Add tiny processor fixture for testing
Browse files- config.json +1 -1
- tokenizer.json +2 -2
- tokenizer_config.json +135 -127
config.json
CHANGED
|
@@ -40,5 +40,5 @@
|
|
| 40 |
},
|
| 41 |
"vision_feature_layer": -1,
|
| 42 |
"vision_feature_select_strategy": "full",
|
| 43 |
-
"vocab_size":
|
| 44 |
}
|
|
|
|
| 40 |
},
|
| 41 |
"vision_feature_layer": -1,
|
| 42 |
"vision_feature_select_strategy": "full",
|
| 43 |
+
"vocab_size": 500
|
| 44 |
}
|
tokenizer.json
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d87e646c8d6d0f0fc7374c9024524f686bcabd121b111b23521651e6d752308
|
| 3 |
+
size 20266
|
tokenizer_config.json
CHANGED
|
@@ -5,363 +5,371 @@
|
|
| 5 |
"added_tokens_decoder": {
|
| 6 |
"0": {
|
| 7 |
"content": "<PAD>",
|
|
|
|
| 8 |
"lstrip": false,
|
| 9 |
-
"normalized": false,
|
| 10 |
"rstrip": false,
|
| 11 |
-
"
|
| 12 |
"special": true
|
| 13 |
},
|
| 14 |
"1": {
|
| 15 |
"content": "<UNK>",
|
|
|
|
| 16 |
"lstrip": false,
|
| 17 |
-
"normalized": false,
|
| 18 |
"rstrip": false,
|
| 19 |
-
"
|
| 20 |
"special": true
|
| 21 |
},
|
| 22 |
"2": {
|
| 23 |
"content": "<CLS>",
|
|
|
|
| 24 |
"lstrip": false,
|
| 25 |
-
"normalized": false,
|
| 26 |
"rstrip": false,
|
| 27 |
-
"
|
| 28 |
"special": true
|
| 29 |
},
|
| 30 |
"3": {
|
| 31 |
"content": "<SEP>",
|
|
|
|
| 32 |
"lstrip": false,
|
| 33 |
-
"normalized": false,
|
| 34 |
"rstrip": false,
|
| 35 |
-
"
|
| 36 |
"special": true
|
| 37 |
},
|
| 38 |
"4": {
|
| 39 |
"content": "<MASK_TOKEN>",
|
|
|
|
| 40 |
"lstrip": false,
|
| 41 |
-
"normalized": false,
|
| 42 |
"rstrip": false,
|
| 43 |
-
"
|
| 44 |
"special": true
|
| 45 |
},
|
| 46 |
"5": {
|
| 47 |
"content": "<BOS_TOKEN>",
|
|
|
|
| 48 |
"lstrip": false,
|
| 49 |
-
"normalized": false,
|
| 50 |
"rstrip": false,
|
| 51 |
-
"
|
| 52 |
"special": true
|
| 53 |
},
|
| 54 |
"6": {
|
| 55 |
"content": "<EOS_TOKEN>",
|
|
|
|
| 56 |
"lstrip": false,
|
| 57 |
-
"normalized": false,
|
| 58 |
"rstrip": false,
|
| 59 |
-
"
|
| 60 |
"special": true
|
| 61 |
},
|
| 62 |
"7": {
|
| 63 |
"content": "<EOP_TOKEN>",
|
|
|
|
| 64 |
"lstrip": false,
|
| 65 |
-
"normalized": false,
|
| 66 |
"rstrip": false,
|
| 67 |
-
"
|
| 68 |
"special": true
|
| 69 |
},
|
| 70 |
-
"
|
| 71 |
"content": "<|START_OF_TURN_TOKEN|>",
|
|
|
|
| 72 |
"lstrip": false,
|
| 73 |
-
"normalized": false,
|
| 74 |
"rstrip": false,
|
| 75 |
-
"
|
| 76 |
"special": false
|
| 77 |
},
|
| 78 |
-
"
|
| 79 |
"content": "<|END_OF_TURN_TOKEN|>",
|
|
|
|
| 80 |
"lstrip": false,
|
| 81 |
-
"normalized": false,
|
| 82 |
