Upload 4 files
Browse files- special_tokens_map.json +4 -117
- tokenizer.json +0 -0
- tokenizer_config.json +8 -1088
special_tokens_map.json
CHANGED
|
@@ -1,119 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
"<extra_id_3>",
|
| 7 |
-
"<extra_id_4>",
|
| 8 |
-
"<extra_id_5>",
|
| 9 |
-
"<extra_id_6>",
|
| 10 |
-
"<extra_id_7>",
|
| 11 |
-
"<extra_id_8>",
|
| 12 |
-
"<extra_id_9>",
|
| 13 |
-
"<extra_id_10>",
|
| 14 |
-
"<extra_id_11>",
|
| 15 |
-
"<extra_id_12>",
|
| 16 |
-
"<extra_id_13>",
|
| 17 |
-
"<extra_id_14>",
|
| 18 |
-
"<extra_id_15>",
|
| 19 |
-
"<extra_id_16>",
|
| 20 |
-
"<extra_id_17>",
|
| 21 |
-
"<extra_id_18>",
|
| 22 |
-
"<extra_id_19>",
|
| 23 |
-
"<extra_id_20>",
|
| 24 |
-
"<extra_id_21>",
|
| 25 |
-
"<extra_id_22>",
|
| 26 |
-
"<extra_id_23>",
|
| 27 |
-
"<extra_id_24>",
|
| 28 |
-
"<extra_id_25>",
|
| 29 |
-
"<extra_id_26>",
|
| 30 |
-
"<extra_id_27>",
|
| 31 |
-
"<extra_id_28>",
|
| 32 |
-
"<extra_id_29>",
|
| 33 |
-
"<extra_id_30>",
|
| 34 |
-
"<extra_id_31>",
|
| 35 |
-
"<extra_id_32>",
|
| 36 |
-
"<extra_id_33>",
|
| 37 |
-
"<extra_id_34>",
|
| 38 |
-
"<extra_id_35>",
|
| 39 |
-
"<extra_id_36>",
|
| 40 |
-
"<extra_id_37>",
|
| 41 |
-
"<extra_id_38>",
|
| 42 |
-
"<extra_id_39>",
|
| 43 |
-
"<extra_id_40>",
|
| 44 |
-
"<extra_id_41>",
|
| 45 |
-
"<extra_id_42>",
|
| 46 |
-
"<extra_id_43>",
|
| 47 |
-
"<extra_id_44>",
|
| 48 |
-
"<extra_id_45>",
|
| 49 |
-
"<extra_id_46>",
|
| 50 |
-
"<extra_id_47>",
|
| 51 |
-
"<extra_id_48>",
|
| 52 |
-
"<extra_id_49>",
|
| 53 |
-
"<extra_id_50>",
|
| 54 |
-
"<extra_id_51>",
|
| 55 |
-
"<extra_id_52>",
|
| 56 |
-
"<extra_id_53>",
|
| 57 |
-
"<extra_id_54>",
|
| 58 |
-
"<extra_id_55>",
|
| 59 |
-
"<extra_id_56>",
|
| 60 |
-
"<extra_id_57>",
|
| 61 |
-
"<extra_id_58>",
|
| 62 |
-
"<extra_id_59>",
|
| 63 |
-
"<extra_id_60>",
|
| 64 |
-
"<extra_id_61>",
|
| 65 |
-
"<extra_id_62>",
|
| 66 |
-
"<extra_id_63>",
|
| 67 |
-
"<extra_id_64>",
|
| 68 |
-
"<extra_id_65>",
|
| 69 |
-
"<extra_id_66>",
|
| 70 |
-
"<extra_id_67>",
|
| 71 |
-
"<extra_id_68>",
|
| 72 |
-
"<extra_id_69>",
|
| 73 |
-
"<extra_id_70>",
|
| 74 |
-
"<extra_id_71>",
|
| 75 |
-
"<extra_id_72>",
|
| 76 |
-
"<extra_id_73>",
|
| 77 |
-
"<extra_id_74>",
|
| 78 |
-
"<extra_id_75>",
|
| 79 |
-
"<extra_id_76>",
|
| 80 |
-
"<extra_id_77>",
|
| 81 |
-
"<extra_id_78>",
|
| 82 |
-
"<extra_id_79>",
|
| 83 |
-
"<extra_id_80>",
|
| 84 |
-
"<extra_id_81>",
|
| 85 |
-
"<extra_id_82>",
|
| 86 |
-
"<extra_id_83>",
|
| 87 |
-
"<extra_id_84>",
|
| 88 |
-
"<extra_id_85>",
|
| 89 |
-
"<extra_id_86>",
|
| 90 |
-
"<extra_id_87>",
|
| 91 |
-
"<extra_id_88>",
|
| 92 |
-
"<extra_id_89>",
|
| 93 |
-
"<extra_id_90>",
|
| 94 |
-
"<extra_id_91>",
|
| 95 |
-
"<extra_id_92>",
|
| 96 |
-
"<extra_id_93>",
|
| 97 |
-
"<extra_id_94>",
|
| 98 |
-
"<extra_id_95>",
|
| 99 |
-
"<extra_id_96>",
|
| 100 |
-
"<extra_id_97>",
|
| 101 |
-
"<extra_id_98>",
|
| 102 |
-
"<extra_id_99>"
|
| 103 |
-
],
|
| 104 |
-
"eos_token": {
|
| 105 |
-
"content": "</s>",
|
| 106 |
-
"lstrip": false,
|
| 107 |
-
"normalized": false,
|
| 108 |
-
"rstrip": false,
|
| 109 |
-
"single_word": false
|
| 110 |
-
},
|
| 111 |
-
"pad_token": "</s>",
|
| 112 |
-
"unk_token": {
|
| 113 |
-
"content": "<unk>",
|
| 114 |
-
"lstrip": false,
|
| 115 |
-
"normalized": false,
|
| 116 |
-
"rstrip": false,
|
| 117 |
-
"single_word": false
|
| 118 |
-
}
|
| 119 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|pad|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
}
|
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
CHANGED
|
@@ -1,1100 +1,20 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"enable": true,
|
| 4 |
-
"max_tokens": 100000,
|
| 5 |
-
"min_frequency": 3,
|
| 6 |
-
"retain_variant_forms": true
|
| 7 |
-
},
|
| 8 |
"added_tokens_decoder": {
|
| 9 |
-
"
|
| 10 |
-
"content": "
|
| 11 |
-
"lstrip": false,
|
| 12 |
-
"normalized": false,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false,
|
| 15 |
-
"special": true
|
| 16 |
-
},
|
| 17 |
-
"1": {
|
| 18 |
-
"content": "</s>",
|
| 19 |
-
"lstrip": false,
|
| 20 |
-
"normalized": false,
|
| 21 |
-
"rstrip": false,
|
| 22 |
-
"single_word": false,
|
| 23 |
-
"special": true
|
| 24 |
-
},
|
| 25 |
-
"2": {
|
| 26 |
-
"content": "<unk>",
|
| 27 |
-
"lstrip": false,
|
| 28 |
-
"normalized": false,
|
| 29 |
-
"rstrip": false,
|
| 30 |
-
"single_word": false,
|
| 31 |
-
"special": true
|
| 32 |
-
},
|
| 33 |
-
"10452": {
|
| 34 |
-
"content": "good",
|
| 35 |
"lstrip": false,
|
| 36 |
