Upload tokenizer
Browse files- tokenizer.json +133 -132
- tokenizer_config.json +5 -5
tokenizer.json
CHANGED
|
@@ -13,7 +13,7 @@
|
|
| 13 |
"special": true
|
| 14 |
},
|
| 15 |
{
|
| 16 |
-
"id":
|
| 17 |
"content": "[bos]",
|
| 18 |
"single_word": false,
|
| 19 |
"lstrip": false,
|
|
@@ -22,7 +22,7 @@
|
|
| 22 |
"special": true
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"id":
|
| 26 |
"content": "[pad]",
|
| 27 |
"single_word": false,
|
| 28 |
"lstrip": false,
|
|
@@ -31,7 +31,7 @@
|
|
| 31 |
"special": true
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"id":
|
| 35 |
"content": "[eos]",
|
| 36 |
"single_word": false,
|
| 37 |
"lstrip": false,
|
|
@@ -40,7 +40,7 @@
|
|
| 40 |
"special": true
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"id":
|
| 44 |
"content": "[sep]",
|
| 45 |
"single_word": false,
|
| 46 |
"lstrip": false,
|
|
@@ -49,7 +49,7 @@
|
|
| 49 |
"special": true
|
| 50 |
},
|
| 51 |
{
|
| 52 |
-
"id":
|
| 53 |
"content": "[unk]",
|
| 54 |
"single_word": false,
|
| 55 |
"lstrip": false,
|
|
@@ -212,133 +212,134 @@
|
|
| 212 |
"H": 129,
|
| 213 |
"H2": 130,
|
| 214 |
"H3": 131,
|
| 215 |
-
"
|
| 216 |
-
"
|
| 217 |
-
"
|
| 218 |
-
"
|
| 219 |
-
"
|
| 220 |
-
"
|
| 221 |
-
"
|
| 222 |
-
"
|
| 223 |
-
"
|
| 224 |
-
"
|
| 225 |
-
"
|
| 226 |
-
"
|
| 227 |
-
"[
|
| 228 |
-
"[
|
| 229 |
-
"
|
| 230 |
-
"
|
| 231 |
-
"
|
| 232 |
-
"
|
| 233 |
-
"
|
| 234 |
-
"
|
| 235 |
-
"
|
| 236 |
-
"
|
| 237 |
-
"
|
| 238 |
-
"
|
| 239 |
-
"
|
| 240 |
-
"
|
| 241 |
-
"
|
| 242 |
-
"
|
| 243 |
-
"
|
| 244 |
-
"
|
| 245 |
-
"
|
| 246 |
-
"
|
| 247 |
-
"
|
| 248 |
-
"
|
| 249 |
-
"
|
| 250 |
-
"
|
| 251 |
-
"
|
| 252 |
-
"
|
| 253 |
-
"
|
| 254 |
-
"
|
| 255 |
-
"
|
| 256 |
-
"
|
| 257 |
-
"
|
| 258 |
-
"
|
| 259 |
-
"
|
| 260 |
-
"
|
| 261 |
-
"
|
| 262 |
-
"
|
| 263 |
-
"
|
| 264 |
-
"
|
| 265 |
-
"
|
| 266 |
-
"
|
| 267 |
-
"
|
| 268 |
-
"
|
| 269 |
-
"
|
| 270 |
-
"
|
| 271 |
-
"
|
| 272 |
-
"
|
| 273 |
-
"
|
| 274 |
-
"
|
| 275 |
-
"
|
| 276 |
-
"
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
"
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
"
|
| 283 |
-
"
|
| 284 |
-
"
|
| 285 |
-
"
|
| 286 |
-
"
|
| 287 |
-
"
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
"
|
| 292 |
-
"
|
| 293 |
-
"
|
| 294 |
-
"
|
| 295 |
-
"
|
| 296 |
-
"
|
| 297 |
-
"
|
| 298 |
-
"
|
| 299 |
-
"
|
| 300 |
-
"
|
| 301 |
-
"
|
| 302 |
-
"
|
| 303 |
-
"
|
| 304 |
-
"
|
| 305 |
-
"
|
| 306 |
-
"
|
| 307 |
-
"
|
| 308 |
-
"
|
| 309 |
-
"
|
| 310 |
-
"
|
| 311 |
-
"
|
| 312 |
-
"
|
| 313 |
-
"
|
| 314 |
-
"
|
| 315 |
-
"
|
| 316 |
-
"
|
| 317 |
-
"
|
| 318 |
-
"
|
| 319 |
-
"
|
| 320 |
-
"
|
| 321 |
-
"
|
| 322 |
-
"
|
| 323 |
-
"
|
| 324 |
-
"
|
| 325 |
-
"
|
| 326 |
-
"
|
| 327 |
-
"
|
| 328 |
-
"
|
| 329 |
-
"
|
| 330 |
-
"
|
| 331 |
-
"
|
| 332 |
-
"
|
| 333 |
-
"
|
| 334 |
-
"
|
| 335 |
-
"
|
| 336 |
-
"
|
| 337 |
-
"
|
| 338 |
-
"[
|
| 339 |
-
"[
|
| 340 |
-
"[
|
| 341 |
-
"[