"rstrip": false,
|
| 83 |
-
"
|
| 84 |
"special": true
|
| 85 |
},
|
| 86 |
-
"
|
| 87 |
"content": "<|YES_TOKEN|>",
|
|
|
|
| 88 |
"lstrip": false,
|
| 89 |
-
"normalized": false,
|
| 90 |
"rstrip": false,
|
| 91 |
-
"
|
| 92 |
"special": false
|
| 93 |
},
|
| 94 |
-
"
|
| 95 |
"content": "<|NO_TOKEN|>",
|
|
|
|
| 96 |
"lstrip": false,
|
| 97 |
-
"normalized": false,
|
| 98 |
"rstrip": false,
|
| 99 |
-
"
|
| 100 |
"special": false
|
| 101 |
},
|
| 102 |
-
"
|
| 103 |
"content": "<|GOOD_TOKEN|>",
|
|
|
|
| 104 |
"lstrip": false,
|
| 105 |
-
"normalized": false,
|
| 106 |
"rstrip": false,
|
| 107 |
-
"
|
| 108 |
"special": false
|
| 109 |
},
|
| 110 |
-
"
|
| 111 |
"content": "<|BAD_TOKEN|>",
|
|
|
|
| 112 |
"lstrip": false,
|
| 113 |
-
"normalized": false,
|
| 114 |
"rstrip": false,
|
| 115 |
-
"
|
| 116 |
"special": false
|
| 117 |
},
|
| 118 |
-
"
|
| 119 |
"content": "<|USER_TOKEN|>",
|
|
|
|
| 120 |
"lstrip": false,
|
| 121 |
-
"normalized": false,
|
| 122 |
"rstrip": false,
|
| 123 |
-
"
|
| 124 |
"special": false
|
| 125 |
},
|
| 126 |
-
"
|
| 127 |
"content": "<|CHATBOT_TOKEN|>",
|
|
|
|
| 128 |
"lstrip": false,
|
| 129 |
-
"normalized": false,
|
| 130 |
"rstrip": false,
|
| 131 |
-
"
|
| 132 |
"special": false
|
| 133 |
},
|
| 134 |
-
"
|
| 135 |
"content": "<|SYSTEM_TOKEN|>",
|
|
|
|
| 136 |
"lstrip": false,
|
| 137 |
-
"normalized": false,
|
| 138 |
"rstrip": false,
|
| 139 |
-
"
|
| 140 |
"special": false
|
| 141 |
},
|
| 142 |
-
"
|
| 143 |
"content": "<|USER_0_TOKEN|>",
|
|
|
|
| 144 |
"lstrip": false,
|
| 145 |
-
"normalized": false,
|
| 146 |
"rstrip": false,
|
| 147 |
-
"
|
| 148 |
"special": false
|
| 149 |
},
|
| 150 |
-
"
|
| 151 |
"content": "<|USER_1_TOKEN|>",
|
|
|
|
| 152 |
"lstrip": false,
|
| 153 |
-
"normalized": false,
|
| 154 |
"rstrip": false,
|
| 155 |
-
"
|
| 156 |
"special": false
|
| 157 |
},
|
| 158 |
-
"
|
| 159 |
"content": "<|USER_2_TOKEN|>",
|
|
|
|
| 160 |
"lstrip": false,
|
| 161 |
-
"normalized": false,
|
| 162 |
"rstrip": false,
|
| 163 |
-
"
|
| 164 |
"special": false
|
| 165 |
},
|
| 166 |
-
"
|
| 167 |
"content": "<|USER_3_TOKEN|>",
|
|
|
|
| 168 |
"lstrip": false,
|
| 169 |
-
"normalized": false,
|
| 170 |
"rstrip": false,
|
| 171 |
-
"
|
| 172 |
"special": false
|
| 173 |
},
|
| 174 |
-
"
|
| 175 |
"content": "<|USER_4_TOKEN|>",
|
|
|
|
| 176 |
"lstrip": false,
|
| 177 |
-
"normalized": false,
|
| 178 |
"rstrip": false,
|
| 179 |
-
"
|
| 180 |
"special": false
|
| 181 |
},
|
| 182 |
-
"
|
| 183 |
"content": "<|USER_5_TOKEN|>",
|
|
|
|
| 184 |