"normalized": true,
|
| 37 |
"rstrip": false,
|
| 38 |
"single_word": false,
|
| 39 |
-
"special": false
|
| 40 |
-
},
|
| 41 |
-
"32000": {
|
| 42 |
-
"content": "<extra_id_99>",
|
| 43 |
-
"lstrip": false,
|
| 44 |
-
"normalized": false,
|
| 45 |
-
"rstrip": false,
|
| 46 |
-
"single_word": false,
|
| 47 |
-
"special": true
|
| 48 |
-
},
|
| 49 |
-
"32001": {
|
| 50 |
-
"content": "<extra_id_98>",
|
| 51 |
-
"lstrip": false,
|
| 52 |
-
"normalized": false,
|
| 53 |
-
"rstrip": false,
|
| 54 |
-
"single_word": false,
|
| 55 |
-
"special": true
|
| 56 |
-
},
|
| 57 |
-
"32002": {
|
| 58 |
-
"content": "<extra_id_97>",
|
| 59 |
-
"lstrip": false,
|
| 60 |
-
"normalized": false,
|
| 61 |
-
"rstrip": false,
|
| 62 |
-
"single_word": false,
|
| 63 |
-
"special": true
|
| 64 |
-
},
|
| 65 |
-
"32003": {
|
| 66 |
-
"content": "<extra_id_96>",
|
| 67 |
-
"lstrip": false,
|
| 68 |
-
"normalized": false,
|
| 69 |
-
"rstrip": false,
|
| 70 |
-
"single_word": false,
|
| 71 |
-
"special": true
|
| 72 |
-
},
|
| 73 |
-
"32004": {
|
| 74 |
-
"content": "<extra_id_95>",
|
| 75 |
-
"lstrip": false,
|
| 76 |
-
"normalized": false,
|
| 77 |
-
"rstrip": false,
|
| 78 |
-
"single_word": false,
|
| 79 |
-
"special": true
|
| 80 |
-
},
|
| 81 |
-
"32005": {
|
| 82 |
-
"content": "<extra_id_94>",
|
| 83 |
-
"lstrip": false,
|
| 84 |
-
"normalized": false,
|
| 85 |
-
"rstrip": false,
|
| 86 |
-
"single_word": false,
|
| 87 |
-
"special": true
|
| 88 |
-
},
|
| 89 |
-
"32006": {
|
| 90 |
-
"content": "<extra_id_93>",
|
| 91 |
-
"lstrip": false,
|
| 92 |
-
"normalized": false,
|
| 93 |
-
"rstrip": false,
|
| 94 |
-
"single_word": false,
|
| 95 |
-
"special": true
|
| 96 |
-
},
|
| 97 |
-
"32007": {
|
| 98 |
-
"content": "<extra_id_92>",
|
| 99 |
-
"lstrip": false,
|
| 100 |
-
"normalized": false,
|
| 101 |
-
"rstrip": false,
|
| 102 |
-
"single_word": false,
|
| 103 |
-
"special": true
|
| 104 |
-
},
|
| 105 |
-
"32008": {
|
| 106 |
-
"content": "<extra_id_91>",
|
| 107 |
-
"lstrip": false,
|
| 108 |
-
"normalized": false,
|
| 109 |
-
"rstrip": false,
|
| 110 |
-
"single_word": false,
|
| 111 |
-
"special": true
|
| 112 |
-
},
|
| 113 |
-
"32009": {
|
| 114 |
-
"content": "<extra_id_90>",
|
| 115 |
-
"lstrip": false,
|
| 116 |
-
"normalized": false,
|
| 117 |
-
"rstrip": false,
|
| 118 |
-
"single_word": false,
|
| 119 |
-
"special": true
|
| 120 |
-
},
|
| 121 |
-
"32010": {
|
| 122 |
-
"content": "<extra_id_89>",
|
| 123 |
-
"lstrip": false,
|
| 124 |
-
"normalized": false,
|
| 125 |
-
"rstrip": false,
|
| 126 |
-
"single_word": false,
|
| 127 |
-
"special": true
|
| 128 |
-
},
|
| 129 |
-
"32011": {
|
| 130 |
-
"content": "<extra_id_88>",
|
| 131 |
-
"lstrip": false,
|
| 132 |
-
"normalized": false,
|
| 133 |
-
"rstrip": false,
|
| 134 |
-
"single_word": false,
|
| 135 |
-
"special": true
|
| 136 |
-
},
|
| 137 |
-
"32012": {
|
| 138 |
-
"content": "<extra_id_87>",
|
| 139 |
-
"lstrip": false,
|
| 140 |
-
"normalized": false,
|
| 141 |
-
"rstrip": false,
|
| 142 |
-
"single_word": false,
|
| 143 |
-
"special": true
|
| 144 |
-
},
|
| 145 |
-
"32013": {
|
| 146 |
-
"content": "<extra_id_86>",
|
| 147 |
-
"lstrip": false,
|
| 148 |
-
"normalized": false,
|
| 149 |
-
"rstrip": false,
|
| 150 |
-
"single_word": false,
|
| 151 |
-
"special": true
|
| 152 |
-
},
|
| 153 |
-
"32014": {
|
| 154 |
-
"content": "<extra_id_85>",
|
| 155 |
-
"lstrip": false,
|
| 156 |
-
"normalized": false,
|
| 157 |
-
"rstrip": false,
|
| 158 |
-
"single_word": false,
|
| 159 |
-
"special": true
|
| 160 |
-
},
|
| 161 |
-
"32015": {
|
| 162 |
-
"content": "<extra_id_84>",
|
| 163 |
-
"lstrip": false,
|
| 164 |
-
"normalized": false,
|
| 165 |
-
"rstrip": false,
|
| 166 |
-
"single_word": false,
|
| 167 |
-
"special": true
|
| 168 |
-
},
|
| 169 |
-
"32016": {
|
| 170 |
-
"content": "<extra_id_83>",
|
| 171 |
-
"lstrip": false,
|
| 172 |
-
"normalized": false,
|
| 173 |
-
"rstrip": false,
|
| 174 |
-
"single_word": false,
|
| 175 |
-
"special": true
|
| 176 |
-
},
|
| 177 |
-
"32017": {
|
| 178 |
-
"content": "<extra_id_82>",
|
| 179 |
-
"lstrip": false,
|
| 180 |
-
"normalized": false,
|
| 181 |
-
"rstrip": false,
|
| 182 |
-
"single_word": false,
|
| 183 |
-
"special": true
|
| 184 |
-
},
|
| 185 |
-
"32018": {
|
| 186 |
-
"content": "<extra_id_81>",
|
| 187 |
-
"lstrip": false,
|
| 188 |
-
"normalized": false,
|
| 189 |
-
"rstrip": false,
|
| 190 |
-
"single_word": false,
|
| 191 |
-
"special": true
|
| 192 |
-
},
|
| 193 |
-
"32019": {
|
| 194 |
