Upload folder using huggingface_hub
Browse files- config.json +353 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +51 -0
- tokenizer_config.json +57 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,353 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "CWE111",
|
| 14 |
+
"1": "CWE113",
|
| 15 |
+
"2": "CWE114",
|
| 16 |
+
"3": "CWE123",
|
| 17 |
+
"4": "CWE129",
|
| 18 |
+
"5": "CWE134",
|
| 19 |
+
"6": "CWE15",
|
| 20 |
+
"7": "CWE176",
|
| 21 |
+
"8": "CWE188",
|
| 22 |
+
"9": "CWE190",
|
| 23 |
+
"10": "CWE191",
|
| 24 |
+
"11": "CWE193",
|
| 25 |
+
"12": "CWE196",
|
| 26 |
+
"13": "CWE197",
|
| 27 |
+
"14": "CWE209",
|
| 28 |
+
"15": "CWE222",
|
| 29 |
+
"16": "CWE223",
|
| 30 |
+
"17": "CWE226",
|
| 31 |
+
"18": "CWE23",
|
| 32 |
+
"19": "CWE242",
|
| 33 |
+
"20": "CWE244",
|
| 34 |
+
"21": "CWE247",
|
| 35 |
+
"22": "CWE248",
|
| 36 |
+
"23": "CWE252",
|
| 37 |
+
"24": "CWE253",
|
| 38 |
+
"25": "CWE256",
|
| 39 |
+
"26": "CWE259",
|
| 40 |
+
"27": "CWE272",
|
| 41 |
+
"28": "CWE273",
|
| 42 |
+
"29": "CWE284",
|
| 43 |
+
"30": "CWE315",
|
| 44 |
+
"31": "CWE319",
|
| 45 |
+
"32": "CWE321",
|
| 46 |
+
"33": "CWE325",
|
| 47 |
+
"34": "CWE327",
|
| 48 |
+
"35": "CWE328",
|
| 49 |
+
"36": "CWE329",
|
| 50 |
+
"37": "CWE336",
|
| 51 |
+
"38": "CWE338",
|
| 52 |
+
"39": "CWE36",
|
| 53 |
+
"40": "CWE364",
|
| 54 |
+
"41": "CWE366",
|
| 55 |
+
"42": "CWE367",
|
| 56 |
+
"43": "CWE369",
|
| 57 |
+
"44": "CWE377",
|
| 58 |
+
"45": "CWE378",
|
| 59 |
+
"46": "CWE379",
|
| 60 |
+
"47": "CWE382",
|
| 61 |
+
"48": "CWE383",
|
| 62 |
+
"49": "CWE390",
|
| 63 |
+
"50": "CWE391",
|
| 64 |
+
"51": "CWE395",
|
| 65 |
+
"52": "CWE396",
|
| 66 |
+
"53": "CWE397",
|
| 67 |
+
"54": "CWE398",
|
| 68 |
+
"55": "CWE400",
|
| 69 |
+
"56": "CWE404",
|
| 70 |
+
"57": "CWE416",
|
| 71 |
+
"58": "CWE426",
|
| 72 |
+
"59": "CWE427",
|
| 73 |
+
"60": "CWE440",
|
| 74 |
+
"61": "CWE459",
|
| 75 |
+
"62": "CWE464",
|
| 76 |
+
"63": "CWE467",
|
| 77 |
+
"64": "CWE468",
|
| 78 |
+
"65": "CWE469",
|
| 79 |
+
"66": "CWE470",
|
| 80 |
+
"67": "CWE475",
|
| 81 |
+
"68": "CWE476",
|
| 82 |
+
"69": "CWE477",
|
| 83 |
+
"70": "CWE478",
|
| 84 |
+
"71": "CWE479",
|
| 85 |
+
"72": "CWE480",
|
| 86 |
+
"73": "CWE481",
|
| 87 |
+
"74": "CWE482",
|
| 88 |
+
"75": "CWE483",