"lstrip": false,
|
| 185 |
-
"normalized": false,
|
| 186 |
"rstrip": false,
|
| 187 |
-
"
|
| 188 |
"special": false
|
| 189 |
},
|
| 190 |
-
"
|
| 191 |
"content": "<|USER_6_TOKEN|>",
|
|
|
|
| 192 |
"lstrip": false,
|
| 193 |
-
"normalized": false,
|
| 194 |
"rstrip": false,
|
| 195 |
-
"
|
| 196 |
"special": false
|
| 197 |
},
|
| 198 |
-
"
|
| 199 |
"content": "<|USER_7_TOKEN|>",
|
|
|
|
| 200 |
"lstrip": false,
|
| 201 |
-
"normalized": false,
|
| 202 |
"rstrip": false,
|
| 203 |
-
"
|
| 204 |
"special": false
|
| 205 |
},
|
| 206 |
-
"
|
| 207 |
"content": "<|USER_8_TOKEN|>",
|
|
|
|
| 208 |
"lstrip": false,
|
| 209 |
-
"normalized": false,
|
| 210 |
"rstrip": false,
|
| 211 |
-
"
|
| 212 |
"special": false
|
| 213 |
},
|
| 214 |
-
"
|
| 215 |
"content": "<|USER_9_TOKEN|>",
|
|
|
|
| 216 |
"lstrip": false,
|
| 217 |
-
"normalized": false,
|
| 218 |
"rstrip": false,
|
| 219 |
-
"
|
| 220 |
"special": false
|
| 221 |
},
|
| 222 |
-
"
|
| 223 |
"content": "<|START_THINKING|>",
|
|
|
|
| 224 |
"lstrip": false,
|
| 225 |
-
"normalized": false,
|
| 226 |
"rstrip": false,
|
| 227 |
-
"
|
| 228 |
"special": false
|
| 229 |
},
|
| 230 |
-
"
|
| 231 |
"content": "<|END_THINKING|>",
|
|
|
|
| 232 |
"lstrip": false,
|
| 233 |
-
"normalized": false,
|
| 234 |
"rstrip": false,
|
| 235 |
-
"
|
| 236 |
"special": false
|
| 237 |
},
|
| 238 |
-
"
|
| 239 |
"content": "<|START_RESPONSE|>",
|
|
|
|
| 240 |
"lstrip": false,
|
| 241 |
-
"normalized": false,
|
| 242 |
"rstrip": false,
|
| 243 |
-
"
|
| 244 |
"special": true
|
| 245 |
},
|
| 246 |
-
"
|
| 247 |
"content": "<|END_RESPONSE|>",
|
|
|
|
| 248 |
"lstrip": false,
|
| 249 |
-
"normalized": false,
|
| 250 |
"rstrip": false,
|
| 251 |
-
"
|
| 252 |
"special": true
|
| 253 |
},
|
| 254 |
-
"
|
| 255 |
"content": "<|START_ACTION|>",
|
|
|
|
| 256 |
"lstrip": false,
|
| 257 |
-
"normalized": false,
|
| 258 |
"rstrip": false,
|
| 259 |
-
"
|
| 260 |
"special": false
|
| 261 |
},
|
| 262 |
-
"
|
| 263 |
"content": "<|END_ACTION|>",
|
|
|
|
| 264 |
"lstrip": false,
|
| 265 |
-
"normalized": false,
|
| 266 |
"rstrip": false,
|
| 267 |
-
"
|
| 268 |
"special": false
|
| 269 |
},
|
| 270 |
-
"
|
| 271 |
"content": "<|START_TOOL_RESULT|>",
|
|
|
|
| 272 |
"lstrip": false,
|
| 273 |
-
"normalized": false,
|
| 274 |
"rstrip": false,
|
| 275 |
-
"
|
| 276 |
"special": false
|
| 277 |
},
|
| 278 |
-
"
|
| 279 |
"content": "<|END_TOOL_RESULT|>",
|
|
|
|
| 280 |
"lstrip": false,
|
| 281 |
-
"normalized": false,
|
| 282 |
"rstrip": false,
|
| 283 |
-
"
|
| 284 |