-
"content": "<extra_id_80>",
|
| 195 |
-
"lstrip": false,
|
| 196 |
-
"normalized": false,
|
| 197 |
-
"rstrip": false,
|
| 198 |
-
"single_word": false,
|
| 199 |
-
"special": true
|
| 200 |
-
},
|
| 201 |
-
"32020": {
|
| 202 |
-
"content": "<extra_id_79>",
|
| 203 |
-
"lstrip": false,
|
| 204 |
-
"normalized": false,
|
| 205 |
-
"rstrip": false,
|
| 206 |
-
"single_word": false,
|
| 207 |
-
"special": true
|
| 208 |
-
},
|
| 209 |
-
"32021": {
|
| 210 |
-
"content": "<extra_id_78>",
|
| 211 |
-
"lstrip": false,
|
| 212 |
-
"normalized": false,
|
| 213 |
-
"rstrip": false,
|
| 214 |
-
"single_word": false,
|
| 215 |
-
"special": true
|
| 216 |
-
},
|
| 217 |
-
"32022": {
|
| 218 |
-
"content": "<extra_id_77>",
|
| 219 |
-
"lstrip": false,
|
| 220 |
-
"normalized": false,
|
| 221 |
-
"rstrip": false,
|
| 222 |
-
"single_word": false,
|
| 223 |
-
"special": true
|
| 224 |
-
},
|
| 225 |
-
"32023": {
|
| 226 |
-
"content": "<extra_id_76>",
|
| 227 |
-
"lstrip": false,
|
| 228 |
-
"normalized": false,
|
| 229 |
-
"rstrip": false,
|
| 230 |
-
"single_word": false,
|
| 231 |
-
"special": true
|
| 232 |
-
},
|
| 233 |
-
"32024": {
|
| 234 |
-
"content": "<extra_id_75>",
|
| 235 |
-
"lstrip": false,
|
| 236 |
-
"normalized": false,
|
| 237 |
-
"rstrip": false,
|
| 238 |
-
"single_word": false,
|
| 239 |
-
"special": true
|
| 240 |
-
},
|
| 241 |
-
"32025": {
|
| 242 |
-
"content": "<extra_id_74>",
|
| 243 |
-
"lstrip": false,
|
| 244 |
-
"normalized": false,
|
| 245 |
-
"rstrip": false,
|
| 246 |
-
"single_word": false,
|
| 247 |
-
"special": true
|
| 248 |
-
},
|
| 249 |
-
"32026": {
|
| 250 |
-
"content": "<extra_id_73>",
|
| 251 |
-
"lstrip": false,
|
| 252 |
-
"normalized": false,
|
| 253 |
-
"rstrip": false,
|
| 254 |
-
"single_word": false,
|
| 255 |
-
"special": true
|
| 256 |
-
},
|
| 257 |
-
"32027": {
|
| 258 |
-
"content": "<extra_id_72>",
|
| 259 |
-
"lstrip": false,
|
| 260 |
-
"normalized": false,
|
| 261 |
-
"rstrip": false,
|
| 262 |
-
"single_word": false,
|
| 263 |
-
"special": true
|
| 264 |
-
},
|
| 265 |
-
"32028": {
|
| 266 |
-
"content": "<extra_id_71>",
|
| 267 |
-
"lstrip": false,
|
| 268 |
-
"normalized": false,
|
| 269 |
-
"rstrip": false,
|
| 270 |
-
"single_word": false,
|
| 271 |
-
"special": true
|
| 272 |
-
},
|
| 273 |
-
"32029": {
|
| 274 |
-
"content": "<extra_id_70>",
|
| 275 |
-
"lstrip": false,
|
| 276 |
-
"normalized": false,
|
| 277 |
-
"rstrip": false,
|
| 278 |
-
"single_word": false,
|
| 279 |
-
"special": true
|
| 280 |
-
},
|
| 281 |
-
"32030": {
|
| 282 |
-
"content": "<extra_id_69>",
|
| 283 |
-
"lstrip": false,
|
| 284 |
-
"normalized": false,
|
| 285 |
-
"rstrip": false,
|
| 286 |
-
"single_word": false,
|
| 287 |
-
"special": true
|
| 288 |
-
},
|
| 289 |
-
"32031": {
|
| 290 |
-
"content": "<extra_id_68>",
|
| 291 |
-
"lstrip": false,
|
| 292 |
-
"normalized": false,
|
| 293 |
-
"rstrip": false,
|
| 294 |
-
"single_word": false,
|
| 295 |
-
"special": true
|
| 296 |
-
},
|
| 297 |
-
"32032": {
|
| 298 |
-
"content": "<extra_id_67>",
|
| 299 |
-
"lstrip": false,
|
| 300 |
-
"normalized": false,
|
| 301 |
-
"rstrip": false,
|
| 302 |
-
"single_word": false,
|
| 303 |
-
"special": true
|
| 304 |
-
},
|
| 305 |
-
"32033": {
|
| 306 |
-
"content": "<extra_id_66>",
|
| 307 |
-
"lstrip": false,
|
| 308 |
-
"normalized": false,
|
| 309 |
-
"rstrip": false,
|
| 310 |
-
"single_word": false,
|
| 311 |
-
"special": true
|
| 312 |
-
},
|
| 313 |
-
"32034": {
|
| 314 |
-
"content": "<extra_id_65>",
|
| 315 |
-
"lstrip": false,
|
| 316 |
-
"normalized": false,
|
| 317 |
-
"rstrip": false,
|
| 318 |
-
"single_word": false,
|
| 319 |
-
"special": true
|
| 320 |
-
},
|
| 321 |
-
"32035": {
|
| 322 |
-
"content": "<extra_id_64>",
|
| 323 |
-
"lstrip": false,
|
| 324 |
-
"normalized": false,
|
| 325 |
-
"rstrip": false,
|
| 326 |
-
"single_word": false,
|
| 327 |
-
"special": true
|
| 328 |
-
},
|
| 329 |
-
"32036": {
|
| 330 |
-
"content": "<extra_id_63>",
|
| 331 |
-
"lstrip": false,
|
| 332 |
-
"normalized": false,
|
| 333 |
-
"rstrip": false,
|
| 334 |
-
"single_word": false,
|
| 335 |
-
"special": true
|
| 336 |
-
},
|
| 337 |
-
"32037": {
|
| 338 |
-
"content": "<extra_id_62>",
|
| 339 |
-
"lstrip": false,
|
| 340 |
-
"normalized": false,
|
| 341 |
-
"rstrip": false,
|
| 342 |
-
"single_word": false,
|
| 343 |
-
"special": true
|
| 344 |
-
},
|
| 345 |
-
"32038": {
|
| 346 |
-
"content": "<extra_id_61>",
|
| 347 |
-
"lstrip": false,
|
| 348 |
-
"normalized": false,