|
|
|
|
| 342 |
},
|
| 343 |
"unk_token": "[unk]"
|
| 344 |
}
|
|
|
|
| 13 |
"special": true
|
| 14 |
},
|
| 15 |
{
|
| 16 |
+
"id": 255,
|
| 17 |
"content": "[bos]",
|
| 18 |
"single_word": false,
|
| 19 |
"lstrip": false,
|
|
|
|
| 22 |
"special": true
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"id": 256,
|
| 26 |
"content": "[pad]",
|
| 27 |
"single_word": false,
|
| 28 |
"lstrip": false,
|
|
|
|
| 31 |
"special": true
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"id": 257,
|
| 35 |
"content": "[eos]",
|
| 36 |
"single_word": false,
|
| 37 |
"lstrip": false,
|
|
|
|
| 40 |
"special": true
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"id": 258,
|
| 44 |
"content": "[sep]",
|
| 45 |
"single_word": false,
|
| 46 |
"lstrip": false,
|
|
|
|
| 49 |
"special": true
|
| 50 |
},
|
| 51 |
{
|
| 52 |
+
"id": 259,
|
| 53 |
"content": "[unk]",
|
| 54 |
"single_word": false,
|
| 55 |
"lstrip": false,
|
|
|
|
| 212 |
"H": 129,
|
| 213 |
"H2": 130,
|
| 214 |
"H3": 131,
|
| 215 |
+
"H4": 132,
|
| 216 |
+
"[": 133,
|
| 217 |
+
"]": 134,
|
| 218 |
+
":": 135,
|
| 219 |
+
"=": 136,
|
| 220 |
+
"#": 137,
|
| 221 |
+
"$": 138,
|
| 222 |
+
"\\": 139,
|
| 223 |
+
"/": 140,
|
| 224 |
+
"(": 141,
|
| 225 |
+
")": 142,
|
| 226 |
+
".": 143,
|
| 227 |
+
"[prod]": 144,
|
| 228 |
+
"[reac]": 145,
|
| 229 |
+
"[mech]": 146,
|
| 230 |
+
"+": 147,
|
| 231 |
+
"-": 148,
|
| 232 |
+
"@": 149,
|
| 233 |
+
"*": 150,
|
| 234 |
+
"0": 151,
|
| 235 |
+
"1": 152,
|
| 236 |
+
"2": 153,
|
| 237 |
+
"3": 154,
|
| 238 |
+
"4": 155,
|
| 239 |
+
"5": 156,
|
| 240 |
+
"6": 157,
|
| 241 |
+
"7": 158,
|
| 242 |
+
"8": 159,
|
| 243 |
+
"9": 160,
|
| 244 |
+
"10": 161,
|
| 245 |
+
"11": 162,
|
| 246 |
+
"12": 163,
|
| 247 |
+
"13": 164,
|
| 248 |
+
"14": 165,
|
| 249 |
+
"15": 166,
|
| 250 |
+
"16": 167,
|
| 251 |
+
"17": 168,
|
| 252 |
+
"18": 169,
|
| 253 |
+
"19": 170,
|
| 254 |
+
"20": 171,
|
| 255 |
+
"21": 172,
|
| 256 |
+
"22": 173,
|
| 257 |
+
"23": 174,
|
| 258 |
+
"24": 175,
|
| 259 |
+
"25": 176,
|
| 260 |
+
"26": 177,
|
| 261 |
+
"27": 178,
|
| 262 |
+
"28": 179,
|
| 263 |
+
"29": 180,
|
| 264 |
+
"30": 181,
|
| 265 |
+
"31": 182,
|
| 266 |
+
"32": 183,
|
| 267 |
+
"33": 184,
|
| 268 |
+
"34": 185,
|
| 269 |
+
"35": 186,
|
| 270 |
+
"36": 187,
|
| 271 |
+
"37": 188,
|
| 272 |
+
"38": 189,
|
| 273 |
+
"39": 190,
|
| 274 |
+
"40": 191,
|
| 275 |
+
"41": 192,
|
| 276 |
+
"42": 193,
|
| 277 |
+
"43": 194,
|
| 278 |
+
"44": 195,
|
| 279 |
+
"45": 196,
|
| 280 |
+
"46": 197,
|
| 281 |
+
"47": 198,
|
| 282 |
+
"48": 199,
|
| 283 |
+
"49": 200,
|
| 284 |
+
"50": 201,
|
| 285 |
+
"51": 202,
|
| 286 |
+
"52": 203,
|
| 287 |
+
"53": 204,
|
| 288 |
+
"54": 205,
|
| 289 |
+
"55": 206,
|
| 290 |
+
"56": 207,
|
| 291 |
+
"57": 208,
|
| 292 |
+
"58": 209,
|
| 293 |
+
"59": 210,
|
| 294 |
+
"60": 211,
|
| 295 |
+
"61": 212,
|
| 296 |
+
"62": 213,
|
| 297 |
+
"63": 214,
|
| 298 |
+
"64": 215,
|
| 299 |
+
"65": 216,