"special": false
|
| 285 |
},
|
| 286 |
-
"
|
| 287 |
"content": "<|EXTRA_8_TOKEN|>",
|
|
|
|
| 288 |
"lstrip": false,
|
| 289 |
-
"normalized": false,
|
| 290 |
"rstrip": false,
|
| 291 |
-
"
|
| 292 |
"special": false
|
| 293 |
},
|
| 294 |
-
"
|
| 295 |
"content": "<|NEW_FILE|>",
|
|
|
|
| 296 |
"lstrip": false,
|
| 297 |
-
"normalized": false,
|
| 298 |
"rstrip": false,
|
| 299 |
-
"
|
| 300 |
"special": true
|
| 301 |
},
|
| 302 |
-
"
|
| 303 |
"content": "<|BEGINNING_OF_PREFIX_FIM_TOKEN|>",
|
|
|
|
| 304 |
"lstrip": false,
|
| 305 |
-
"normalized": false,
|
| 306 |
"rstrip": false,
|
| 307 |
-
"
|
| 308 |
"special": false
|
| 309 |
},
|
| 310 |
-
"
|
| 311 |
"content": "<|BEGINNING_OF_MIDDLE_FIM_TOKEN|>",
|
|
|
|
| 312 |
"lstrip": false,
|
| 313 |
-
"normalized": false,
|
| 314 |
"rstrip": false,
|
| 315 |
-
"
|
| 316 |
"special": false
|
| 317 |
},
|
| 318 |
-
"
|
| 319 |
"content": "<|BEGINNING_OF_SUFFIX_FIM_TOKEN|>",
|
|
|
|
| 320 |
"lstrip": false,
|
| 321 |
-
"normalized": false,
|
| 322 |
"rstrip": false,
|
| 323 |
-
"
|
| 324 |
"special": false
|
| 325 |
},
|
| 326 |
-
"
|
| 327 |
"content": "<|END_OF_MIDDLE_FIM_TOKEN|>",
|
|
|
|
| 328 |
"lstrip": false,
|
| 329 |
-
"normalized": false,
|
| 330 |
"rstrip": false,
|
| 331 |
-
"
|
| 332 |
"special": false
|
| 333 |
},
|
| 334 |
-
"
|
| 335 |
"content": "<|START_OF_IMG|>",
|
|
|
|
| 336 |
"lstrip": false,
|
| 337 |
-
"normalized": false,
|
| 338 |
"rstrip": false,
|
| 339 |
-
"
|
| 340 |
"special": false
|
| 341 |
},
|
| 342 |
-
"
|
| 343 |
"content": "<|END_OF_IMG|>",
|
|
|
|
| 344 |
"lstrip": false,
|
| 345 |
-
"normalized": false,
|
| 346 |
"rstrip": false,
|
| 347 |
-
"
|
| 348 |
"special": false
|
| 349 |
},
|
| 350 |
-
"
|
| 351 |
"content": "<|IMG_LINE_BREAK|>",
|
|
|
|
| 352 |
"lstrip": false,
|
| 353 |
-
"normalized": false,
|
| 354 |
"rstrip": false,
|
| 355 |
-
"
|
| 356 |
"special": false
|
| 357 |
},
|
| 358 |
-
"
|
| 359 |
"content": "<|IMG_PATCH|>",
|
|
|
|
| 360 |
"lstrip": false,
|
| 361 |
-
"normalized": false,
|
| 362 |
"rstrip": false,
|
| 363 |
-
"
|
| 364 |
"special": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 365 |
}
|
| 366 |
},
|
| 367 |
"bos_token": "<BOS_TOKEN>",
|
|
|
|
| 5 |
"added_tokens_decoder": {
|
| 6 |
"0": {
|
| 7 |
"content": "<PAD>",
|
| 8 |
+
"single_word": false,
|
| 9 |
"lstrip": false,
|
|
|
|
| 10 |
"rstrip": false,
|
| 11 |
+
"normalized": false,
|
| 12 |
"special": true
|
| 13 |
},
|
| 14 |
"1": {
|
| 15 |
"content": "<UNK>",
|
| 16 |
+
"single_word": false,
|
| 17 |
"lstrip": false,