|
| 349 |
-
"rstrip": false,
|
| 350 |
-
"single_word": false,
|
| 351 |
-
"special": true
|
| 352 |
-
},
|
| 353 |
-
"32039": {
|
| 354 |
-
"content": "<extra_id_60>",
|
| 355 |
-
"lstrip": false,
|
| 356 |
-
"normalized": false,
|
| 357 |
-
"rstrip": false,
|
| 358 |
-
"single_word": false,
|
| 359 |
-
"special": true
|
| 360 |
-
},
|
| 361 |
-
"32040": {
|
| 362 |
-
"content": "<extra_id_59>",
|
| 363 |
-
"lstrip": false,
|
| 364 |
-
"normalized": false,
|
| 365 |
-
"rstrip": false,
|
| 366 |
-
"single_word": false,
|
| 367 |
-
"special": true
|
| 368 |
-
},
|
| 369 |
-
"32041": {
|
| 370 |
-
"content": "<extra_id_58>",
|
| 371 |
-
"lstrip": false,
|
| 372 |
-
"normalized": false,
|
| 373 |
-
"rstrip": false,
|
| 374 |
-
"single_word": false,
|
| 375 |
-
"special": true
|
| 376 |
-
},
|
| 377 |
-
"32042": {
|
| 378 |
-
"content": "<extra_id_57>",
|
| 379 |
-
"lstrip": false,
|
| 380 |
-
"normalized": false,
|
| 381 |
-
"rstrip": false,
|
| 382 |
-
"single_word": false,
|
| 383 |
-
"special": true
|
| 384 |
-
},
|
| 385 |
-
"32043": {
|
| 386 |
-
"content": "<extra_id_56>",
|
| 387 |
-
"lstrip": false,
|
| 388 |
-
"normalized": false,
|
| 389 |
-
"rstrip": false,
|
| 390 |
-
"single_word": false,
|
| 391 |
-
"special": true
|
| 392 |
-
},
|
| 393 |
-
"32044": {
|
| 394 |
-
"content": "<extra_id_55>",
|
| 395 |
-
"lstrip": false,
|
| 396 |
-
"normalized": false,
|
| 397 |
-
"rstrip": false,
|
| 398 |
-
"single_word": false,
|
| 399 |
-
"special": true
|
| 400 |
-
},
|
| 401 |
-
"32045": {
|
| 402 |
-
"content": "<extra_id_54>",
|
| 403 |
-
"lstrip": false,
|
| 404 |
-
"normalized": false,
|
| 405 |
-
"rstrip": false,
|
| 406 |
-
"single_word": false,
|
| 407 |
-
"special": true
|
| 408 |
-
},
|
| 409 |
-
"32046": {
|
| 410 |
-
"content": "<extra_id_53>",
|
| 411 |
-
"lstrip": false,
|
| 412 |
-
"normalized": false,
|
| 413 |
-
"rstrip": false,
|
| 414 |
-
"single_word": false,
|
| 415 |
-
"special": true
|
| 416 |
-
},
|
| 417 |
-
"32047": {
|
| 418 |
-
"content": "<extra_id_52>",
|
| 419 |
-
"lstrip": false,
|
| 420 |
-
"normalized": false,
|
| 421 |
-
"rstrip": false,
|
| 422 |
-
"single_word": false,
|
| 423 |
-
"special": true
|
| 424 |
-
},
|
| 425 |
-
"32048": {
|
| 426 |
-
"content": "<extra_id_51>",
|
| 427 |
-
"lstrip": false,
|
| 428 |
-
"normalized": false,
|
| 429 |
-
"rstrip": false,
|
| 430 |
-
"single_word": false,
|
| 431 |
-
"special": true
|
| 432 |
-
},
|
| 433 |
-
"32049": {
|
| 434 |
-
"content": "<extra_id_50>",
|
| 435 |
-
"lstrip": false,
|
| 436 |
-
"normalized": false,
|
| 437 |
-
"rstrip": false,
|
| 438 |
-
"single_word": false,
|
| 439 |
-
"special": true
|
| 440 |
-
},
|
| 441 |
-
"32050": {
|
| 442 |
-
"content": "<extra_id_49>",
|
| 443 |
-
"lstrip": false,
|
| 444 |
-
"normalized": false,
|
| 445 |
-
"rstrip": false,
|
| 446 |
-
"single_word": false,
|
| 447 |
-
"special": true
|
| 448 |
-
},
|
| 449 |
-
"32051": {
|
| 450 |
-
"content": "<extra_id_48>",
|
| 451 |
-
"lstrip": false,
|
| 452 |
-
"normalized": false,
|
| 453 |
-
"rstrip": false,
|
| 454 |
-
"single_word": false,
|
| 455 |
-
"special": true
|
| 456 |
-
},
|
| 457 |
-
"32052": {
|
| 458 |
-
"content": "<extra_id_47>",
|
| 459 |
-
"lstrip": false,
|
| 460 |
-
"normalized": false,
|
| 461 |
-
"rstrip": false,
|
| 462 |
-
"single_word": false,
|
| 463 |
-
"special": true
|
| 464 |
-
},
|
| 465 |
-
"32053": {
|
| 466 |
-
"content": "<extra_id_46>",
|
| 467 |
-
"lstrip": false,
|
| 468 |
-
"normalized": false,
|
| 469 |
-
"rstrip": false,
|
| 470 |
-
"single_word": false,
|
| 471 |
-
"special": true
|
| 472 |
-
},
|
| 473 |
-
"32054": {
|
| 474 |
-
"content": "<extra_id_45>",
|
| 475 |
-
"lstrip": false,
|
| 476 |
-
"normalized": false,
|
| 477 |
-
"rstrip": false,
|
| 478 |
-
"single_word": false,
|
| 479 |
-
"special": true
|
| 480 |
-
},
|
| 481 |
-
"32055": {
|
| 482 |
-
"content": "<extra_id_44>",
|
| 483 |
-
"lstrip": false,
|
| 484 |
-
"normalized": false,
|
| 485 |
-
"rstrip": false,
|
| 486 |
-
"single_word": false,
|
| 487 |
-
"special": true
|
| 488 |
-
},
|
| 489 |
-
"32056": {
|
| 490 |
-
"content": "<extra_id_43>",
|
| 491 |
-
"lstrip": false,
|
| 492 |
-
"normalized": false,
|
| 493 |
-
"rstrip": false,
|
| 494 |
-
"single_word": false,
|
| 495 |
-
"special": true
|
| 496 |
-
},
|
| 497 |
-
"32057": {
|
| 498 |
-
"content": "<extra_id_42>",
|
| 499 |
-
"lstrip": false,
|
| 500 |
-
"normalized": false,
|
| 501 |
-
"rstrip": false,
|
| 502 |
-
"single_word": false,
|
| 503 |
-
"special": true
