -9
special_tokens_map.json
CHANGED
@@ -21,5 +21,11 @@
|
|
21 |
"rstrip": false,
|
22 |
"single_word": false
|
23 |
},
|
24 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
}
|
|
|
21 |
"rstrip": false,
|
22 |
"single_word": false
|
23 |
},
|
24 |
+
"pad_token": {
|
25 |
+
"content": "<|reserved_special_token_250|>",
|
26 |
+
"lstrip": false,
|
27 |
+
"normalized": false,
|
28 |
+
"rstrip": false,
|
29 |
+
"single_word": false
|
30 |
+
}
|
31 |
}
|
tokenizer.json
CHANGED
@@ -2360,15 +2360,6 @@
|
|
2360 |
"rstrip": false,
|
2361 |
"normalized": false,
|
2362 |
"special": true
|
2363 |
-
},
|
2364 |
-
{
|
2365 |
-
"id": 128262,
|
2366 |
-
"content": "<|finetune_right_pad_id|>",
|
2367 |
-
"single_word": false,
|
2368 |
-
"lstrip": false,
|
2369 |
-
"rstrip": false,
|
2370 |
-
"normalized": false,
|
2371 |
-
"special": true
|
2372 |
}
|
2373 |
],
|
2374 |
"normalizer": null,
|
|
|
2360 |
"rstrip": false,
|
2361 |
"normalized": false,
|
2362 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2363 |
}
|
2364 |
],
|
2365 |
"normalizer": null,
|
tokenizer_config.json
CHANGED
@@ -2095,14 +2095,6 @@
|
|
2095 |
"rstrip": false,
|
2096 |
"single_word": false,
|
2097 |
"special": true
|
2098 |
-
},
|
2099 |
-
"128262": {
|
2100 |
-
"content": "<|finetune_right_pad_id|>",
|
2101 |
-
"lstrip": false,
|
2102 |
-
"normalized": false,
|
2103 |
-
"rstrip": false,
|
2104 |
-
"single_word": false,
|
2105 |
-
"special": true
|
2106 |
}
|
2107 |
},
|
2108 |
"additional_special_tokens": [
|
@@ -2122,7 +2114,7 @@
|
|
2122 |
"attention_mask"
|
2123 |
],
|
2124 |
"model_max_length": 1000000000000000019884624838656,
|
2125 |
-
"pad_token": "<|
|
2126 |
"padding_side": "left",
|
2127 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
2128 |
}
|
|
|
2095 |
"rstrip": false,
|
2096 |
"single_word": false,
|
2097 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2098 |
}
|
2099 |
},
|
2100 |
"additional_special_tokens": [
|
|
|
2114 |
"attention_mask"
|
2115 |
],
|
2116 |
"model_max_length": 1000000000000000019884624838656,
|
2117 |
+
"pad_token": "<|reserved_special_token_250|>",
|
2118 |
"padding_side": "left",
|
2119 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
2120 |
}
|