|
|
|
|
| 18 |
"rstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
"special": true
|
| 21 |
},
|
| 22 |
"2": {
|
| 23 |
"content": "<CLS>",
|
| 24 |
+
"single_word": false,
|
| 25 |
"lstrip": false,
|
|
|
|
| 26 |
"rstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
"special": true
|
| 29 |
},
|
| 30 |
"3": {
|
| 31 |
"content": "<SEP>",
|
| 32 |
+
"single_word": false,
|
| 33 |
"lstrip": false,
|
|
|
|
| 34 |
"rstrip": false,
|
| 35 |
+
"normalized": false,
|
| 36 |
"special": true
|
| 37 |
},
|
| 38 |
"4": {
|
| 39 |
"content": "<MASK_TOKEN>",
|
| 40 |
+
"single_word": false,
|
| 41 |
"lstrip": false,
|
|
|
|
| 42 |
"rstrip": false,
|
| 43 |
+
"normalized": false,
|
| 44 |
"special": true
|
| 45 |
},
|
| 46 |
"5": {
|
| 47 |
"content": "<BOS_TOKEN>",
|
| 48 |
+
"single_word": false,
|
| 49 |
"lstrip": false,
|
|
|
|
| 50 |
"rstrip": false,
|
| 51 |
+
"normalized": false,
|
| 52 |
"special": true
|
| 53 |
},
|
| 54 |
"6": {
|
| 55 |
"content": "<EOS_TOKEN>",
|
| 56 |
+
"single_word": false,
|
| 57 |
"lstrip": false,
|
|
|
|
| 58 |
"rstrip": false,
|
| 59 |
+
"normalized": false,
|
| 60 |
"special": true
|
| 61 |
},
|
| 62 |
"7": {
|
| 63 |
"content": "<EOP_TOKEN>",
|
| 64 |
+
"single_word": false,
|
| 65 |
"lstrip": false,
|
|
|
|
| 66 |
"rstrip": false,
|
| 67 |
+
"normalized": false,
|
| 68 |
"special": true
|
| 69 |
},
|
| 70 |
+
"500": {
|
| 71 |
"content": "<|START_OF_TURN_TOKEN|>",
|
| 72 |
+
"single_word": false,
|
| 73 |
"lstrip": false,
|
|
|
|
| 74 |
"rstrip": false,
|
| 75 |
+
"normalized": false,
|
| 76 |
"special": false
|
| 77 |
},
|
| 78 |
+
"501": {
|
| 79 |
"content": "<|END_OF_TURN_TOKEN|>",
|
| 80 |
+
"single_word": false,
|
| 81 |
"lstrip": false,
|
|
|
|
| 82 |
"rstrip": false,
|
| 83 |
+
"normalized": false,
|
| 84 |
"special": true
|
| 85 |
},
|
| 86 |
+
"502": {
|
| 87 |
"content": "<|YES_TOKEN|>",
|
| 88 |
+
"single_word": false,
|
| 89 |
"lstrip": false,
|
|
|
|
| 90 |
"rstrip": false,
|
| 91 |
+
"normalized": false,
|
| 92 |
"special": false
|
| 93 |
},
|
| 94 |
+
"503": {
|
| 95 |
"content": "<|NO_TOKEN|>",
|
| 96 |
+
"single_word": false,
|
| 97 |
"lstrip": false,
|
|
|
|
| 98 |
"rstrip": false,
|
| 99 |
+
"normalized": false,
|
| 100 |
"special": false
|
| 101 |
},
|
| 102 |
+
"504": {
|
| 103 |
"content": "<|GOOD_TOKEN|>",
|
| 104 |
+
"single_word": false,
|
| 105 |
"lstrip": false,
|
|
|
|
| 106 |
"rstrip": false,
|
| 107 |
+
"normalized": false,
|
| 108 |
"special": false
|
| 109 |
},
|
| 110 |
+
"505": {
|
| 111 |
"content": "<|BAD_TOKEN|>",
|