|
| 504 |
-
},
|
| 505 |
-
"32058": {
|
| 506 |
-
"content": "<extra_id_41>",
|
| 507 |
-
"lstrip": false,
|
| 508 |
-
"normalized": false,
|
| 509 |
-
"rstrip": false,
|
| 510 |
-
"single_word": false,
|
| 511 |
-
"special": true
|
| 512 |
-
},
|
| 513 |
-
"32059": {
|
| 514 |
-
"content": "<extra_id_40>",
|
| 515 |
-
"lstrip": false,
|
| 516 |
-
"normalized": false,
|
| 517 |
-
"rstrip": false,
|
| 518 |
-
"single_word": false,
|
| 519 |
-
"special": true
|
| 520 |
-
},
|
| 521 |
-
"32060": {
|
| 522 |
-
"content": "<extra_id_39>",
|
| 523 |
-
"lstrip": false,
|
| 524 |
-
"normalized": false,
|
| 525 |
-
"rstrip": false,
|
| 526 |
-
"single_word": false,
|
| 527 |
-
"special": true
|
| 528 |
-
},
|
| 529 |
-
"32061": {
|
| 530 |
-
"content": "<extra_id_38>",
|
| 531 |
-
"lstrip": false,
|
| 532 |
-
"normalized": false,
|
| 533 |
-
"rstrip": false,
|
| 534 |
-
"single_word": false,
|
| 535 |
-
"special": true
|
| 536 |
-
},
|
| 537 |
-
"32062": {
|
| 538 |
-
"content": "<extra_id_37>",
|
| 539 |
-
"lstrip": false,
|
| 540 |
-
"normalized": false,
|
| 541 |
-
"rstrip": false,
|
| 542 |
-
"single_word": false,
|
| 543 |
-
"special": true
|
| 544 |
-
},
|
| 545 |
-
"32063": {
|
| 546 |
-
"content": "<extra_id_36>",
|
| 547 |
-
"lstrip": false,
|
| 548 |
-
"normalized": false,
|
| 549 |
-
"rstrip": false,
|
| 550 |
-
"single_word": false,
|
| 551 |
-
"special": true
|
| 552 |
-
},
|
| 553 |
-
"32064": {
|
| 554 |
-
"content": "<extra_id_35>",
|
| 555 |
-
"lstrip": false,
|
| 556 |
-
"normalized": false,
|
| 557 |
-
"rstrip": false,
|
| 558 |
-
"single_word": false,
|
| 559 |
-
"special": true
|
| 560 |
-
},
|
| 561 |
-
"32065": {
|
| 562 |
-
"content": "<extra_id_34>",
|
| 563 |
-
"lstrip": false,
|
| 564 |
-
"normalized": false,
|
| 565 |
-
"rstrip": false,
|
| 566 |
-
"single_word": false,
|
| 567 |
-
"special": true
|
| 568 |
-
},
|
| 569 |
-
"32066": {
|
| 570 |
-
"content": "<extra_id_33>",
|
| 571 |
-
"lstrip": false,
|
| 572 |
-
"normalized": false,
|
| 573 |
-
"rstrip": false,
|
| 574 |
-
"single_word": false,
|
| 575 |
-
"special": true
|
| 576 |
-
},
|
| 577 |
-
"32067": {
|
| 578 |
-
"content": "<extra_id_32>",
|
| 579 |
-
"lstrip": false,
|
| 580 |
-
"normalized": false,
|
| 581 |
-
"rstrip": false,
|
| 582 |
-
"single_word": false,
|
| 583 |
-
"special": true
|
| 584 |
-
},
|
| 585 |
-
"32068": {
|
| 586 |
-
"content": "<extra_id_31>",
|
| 587 |
-
"lstrip": false,
|
| 588 |
-
"normalized": false,
|
| 589 |
-
"rstrip": false,
|
| 590 |
-
"single_word": false,
|
| 591 |
-
"special": true
|
| 592 |
-
},
|
| 593 |
-
"32069": {
|
| 594 |
-
"content": "<extra_id_30>",
|
| 595 |
-
"lstrip": false,
|
| 596 |
-
"normalized": false,
|
| 597 |
-
"rstrip": false,
|
| 598 |
-
"single_word": false,
|
| 599 |
-
"special": true
|
| 600 |
-
},
|
| 601 |
-
"32070": {
|
| 602 |
-
"content": "<extra_id_29>",
|
| 603 |
-
"lstrip": false,
|
| 604 |
-
"normalized": false,
|
| 605 |
-
"rstrip": false,
|
| 606 |
-
"single_word": false,
|
| 607 |
-
"special": true
|
| 608 |
-
},
|
| 609 |
-
"32071": {
|
| 610 |
-
"content": "<extra_id_28>",
|
| 611 |
-
"lstrip": false,
|
| 612 |
-
"normalized": false,
|
| 613 |
-
"rstrip": false,
|
| 614 |
-
"single_word": false,
|
| 615 |
-
"special": true
|
| 616 |
-
},
|
| 617 |
-
"32072": {
|
| 618 |
-
"content": "<extra_id_27>",
|
| 619 |
-
"lstrip": false,
|
| 620 |
-
"normalized": false,
|
| 621 |
-
"rstrip": false,
|
| 622 |
-
"single_word": false,
|
| 623 |
-
"special": true
|
| 624 |
-
},
|
| 625 |
-
"32073": {
|
| 626 |
-
"content": "<extra_id_26>",
|
| 627 |
-
"lstrip": false,
|
| 628 |
-
"normalized": false,
|
| 629 |
-
"rstrip": false,
|
| 630 |
-
"single_word": false,
|
| 631 |
-
"special": true
|
| 632 |
-
},
|
| 633 |
-
"32074": {
|
| 634 |
-
"content": "<extra_id_25>",
|
| 635 |
-
"lstrip": false,
|
| 636 |
-
"normalized": false,
|
| 637 |
-
"rstrip": false,
|
| 638 |
-
"single_word": false,
|
| 639 |
-
"special": true
|
| 640 |
-
},
|
| 641 |
-
"32075": {
|
| 642 |
-
"content": "<extra_id_24>",
|
| 643 |
-
"lstrip": false,
|
| 644 |
-
"normalized": false,
|
| 645 |
-
"rstrip": false,
|
| 646 |
-
"single_word": false,
|
| 647 |
-
"special": true
|
| 648 |
-
},
|
| 649 |
-
"32076": {
|
| 650 |
-
"content": "<extra_id_23>",
|
| 651 |
-
"lstrip": false,
|
| 652 |
-
"normalized": false,
|
| 653 |
-
"rstrip": false,
|
| 654 |
-
"single_word": false,
|
| 655 |
-
"special": true
|
| 656 |
-
},
|
| 657 |
-
"32077": {
|
| 658 |
-
"content": "<extra_id_22>",
|
| 659 |