|
| 89 |
+
"76": "CWE484",
|
| 90 |
+
"77": "CWE486",
|
| 91 |
+
"78": "CWE491",
|
| 92 |
+
"79": "CWE499",
|
| 93 |
+
"80": "CWE500",
|
| 94 |
+
"81": "CWE506",
|
| 95 |
+
"82": "CWE510",
|
| 96 |
+
"83": "CWE511",
|
| 97 |
+
"84": "CWE523",
|
| 98 |
+
"85": "CWE526",
|
| 99 |
+
"86": "CWE533",
|
| 100 |
+
"87": "CWE534",
|
| 101 |
+
"88": "CWE535",
|
| 102 |
+
"89": "CWE539",
|
| 103 |
+
"90": "CWE546",
|
| 104 |
+
"91": "CWE549",
|
| 105 |
+
"92": "CWE561",
|
| 106 |
+
"93": "CWE562",
|
| 107 |
+
"94": "CWE563",
|
| 108 |
+
"95": "CWE566",
|
| 109 |
+
"96": "CWE568",
|
| 110 |
+
"97": "CWE570",
|
| 111 |
+
"98": "CWE571",
|
| 112 |
+
"99": "CWE572",
|
| 113 |
+
"100": "CWE579",
|
| 114 |
+
"101": "CWE580",
|
| 115 |
+
"102": "CWE581",
|
| 116 |
+
"103": "CWE582",
|
| 117 |
+
"104": "CWE584",
|
| 118 |
+
"105": "CWE585",
|
| 119 |
+
"106": "CWE586",
|
| 120 |
+
"107": "CWE587",
|
| 121 |
+
"108": "CWE588",
|
| 122 |
+
"109": "CWE591",
|
| 123 |
+
"110": "CWE597",
|
| 124 |
+
"111": "CWE598",
|
| 125 |
+
"112": "CWE600",
|
| 126 |
+
"113": "CWE601",
|
| 127 |
+
"114": "CWE605",
|
| 128 |
+
"115": "CWE606",
|
| 129 |
+
"116": "CWE607",
|
| 130 |
+
"117": "CWE609",
|
| 131 |
+
"118": "CWE613",
|
| 132 |
+
"119": "CWE614",
|
| 133 |
+
"120": "CWE615",
|
| 134 |
+
"121": "CWE617",
|
| 135 |
+
"122": "CWE620",
|
| 136 |
+
"123": "CWE643",
|
| 137 |
+
"124": "CWE665",
|
| 138 |
+
"125": "CWE666",
|
| 139 |
+
"126": "CWE667",
|
| 140 |
+
"127": "CWE672",
|
| 141 |
+
"128": "CWE674",
|
| 142 |
+
"129": "CWE675",
|
| 143 |
+
"130": "CWE676",
|
| 144 |
+
"131": "CWE680",
|
| 145 |
+
"132": "CWE681",
|
| 146 |
+
"133": "CWE685",
|
| 147 |
+
"134": "CWE688",
|
| 148 |
+
"135": "CWE690",
|
| 149 |
+
"136": "CWE698",
|
| 150 |
+
"137": "CWE758",
|
| 151 |
+
"138": "CWE759",
|
| 152 |
+
"139": "CWE760",
|
| 153 |
+
"140": "CWE761",
|
| 154 |
+
"141": "CWE764",
|
| 155 |
+
"142": "CWE765",
|
| 156 |
+
"143": "CWE772",
|
| 157 |
+
"144": "CWE773",
|
| 158 |
+
"145": "CWE775",
|
| 159 |
+
"146": "CWE78",
|
| 160 |
+
"147": "CWE780",
|
| 161 |
+
"148": "CWE785",
|
| 162 |
+
"149": "CWE789",
|
| 163 |
+
"150": "CWE80",
|
| 164 |
+
"151": "CWE81",
|
| 165 |
+
"152": "CWE83",
|
| 166 |
+
"153": "CWE832",
|
| 167 |
+
"154": "CWE833",
|
| 168 |
+
"155": "CWE835",
|