|
| 300 |
+
"66": 217,
|
| 301 |
+
"67": 218,
|
| 302 |
+
"68": 219,
|
| 303 |
+
"69": 220,
|
| 304 |
+
"70": 221,
|
| 305 |
+
"71": 222,
|
| 306 |
+
"72": 223,
|
| 307 |
+
"73": 224,
|
| 308 |
+
"74": 225,
|
| 309 |
+
"75": 226,
|
| 310 |
+
"76": 227,
|
| 311 |
+
"77": 228,
|
| 312 |
+
"78": 229,
|
| 313 |
+
"79": 230,
|
| 314 |
+
"80": 231,
|
| 315 |
+
"81": 232,
|
| 316 |
+
"82": 233,
|
| 317 |
+
"83": 234,
|
| 318 |
+
"84": 235,
|
| 319 |
+
"85": 236,
|
| 320 |
+
"86": 237,
|
| 321 |
+
"87": 238,
|
| 322 |
+
"88": 239,
|
| 323 |
+
"89": 240,
|
| 324 |
+
"90": 241,
|
| 325 |
+
"91": 242,
|
| 326 |
+
"92": 243,
|
| 327 |
+
"93": 244,
|
| 328 |
+
"94": 245,
|
| 329 |
+
"95": 246,
|
| 330 |
+
"96": 247,
|
| 331 |
+
"97": 248,
|
| 332 |
+
"98": 249,
|
| 333 |
+
"99": 250,
|
| 334 |
+
"%": 251,
|
| 335 |
+
",": 252,
|
| 336 |
+
";": 253,
|
| 337 |
+
"|": 254,
|
| 338 |
+
"[bos]": 255,
|
| 339 |
+
"[pad]": 256,
|
| 340 |
+
"[eos]": 257,
|
| 341 |
+
"[sep]": 258,
|
| 342 |
+
"[unk]": 259
|
| 343 |
},
|
| 344 |
"unk_token": "[unk]"
|
| 345 |
}
|
tokenizer_config.json
CHANGED
|
@@ -8,7 +8,7 @@
|
|
| 8 |
"single_word": false,
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
-
"
|
| 12 |
"content": "[bos]",
|
| 13 |
"lstrip": false,
|
| 14 |
"normalized": false,
|
|
@@ -16,7 +16,7 @@
|
|
| 16 |
"single_word": false,
|
| 17 |
"special": true
|
| 18 |
},
|
| 19 |
-
"
|
| 20 |
"content": "[pad]",
|
| 21 |
"lstrip": false,
|
| 22 |
"normalized": false,
|
|
@@ -24,7 +24,7 @@
|
|
| 24 |
"single_word": false,
|
| 25 |
"special": true
|
| 26 |
},
|
| 27 |
-
"
|
| 28 |
"content": "[eos]",
|
| 29 |
"lstrip": false,
|
| 30 |
"normalized": false,
|
|
@@ -32,7 +32,7 @@
|
|
| 32 |
"single_word": false,
|
| 33 |
"special": true
|
| 34 |
},
|
| 35 |
-
"
|
| 36 |
"content": "[sep]",
|
| 37 |
"lstrip": false,
|
| 38 |
"normalized": false,
|
|
@@ -40,7 +40,7 @@
|
|
| 40 |
"single_word": false,
|
| 41 |
"special": true
|
| 42 |
},
|
| 43 |
-
"
|
| 44 |
"content": "[unk]",
|
| 45 |
"lstrip": false,
|
| 46 |
"normalized": false,
|
|
|
|
| 8 |
"single_word": false,
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
+
"255": {
|
| 12 |
"content": "[bos]",
|
| 13 |
"lstrip": false,
|
| 14 |
"normalized": false,
|
|
|
|
| 16 |
"single_word": false,
|
| 17 |
"special": true
|
| 18 |
},
|
| 19 |
+
"256": {
|
| 20 |
"content": "[pad]",
|
| 21 |
"lstrip": false,
|
| 22 |
"normalized": false,
|
|
|
|
| 24 |
"single_word": false,
|
| 25 |
"special": true
|
| 26 |
},
|
| 27 |
+
"257": {
|
| 28 |
"content": "[eos]",
|
| 29 |
"lstrip": false,
|
| 30 |
"normalized": false,
|
|
|
|
| 32 |
"single_word": false,
|
| 33 |
"special": true
|
| 34 |
},
|
| 35 |
+
"258": {
|
| 36 |
"content": "[sep]",
|
| 37 |
"lstrip": false,
|
| 38 |
"normalized": false,
|
|
|
|
| 40 |
"single_word": false,
|
| 41 |
"special": true
|
| 42 |
},
|
| 43 |
+
"259": {
|
| 44 |
"content": "[unk]",
|
| 45 |
"lstrip": false,
|
| 46 |
"normalized": false,
|