-
"lstrip": false,
|
| 660 |
-
"normalized": false,
|
| 661 |
-
"rstrip": false,
|
| 662 |
-
"single_word": false,
|
| 663 |
-
"special": true
|
| 664 |
-
},
|
| 665 |
-
"32078": {
|
| 666 |
-
"content": "<extra_id_21>",
|
| 667 |
-
"lstrip": false,
|
| 668 |
-
"normalized": false,
|
| 669 |
-
"rstrip": false,
|
| 670 |
-
"single_word": false,
|
| 671 |
-
"special": true
|
| 672 |
-
},
|
| 673 |
-
"32079": {
|
| 674 |
-
"content": "<extra_id_20>",
|
| 675 |
-
"lstrip": false,
|
| 676 |
-
"normalized": false,
|
| 677 |
-
"rstrip": false,
|
| 678 |
-
"single_word": false,
|
| 679 |
-
"special": true
|
| 680 |
-
},
|
| 681 |
-
"32080": {
|
| 682 |
-
"content": "<extra_id_19>",
|
| 683 |
-
"lstrip": false,
|
| 684 |
-
"normalized": false,
|
| 685 |
-
"rstrip": false,
|
| 686 |
-
"single_word": false,
|
| 687 |
-
"special": true
|
| 688 |
-
},
|
| 689 |
-
"32081": {
|
| 690 |
-
"content": "<extra_id_18>",
|
| 691 |
-
"lstrip": false,
|
| 692 |
-
"normalized": false,
|
| 693 |
-
"rstrip": false,
|
| 694 |
-
"single_word": false,
|
| 695 |
-
"special": true
|
| 696 |
-
},
|
| 697 |
-
"32082": {
|
| 698 |
-
"content": "<extra_id_17>",
|
| 699 |
-
"lstrip": false,
|
| 700 |
-
"normalized": false,
|
| 701 |
-
"rstrip": false,
|
| 702 |
-
"single_word": false,
|
| 703 |
-
"special": true
|
| 704 |
-
},
|
| 705 |
-
"32083": {
|
| 706 |
-
"content": "<extra_id_16>",
|
| 707 |
-
"lstrip": false,
|
| 708 |
-
"normalized": false,
|
| 709 |
-
"rstrip": false,
|
| 710 |
-
"single_word": false,
|
| 711 |
-
"special": true
|
| 712 |
-
},
|
| 713 |
-
"32084": {
|
| 714 |
-
"content": "<extra_id_15>",
|
| 715 |
-
"lstrip": false,
|
| 716 |
-
"normalized": false,
|
| 717 |
-
"rstrip": false,
|
| 718 |
-
"single_word": false,
|
| 719 |
-
"special": true
|
| 720 |
-
},
|
| 721 |
-
"32085": {
|
| 722 |
-
"content": "<extra_id_14>",
|
| 723 |
-
"lstrip": false,
|
| 724 |
-
"normalized": false,
|
| 725 |
-
"rstrip": false,
|
| 726 |
-
"single_word": false,
|
| 727 |
-
"special": true
|
| 728 |
-
},
|
| 729 |
-
"32086": {
|
| 730 |
-
"content": "<extra_id_13>",
|
| 731 |
-
"lstrip": false,
|
| 732 |
-
"normalized": false,
|
| 733 |
-
"rstrip": false,
|
| 734 |
-
"single_word": false,
|
| 735 |
-
"special": true
|
| 736 |
-
},
|
| 737 |
-
"32087": {
|
| 738 |
-
"content": "<extra_id_12>",
|
| 739 |
-
"lstrip": false,
|
| 740 |
-
"normalized": false,
|
| 741 |
-
"rstrip": false,
|
| 742 |
-
"single_word": false,
|
| 743 |
-
"special": true
|
| 744 |
-
},
|
| 745 |
-
"32088": {
|
| 746 |
-
"content": "<extra_id_11>",
|
| 747 |
-
"lstrip": false,
|
| 748 |
-
"normalized": false,
|
| 749 |
-
"rstrip": false,
|
| 750 |
-
"single_word": false,
|
| 751 |
-
"special": true
|
| 752 |
-
},
|
| 753 |
-
"32089": {
|
| 754 |
-
"content": "<extra_id_10>",
|
| 755 |
-
"lstrip": false,
|
| 756 |
-
"normalized": false,
|
| 757 |
-
"rstrip": false,
|
| 758 |
-
"single_word": false,
|
| 759 |
-
"special": true
|
| 760 |
-
},
|
| 761 |
-
"32090": {
|
| 762 |
-
"content": "<extra_id_9>",
|
| 763 |
-
"lstrip": false,
|
| 764 |
-
"normalized": false,
|
| 765 |
-
"rstrip": false,
|
| 766 |
-
"single_word": false,
|
| 767 |
-
"special": true
|
| 768 |
-
},
|
| 769 |
-
"32091": {
|
| 770 |
-
"content": "<extra_id_8>",
|
| 771 |
-
"lstrip": false,
|
| 772 |
-
"normalized": false,
|
| 773 |
-
"rstrip": false,
|
| 774 |
-
"single_word": false,
|
| 775 |
-
"special": true
|
| 776 |
-
},
|
| 777 |
-
"32092": {
|
| 778 |
-
"content": "<extra_id_7>",
|
| 779 |
-
"lstrip": false,
|
| 780 |
-
"normalized": false,
|
| 781 |
-
"rstrip": false,
|
| 782 |
-
"single_word": false,
|
| 783 |
-
"special": true
|
| 784 |
-
},
|
| 785 |
-
"32093": {
|
| 786 |
-
"content": "<extra_id_6>",
|
| 787 |
-
"lstrip": false,
|
| 788 |
-
"normalized": false,
|
| 789 |
-
"rstrip": false,
|
| 790 |
-
"single_word": false,
|
| 791 |
-
"special": true
|
| 792 |
-
},
|
| 793 |
-
"32094": {
|
| 794 |
-
"content": "<extra_id_5>",
|
| 795 |
-
"lstrip": false,
|
| 796 |
-
"normalized": false,
|
| 797 |
-
"rstrip": false,
|
| 798 |
-
"single_word": false,
|
| 799 |
-
"special": true
|
| 800 |
-
},
|
| 801 |
-
"32095": {
|
| 802 |
-
"content": "<extra_id_4>",
|
| 803 |
-
"lstrip": false,
|
| 804 |
-
"normalized": false,
|
| 805 |
-
"rstrip": false,
|
| 806 |
-
"single_word": false,
|
| 807 |
-
"special": true
|
| 808 |
-
},
|
| 809 |
-
"32096": {
|
| 810 |
-
"content": "<extra_id_3>",
|
| 811 |
-
"lstrip": false,
|
| 812 |
-
"normalized": false,
|
| 813 |
-
"rstrip": false,
|
| 814 |
-