| 112 |
+
"single_word": false,
|
| 113 |
"lstrip": false,
|
|
|
|
| 114 |
"rstrip": false,
|
| 115 |
+
"normalized": false,
|
| 116 |
"special": false
|
| 117 |
},
|
| 118 |
+
"506": {
|
| 119 |
"content": "<|USER_TOKEN|>",
|
| 120 |
+
"single_word": false,
|
| 121 |
"lstrip": false,
|
|
|
|
| 122 |
"rstrip": false,
|
| 123 |
+
"normalized": false,
|
| 124 |
"special": false
|
| 125 |
},
|
| 126 |
+
"507": {
|
| 127 |
"content": "<|CHATBOT_TOKEN|>",
|
| 128 |
+
"single_word": false,
|
| 129 |
"lstrip": false,
|
|
|
|
| 130 |
"rstrip": false,
|
| 131 |
+
"normalized": false,
|
| 132 |
"special": false
|
| 133 |
},
|
| 134 |
+
"508": {
|
| 135 |
"content": "<|SYSTEM_TOKEN|>",
|
| 136 |
+
"single_word": false,
|
| 137 |
"lstrip": false,
|
|
|
|
| 138 |
"rstrip": false,
|
| 139 |
+
"normalized": false,
|
| 140 |
"special": false
|
| 141 |
},
|
| 142 |
+
"509": {
|
| 143 |
"content": "<|USER_0_TOKEN|>",
|
| 144 |
+
"single_word": false,
|
| 145 |
"lstrip": false,
|
|
|
|
| 146 |
"rstrip": false,
|
| 147 |
+
"normalized": false,
|
| 148 |
"special": false
|
| 149 |
},
|
| 150 |
+
"510": {
|
| 151 |
"content": "<|USER_1_TOKEN|>",
|
| 152 |
+
"single_word": false,
|
| 153 |
"lstrip": false,
|
|
|
|
| 154 |
"rstrip": false,
|
| 155 |
+
"normalized": false,
|
| 156 |
"special": false
|
| 157 |
},
|
| 158 |
+
"511": {
|
| 159 |
"content": "<|USER_2_TOKEN|>",
|
| 160 |
+
"single_word": false,
|
| 161 |
"lstrip": false,
|
|
|
|
| 162 |
"rstrip": false,
|
| 163 |
+
"normalized": false,
|
| 164 |
"special": false
|
| 165 |
},
|
| 166 |
+
"512": {
|
| 167 |
"content": "<|USER_3_TOKEN|>",
|
| 168 |
+
"single_word": false,
|
| 169 |
"lstrip": false,
|
|
|
|
| 170 |
"rstrip": false,
|
| 171 |
+
"normalized": false,
|
| 172 |
"special": false
|
| 173 |
},
|
| 174 |
+
"513": {
|
| 175 |
"content": "<|USER_4_TOKEN|>",
|
| 176 |
+
"single_word": false,
|
| 177 |
"lstrip": false,
|
|
|
|
| 178 |
"rstrip": false,
|
| 179 |
+
"normalized": false,
|
| 180 |
"special": false
|
| 181 |
},
|
| 182 |
+
"514": {
|
| 183 |
"content": "<|USER_5_TOKEN|>",
|
| 184 |
+
"single_word": false,
|
| 185 |
"lstrip": false,
|
|
|
|
| 186 |
"rstrip": false,
|
| 187 |
+
"normalized": false,
|
| 188 |
"special": false
|
| 189 |
},
|
| 190 |
+
"515": {
|
| 191 |
"content": "<|USER_6_TOKEN|>",
|
| 192 |
+
"single_word": false,
|
| 193 |
"lstrip": false,
|
|
|
|
| 194 |
"rstrip": false,
|
| 195 |
+
"normalized": false,
|
| 196 |
"special": false
|
| 197 |
},
|
| 198 |
+
"516": {
|
| 199 |
"content": "<|USER_7_TOKEN|>",
|
| 200 |
+