| 169 |
+
"156": "CWE843",
|
| 170 |
+
"157": "CWE89",
|
| 171 |
+
"158": "CWE90",
|
| 172 |
+
"159": "safe",
|
| 173 |
+
"160": "vulnerable"
|
| 174 |
+
},
|
| 175 |
+
"initializer_range": 0.02,
|
| 176 |
+
"intermediate_size": 3072,
|
| 177 |
+
"label2id": {
|
| 178 |
+
"CWE111": 0,
|
| 179 |
+
"CWE113": 1,
|
| 180 |
+
"CWE114": 2,
|
| 181 |
+
"CWE123": 3,
|
| 182 |
+
"CWE129": 4,
|
| 183 |
+
"CWE134": 5,
|
| 184 |
+
"CWE15": 6,
|
| 185 |
+
"CWE176": 7,
|
| 186 |
+
"CWE188": 8,
|
| 187 |
+
"CWE190": 9,
|
| 188 |
+
"CWE191": 10,
|
| 189 |
+
"CWE193": 11,
|
| 190 |
+
"CWE196": 12,
|
| 191 |
+
"CWE197": 13,
|
| 192 |
+
"CWE209": 14,
|
| 193 |
+
"CWE222": 15,
|
| 194 |
+
"CWE223": 16,
|
| 195 |
+
"CWE226": 17,
|
| 196 |
+
"CWE23": 18,
|
| 197 |
+
"CWE242": 19,
|
| 198 |
+
"CWE244": 20,
|
| 199 |
+
"CWE247": 21,
|
| 200 |
+
"CWE248": 22,
|
| 201 |
+
"CWE252": 23,
|
| 202 |
+
"CWE253": 24,
|
| 203 |
+
"CWE256": 25,
|
| 204 |
+
"CWE259": 26,
|
| 205 |
+
"CWE272": 27,
|
| 206 |
+
"CWE273": 28,
|
| 207 |
+
"CWE284": 29,
|
| 208 |
+
"CWE315": 30,
|
| 209 |
+
"CWE319": 31,
|
| 210 |
+
"CWE321": 32,
|
| 211 |
+
"CWE325": 33,
|
| 212 |
+
"CWE327": 34,
|
| 213 |
+
"CWE328": 35,
|
| 214 |
+
"CWE329": 36,
|
| 215 |
+
"CWE336": 37,
|
| 216 |
+
"CWE338": 38,
|
| 217 |
+
"CWE36": 39,
|
| 218 |
+
"CWE364": 40,
|
| 219 |
+
"CWE366": 41,
|
| 220 |
+
"CWE367": 42,
|
| 221 |
+
"CWE369": 43,
|
| 222 |
+
"CWE377": 44,
|
| 223 |
+
"CWE378": 45,
|
| 224 |
+
"CWE379": 46,
|
| 225 |
+
"CWE382": 47,
|
| 226 |
+
"CWE383": 48,
|
| 227 |
+
"CWE390": 49,
|
| 228 |
+
"CWE391": 50,
|
| 229 |
+
"CWE395": 51,
|
| 230 |
+
"CWE396": 52,
|
| 231 |
+
"CWE397": 53,
|
| 232 |
+
"CWE398": 54,
|
| 233 |
+
"CWE400": 55,
|
| 234 |
+
"CWE404": 56,
|
| 235 |
+
"CWE416": 57,
|
| 236 |
+
"CWE426": 58,
|
| 237 |
+
"CWE427": 59,
|
| 238 |
+
"CWE440": 60,
|
| 239 |
+
"CWE459": 61,
|
| 240 |
+
"CWE464": 62,
|
| 241 |
+
"CWE467": 63,
|
| 242 |
+
"CWE468": 64,
|
| 243 |
+
"CWE469": 65,
|
| 244 |
+
"CWE470": 66,
|
| 245 |
+
"CWE475": 67,
|
| 246 |
+
"CWE476": 68,
|
| 247 |
+
"CWE477": 69,
|
| 248 |
+
"CWE478": 70,
|
| 249 |
+
"CWE479": 71,
|
| 250 |
+
"CWE480": 72,
|
| 251 |
+
"CWE481": 73,
|
| 252 |
+
"CWE482": 74,
|
| 253 |
+
"CWE483": 75,
|
| 254 |
+
"CWE484": 76,
|
| 255 |
+
"CWE486": 77,
|
| 256 |