"single_word": false,
|
| 815 |
-
"special": true
|
| 816 |
-
},
|
| 817 |
-
"32097": {
|
| 818 |
-
"content": "<extra_id_2>",
|
| 819 |
-
"lstrip": false,
|
| 820 |
-
"normalized": false,
|
| 821 |
-
"rstrip": false,
|
| 822 |
-
"single_word": false,
|
| 823 |
-
"special": true
|
| 824 |
-
},
|
| 825 |
-
"32098": {
|
| 826 |
-
"content": "<extra_id_1>",
|
| 827 |
-
"lstrip": false,
|
| 828 |
-
"normalized": false,
|
| 829 |
-
"rstrip": false,
|
| 830 |
-
"single_word": false,
|
| 831 |
-
"special": true
|
| 832 |
-
},
|
| 833 |
-
"32099": {
|
| 834 |
-
"content": "<extra_id_0>",
|
| 835 |
-
"lstrip": false,
|
| 836 |
-
"normalized": false,
|
| 837 |
-
"rstrip": false,
|
| 838 |
-
"single_word": false,
|
| 839 |
-
"special": true
|
| 840 |
-
},
|
| 841 |
-
"32100": {
|
| 842 |
-
"content": "happy",
|
| 843 |
-
"lstrip": false,
|
| 844 |
-
"normalized": true,
|
| 845 |
-
"rstrip": false,
|
| 846 |
-
"single_word": false,
|
| 847 |
-
"special": false
|
| 848 |
-
},
|
| 849 |
-
"32101": {
|
| 850 |
-
"content": "intelligent",
|
| 851 |
-
"lstrip": false,
|
| 852 |
-
"normalized": true,
|
| 853 |
-
"rstrip": false,
|
| 854 |
-
"single_word": false,
|
| 855 |
-
"special": false
|
| 856 |
-
},
|
| 857 |
-
"32102": {
|
| 858 |
-
"content": "response",
|
| 859 |
-
"lstrip": false,
|
| 860 |
-
"normalized": true,
|
| 861 |
-
"rstrip": false,
|
| 862 |
-
"single_word": false,
|
| 863 |
-
"special": false
|
| 864 |
-
},
|
| 865 |
-
"32103": {
|
| 866 |
-
"content": "environment",
|
| 867 |
-
"lstrip": false,
|
| 868 |
-
"normalized": true,
|
| 869 |
-
"rstrip": false,
|
| 870 |
-
"single_word": false,
|
| 871 |
-
"special": false
|
| 872 |
-
},
|
| 873 |
-
"32104": {
|
| 874 |
-
"content": "amazeballs",
|
| 875 |
-
"lstrip": false,
|
| 876 |
-
"normalized": true,
|
| 877 |
-
"rstrip": false,
|
| 878 |
-
"single_word": false,
|
| 879 |
-
"special": false
|
| 880 |
-
},
|
| 881 |
-
"32105": {
|
| 882 |
-
"content": "cryptocurrency",
|
| 883 |
-
"lstrip": false,
|
| 884 |
-
"normalized": true,
|
| 885 |
-
"rstrip": false,
|
| 886 |
-
"single_word": false,
|
| 887 |
-
"special": false
|
| 888 |
-
},
|
| 889 |
-
"32106": {
|
| 890 |
-
"content": "webinar",
|
| 891 |
-
"lstrip": false,
|
| 892 |
-
"normalized": true,
|
| 893 |
-
"rstrip": false,
|
| 894 |
-
"single_word": false,
|
| 895 |
-
"special": false
|
| 896 |
-
},
|
| 897 |
-
"32107": {
|
| 898 |
-
"content": "vlog",
|
| 899 |
-
"lstrip": false,
|
| 900 |
-
"normalized": true,
|
| 901 |
-
"rstrip": false,
|
| 902 |
-
"single_word": false,
|
| 903 |
-
"special": false
|
| 904 |
-
},
|
| 905 |
-
"32108": {
|
| 906 |
-
"content": "upcycle",
|
| 907 |
-
"lstrip": false,
|
| 908 |
-
"normalized": true,
|
| 909 |
-
"rstrip": false,
|
| 910 |
-
"single_word": false,
|
| 911 |
-
"special": false
|
| 912 |
-
},
|
| 913 |
-
"32109": {
|
| 914 |
-
"content": "photobomb",
|
| 915 |
-
"lstrip": false,
|
| 916 |
-
"normalized": true,
|
| 917 |
-
"rstrip": false,
|
| 918 |
-
"single_word": false,
|
| 919 |
-
"special": false
|
| 920 |
-
},
|
| 921 |
-
"32110": {
|
| 922 |
-
"content": "facepalm",
|
| 923 |
-
"lstrip": false,
|
| 924 |
-
"normalized": true,
|
| 925 |
-
"rstrip": false,
|
| 926 |
-
"single_word": false,
|
| 927 |
-
"special": false
|
| 928 |
-
},
|
| 929 |
-
"32111": {
|
| 930 |
-
"content": "crowdfunding",
|
| 931 |
-
"lstrip": false,
|
| 932 |
-
"normalized": true,
|
| 933 |
-
"rstrip": false,
|
| 934 |
-
"single_word": false,
|
| 935 |
-
"special": false
|
| 936 |
-
},
|
| 937 |
-
"32112": {
|
| 938 |
-
"content": "bromance",
|
| 939 |
-
"lstrip": false,
|
| 940 |
-
"normalized": true,
|
| 941 |
-
"rstrip": false,
|
| 942 |
-
"single_word": false,
|
| 943 |
-
"special": false
|
| 944 |
-
},
|
| 945 |
-
"32113": {
|
| 946 |
-
"content": "hangry",
|
| 947 |
-
"lstrip": false,
|
| 948 |
-
"normalized": true,
|
| 949 |
-
"rstrip": false,
|
| 950 |
-
"single_word": false,
|
| 951 |
-
"special": false
|
| 952 |
-
},
|
| 953 |
-
"32114": {
|
| 954 |
-
"content": "[PAD]",
|
| 955 |
-
"lstrip": false,
|
| 956 |
-
"normalized": false,
|
| 957 |
-
"rstrip": false,
|
| 958 |
-
"single_word": false,
|
| 959 |
"special": true
|
| 960 |
}
|
| 961 |
},
|
| 962 |
-
"
|
| 963 |
-
"<extra_id_0>",
|
| 964 |
-
"<extra_id_1>",
|
| 965 |
-
"<extra_id_2>",
|
| 966 |
-
"<extra_id_3>",
|
| 967 |
-
"<extra_id_4>",
|
| 968 |
-
"<extra_id_5>",
|
| 969 |
-
"<extra_id_6>",
|
| 970 |
-
"<extra_id_7>",
|
| 971 |
-
"<extra_id_8>",
|
| 972 |
-
"<extra_id_9>",
|
| 973 |
-
"<extra_id_10>",
|