"single_word": false,
|
| 201 |
"lstrip": false,
|
|
|
|
| 202 |
"rstrip": false,
|
| 203 |
+
"normalized": false,
|
| 204 |
"special": false
|
| 205 |
},
|
| 206 |
+
"517": {
|
| 207 |
"content": "<|USER_8_TOKEN|>",
|
| 208 |
+
"single_word": false,
|
| 209 |
"lstrip": false,
|
|
|
|
| 210 |
"rstrip": false,
|
| 211 |
+
"normalized": false,
|
| 212 |
"special": false
|
| 213 |
},
|
| 214 |
+
"518": {
|
| 215 |
"content": "<|USER_9_TOKEN|>",
|
| 216 |
+
"single_word": false,
|
| 217 |
"lstrip": false,
|
|
|
|
| 218 |
"rstrip": false,
|
| 219 |
+
"normalized": false,
|
| 220 |
"special": false
|
| 221 |
},
|
| 222 |
+
"519": {
|
| 223 |
"content": "<|START_THINKING|>",
|
| 224 |
+
"single_word": false,
|
| 225 |
"lstrip": false,
|
|
|
|
| 226 |
"rstrip": false,
|
| 227 |
+
"normalized": false,
|
| 228 |
"special": false
|
| 229 |
},
|
| 230 |
+
"520": {
|
| 231 |
"content": "<|END_THINKING|>",
|
| 232 |
+
"single_word": false,
|
| 233 |
"lstrip": false,
|
|
|
|
| 234 |
"rstrip": false,
|
| 235 |
+
"normalized": false,
|
| 236 |
"special": false
|
| 237 |
},
|
| 238 |
+
"521": {
|
| 239 |
"content": "<|START_RESPONSE|>",
|
| 240 |
+
"single_word": false,
|
| 241 |
"lstrip": false,
|
|
|
|
| 242 |
"rstrip": false,
|
| 243 |
+
"normalized": false,
|
| 244 |
"special": true
|
| 245 |
},
|
| 246 |
+
"522": {
|
| 247 |
"content": "<|END_RESPONSE|>",
|
| 248 |
+
"single_word": false,
|
| 249 |
"lstrip": false,
|
|
|
|
| 250 |
"rstrip": false,
|
| 251 |
+
"normalized": false,
|
| 252 |
"special": true
|
| 253 |
},
|
| 254 |
+
"523": {
|
| 255 |
"content": "<|START_ACTION|>",
|
| 256 |
+
"single_word": false,
|
| 257 |
"lstrip": false,
|
|
|
|
| 258 |
"rstrip": false,
|
| 259 |
+
"normalized": false,
|
| 260 |
"special": false
|
| 261 |
},
|
| 262 |
+
"524": {
|
| 263 |
"content": "<|END_ACTION|>",
|
| 264 |
+
"single_word": false,
|
| 265 |
"lstrip": false,
|
|
|
|
| 266 |
"rstrip": false,
|
| 267 |
+
"normalized": false,
|
| 268 |
"special": false
|
| 269 |
},
|
| 270 |
+
"525": {
|
| 271 |
"content": "<|START_TOOL_RESULT|>",
|
| 272 |
+
"single_word": false,
|
| 273 |
"lstrip": false,
|
|
|
|
| 274 |
"rstrip": false,
|
| 275 |
+
"normalized": false,
|
| 276 |
"special": false
|
| 277 |
},
|
| 278 |
+
"526": {
|
| 279 |
"content": "<|END_TOOL_RESULT|>",
|
| 280 |
+
"single_word": false,
|
| 281 |
"lstrip": false,
|
|
|
|
| 282 |
"rstrip": false,
|
| 283 |
+
"normalized": false,
|
| 284 |
"special": false
|
| 285 |
},
|
| 286 |
+
"527": {
|
| 287 |
"content": "<|EXTRA_8_TOKEN|>",
|
| 288 |
+
"single_word": false,
|
| 289 |
"lstrip": false,
|
|
|
|
| 290 |
"rstrip": false,
|
| 291 |
+
"normalized": false,
|
| 292 |
"special": false
|
| 293 |
},
|
| 294 |
+
"528": {
|
| 295 |
"content": "<|NEW_FILE|>",
|
| 296 |
+
"single_word": false,
|
| 297 |
"lstrip": false,
|
|
|
|
| 298 |
"rstrip": false,
|
| 299 |
+
"normalized": false,
|
| 300 |
"special": true
|
| 301 |
},
|
| 302 |
+
"529": {
|
| 303 |
"content": "<|BEGINNING_OF_PREFIX_FIM_TOKEN|>",
|
| 304 |
+
"single_word": false,
|
| 305 |
"lstrip": false,
|
|
|
|
| 306 |
"rstrip": false,
|
| 307 |
+
"normalized": false,
|
| 308 |
"special": false
|
| 309 |
},
|
| 310 |
+
"530": {
|
| 311 |
"content": "<|BEGINNING_OF_MIDDLE_FIM_TOKEN|>",
|
| 312 |
+
"single_word": false,
|
| 313 |
"lstrip": false,
|
|
|
|
| 314 |
"rstrip": false,
|
| 315 |
+
"normalized": false,
|
| 316 |
"special": false
|
| 317 |
},
|
| 318 |
+
"531": {
|
| 319 |
"content": "<|BEGINNING_OF_SUFFIX_FIM_TOKEN|>",
|
| 320 |
+
"single_word": false,
|
| 321 |
"lstrip": false,
|
|
|
|
| 322 |
"rstrip": false,
|
| 323 |
+
"normalized": false,
|
| 324 |
"special": false
|
| 325 |
},
|
| 326 |
+
"532": {
|
| 327 |
"content": "<|END_OF_MIDDLE_FIM_TOKEN|>",
|
| 328 |
+
"single_word": false,
|
| 329 |
"lstrip": false,
|
|
|
|
| 330 |
"rstrip": false,
|
| 331 |
+
"normalized": false,
|
| 332 |
"special": false
|
| 333 |
},
|
| 334 |
+
"533": {
|
| 335 |
"content": "<|START_OF_IMG|>",
|
| 336 |
+
"single_word": false,
|
| 337 |
"lstrip": false,
|
|
|
|
| 338 |
"rstrip": false,
|
| 339 |
+
"normalized": false,
|
| 340 |
"special": false
|
| 341 |
},
|
| 342 |
+
"534": {
|
| 343 |
"content": "<|END_OF_IMG|>",
|
| 344 |
+
"single_word": false,
|
| 345 |
"lstrip": false,
|
|
|
|
| 346 |
"rstrip": false,
|
| 347 |
+
"normalized": false,
|
| 348 |
"special": false
|
| 349 |
},
|
| 350 |
+
"535": {
|
| 351 |
"content": "<|IMG_LINE_BREAK|>",
|
| 352 |
+
"single_word": false,
|
| 353 |
"lstrip": false,
|
|
|
|
| 354 |
"rstrip": false,
|
| 355 |
+
"normalized": false,
|
| 356 |
"special": false
|
| 357 |
},
|
| 358 |
+
"536": {
|
| 359 |
"content": "<|IMG_PATCH|>",
|
| 360 |
+
"single_word": false,
|
| 361 |
"lstrip": false,
|
|
|
|
| 362 |
"rstrip": false,
|
| 363 |
+
"normalized": false,
|
| 364 |
"special": false
|
| 365 |
+
},
|
| 366 |
+
"537": {
|
| 367 |
+
"content": "TILE",
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"lstrip": false,
|
| 370 |
+
"rstrip": false,
|
| 371 |
+
"normalized": false,
|
| 372 |
+
"special": true
|
| 373 |
}
|
| 374 |
},
|
| 375 |
"bos_token": "<BOS_TOKEN>",
|