| 974 |
-
"<extra_id_11>",
|
| 975 |
-
"<extra_id_12>",
|
| 976 |
-
"<extra_id_13>",
|
| 977 |
-
"<extra_id_14>",
|
| 978 |
-
"<extra_id_15>",
|
| 979 |
-
"<extra_id_16>",
|
| 980 |
-
"<extra_id_17>",
|
| 981 |
-
"<extra_id_18>",
|
| 982 |
-
"<extra_id_19>",
|
| 983 |
-
"<extra_id_20>",
|
| 984 |
-
"<extra_id_21>",
|
| 985 |
-
"<extra_id_22>",
|
| 986 |
-
"<extra_id_23>",
|
| 987 |
-
"<extra_id_24>",
|
| 988 |
-
"<extra_id_25>",
|
| 989 |
-
"<extra_id_26>",
|
| 990 |
-
"<extra_id_27>",
|
| 991 |
-
"<extra_id_28>",
|
| 992 |
-
"<extra_id_29>",
|
| 993 |
-
"<extra_id_30>",
|
| 994 |
-
"<extra_id_31>",
|
| 995 |
-
"<extra_id_32>",
|
| 996 |
-
"<extra_id_33>",
|
| 997 |
-
"<extra_id_34>",
|
| 998 |
-
"<extra_id_35>",
|
| 999 |
-
"<extra_id_36>",
|
| 1000 |
-
"<extra_id_37>",
|
| 1001 |
-
"<extra_id_38>",
|
| 1002 |
-
"<extra_id_39>",
|
| 1003 |
-
"<extra_id_40>",
|
| 1004 |
-
"<extra_id_41>",
|
| 1005 |
-
"<extra_id_42>",
|
| 1006 |
-
"<extra_id_43>",
|
| 1007 |
-
"<extra_id_44>",
|
| 1008 |
-
"<extra_id_45>",
|
| 1009 |
-
"<extra_id_46>",
|
| 1010 |
-
"<extra_id_47>",
|
| 1011 |
-
"<extra_id_48>",
|
| 1012 |
-
"<extra_id_49>",
|
| 1013 |
-
"<extra_id_50>",
|
| 1014 |
-
"<extra_id_51>",
|
| 1015 |
-
"<extra_id_52>",
|
| 1016 |
-
"<extra_id_53>",
|
| 1017 |
-
"<extra_id_54>",
|
| 1018 |
-
"<extra_id_55>",
|
| 1019 |
-
"<extra_id_56>",
|
| 1020 |
-
"<extra_id_57>",
|
| 1021 |
-
"<extra_id_58>",
|
| 1022 |
-
"<extra_id_59>",
|
| 1023 |
-
"<extra_id_60>",
|
| 1024 |
-
"<extra_id_61>",
|
| 1025 |
-
"<extra_id_62>",
|
| 1026 |
-
"<extra_id_63>",
|
| 1027 |
-
"<extra_id_64>",
|
| 1028 |
-
"<extra_id_65>",
|
| 1029 |
-
"<extra_id_66>",
|
| 1030 |
-
"<extra_id_67>",
|
| 1031 |
-
"<extra_id_68>",
|
| 1032 |
-
"<extra_id_69>",
|
| 1033 |
-
"<extra_id_70>",
|
| 1034 |
-
"<extra_id_71>",
|
| 1035 |
-
"<extra_id_72>",
|
| 1036 |
-
"<extra_id_73>",
|
| 1037 |
-
"<extra_id_74>",
|
| 1038 |
-
"<extra_id_75>",
|
| 1039 |
-
"<extra_id_76>",
|
| 1040 |
-
"<extra_id_77>",
|
| 1041 |
-
"<extra_id_78>",
|
| 1042 |
-
"<extra_id_79>",
|
| 1043 |
-
"<extra_id_80>",
|
| 1044 |
-
"<extra_id_81>",
|
| 1045 |
-
"<extra_id_82>",
|
| 1046 |
-
"<extra_id_83>",
|
| 1047 |
-
"<extra_id_84>",
|
| 1048 |
-
"<extra_id_85>",
|
| 1049 |
-
"<extra_id_86>",
|
| 1050 |
-
"<extra_id_87>",
|
| 1051 |
-
"<extra_id_88>",
|
| 1052 |
-
"<extra_id_89>",
|
| 1053 |
-
"<extra_id_90>",
|
| 1054 |
-
"<extra_id_91>",
|
| 1055 |
-
"<extra_id_92>",
|
| 1056 |
-
"<extra_id_93>",
|
| 1057 |
-
"<extra_id_94>",
|
| 1058 |
-
"<extra_id_95>",
|
| 1059 |
-
"<extra_id_96>",
|
| 1060 |
-
"<extra_id_97>",
|
| 1061 |
-
"<extra_id_98>",
|
| 1062 |
-
"<extra_id_99>"
|
| 1063 |
-
],
|
| 1064 |
-
"bpe_dropout": 0.1,
|
| 1065 |
"clean_up_tokenization_spaces": true,
|
| 1066 |
-
"
|
| 1067 |
-
"eos_token": "</s>",
|
| 1068 |
-
"extra_ids": 100,
|
| 1069 |
-
"max_length": 1024,
|
| 1070 |
"model_max_length": 1024,
|
| 1071 |
-
"
|
| 1072 |
-
|
| 1073 |
-
|
| 1074 |
-
"nmt_normalization": true,
|
| 1075 |
-
"pre_tokenization": {
|
| 1076 |
-
"punctuation_split": true,
|
| 1077 |
-
"split_digits": true
|
| 1078 |
-
},
|
| 1079 |
-
"strip_accents": true
|
| 1080 |
-
},
|
| 1081 |
-
"pad_to_multiple_of": null,
|
| 1082 |
-
"pad_token": "</s>",
|
| 1083 |
-
"pad_token_type_id": 0,
|
| 1084 |
-
"padding_side": "right",
|
| 1085 |
-
"spe_tokenization": {
|
| 1086 |
-
"coverage": 0.9995,
|
| 1087 |
-
"enable": true,
|
| 1088 |
-
"nbest_size": 64
|
| 1089 |
-
},
|
| 1090 |
-
"stride": 0,
|
| 1091 |
-
"subword_regularization": {
|
| 1092 |
-
"alpha": 0.1,
|
| 1093 |
-
"dropout": 0.1,
|
| 1094 |
-
"enable": true
|
| 1095 |
-
},
|
| 1096 |
-
"tokenizer_class": "T5Tokenizer",
|
| 1097 |
-
"truncation_side": "right",
|
| 1098 |
-
"truncation_strategy": "longest_first",
|
| 1099 |
-
"unk_token": "<unk>"
|
| 1100 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"add_prefix_space": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"lstrip": false,
|
| 7 |
"normalized": true,
|
| 8 |
"rstrip": false,
|
| 9 |
"single_word": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
"special": true
|
| 11 |
}
|
| 12 |
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
"clean_up_tokenization_spaces": true,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
|
|
|
|
|
|
|
|
|
| 16 |
"model_max_length": 1024,
|
| 17 |
+
"pad_token": "<|pad|>",
|
| 18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 19 |
+
"unk_token": "<|endoftext|>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
}
|