+
"CWE491": 78,
|
| 257 |
+
"CWE499": 79,
|
| 258 |
+
"CWE500": 80,
|
| 259 |
+
"CWE506": 81,
|
| 260 |
+
"CWE510": 82,
|
| 261 |
+
"CWE511": 83,
|
| 262 |
+
"CWE523": 84,
|
| 263 |
+
"CWE526": 85,
|
| 264 |
+
"CWE533": 86,
|
| 265 |
+
"CWE534": 87,
|
| 266 |
+
"CWE535": 88,
|
| 267 |
+
"CWE539": 89,
|
| 268 |
+
"CWE546": 90,
|
| 269 |
+
"CWE549": 91,
|
| 270 |
+
"CWE561": 92,
|
| 271 |
+
"CWE562": 93,
|
| 272 |
+
"CWE563": 94,
|
| 273 |
+
"CWE566": 95,
|
| 274 |
+
"CWE568": 96,
|
| 275 |
+
"CWE570": 97,
|
| 276 |
+
"CWE571": 98,
|
| 277 |
+
"CWE572": 99,
|
| 278 |
+
"CWE579": 100,
|
| 279 |
+
"CWE580": 101,
|
| 280 |
+
"CWE581": 102,
|
| 281 |
+
"CWE582": 103,
|
| 282 |
+
"CWE584": 104,
|
| 283 |
+
"CWE585": 105,
|
| 284 |
+
"CWE586": 106,
|
| 285 |
+
"CWE587": 107,
|
| 286 |
+
"CWE588": 108,
|
| 287 |
+
"CWE591": 109,
|
| 288 |
+
"CWE597": 110,
|
| 289 |
+
"CWE598": 111,
|
| 290 |
+
"CWE600": 112,
|
| 291 |
+
"CWE601": 113,
|
| 292 |
+
"CWE605": 114,
|
| 293 |
+
"CWE606": 115,
|
| 294 |
+
"CWE607": 116,
|
| 295 |
+
"CWE609": 117,
|
| 296 |
+
"CWE613": 118,
|
| 297 |
+
"CWE614": 119,
|
| 298 |
+
"CWE615": 120,
|
| 299 |
+
"CWE617": 121,
|
| 300 |
+
"CWE620": 122,
|
| 301 |
+
"CWE643": 123,
|
| 302 |
+
"CWE665": 124,
|
| 303 |
+
"CWE666": 125,
|
| 304 |
+
"CWE667": 126,
|
| 305 |
+
"CWE672": 127,
|
| 306 |
+
"CWE674": 128,
|
| 307 |
+
"CWE675": 129,
|
| 308 |
+
"CWE676": 130,
|
| 309 |
+
"CWE680": 131,
|
| 310 |
+
"CWE681": 132,
|
| 311 |
+
"CWE685": 133,
|
| 312 |
+
"CWE688": 134,
|
| 313 |
+
"CWE690": 135,
|
| 314 |
+
"CWE698": 136,
|
| 315 |
+
"CWE758": 137,
|
| 316 |
+
"CWE759": 138,
|
| 317 |
+
"CWE760": 139,
|
| 318 |
+
"CWE761": 140,
|
| 319 |
+
"CWE764": 141,
|
| 320 |
+
"CWE765": 142,
|
| 321 |
+
"CWE772": 143,
|
| 322 |
+
"CWE773": 144,
|
| 323 |
+
"CWE775": 145,
|
| 324 |
+
"CWE78": 146,
|
| 325 |
+
"CWE780": 147,
|
| 326 |
+
"CWE785": 148,
|
| 327 |
+
"CWE789": 149,
|
| 328 |
+
"CWE80": 150,
|
| 329 |
+
"CWE81": 151,
|
| 330 |
+
"CWE83": 152,
|
| 331 |
+
"CWE832": 153,
|
| 332 |
+
"CWE833": 154,
|
| 333 |
+
"CWE835": 155,
|
| 334 |
+
"CWE843": 156,
|
| 335 |
+
"CWE89": 157,
|
| 336 |
+
"CWE90": 158,
|
| 337 |
+
"safe": 159,
|
| 338 |
+
"vulnerable": 160
|
| 339 |
+
},
|
| 340 |
+
"layer_norm_eps": 1e-05,
|
| 341 |
+
"max_position_embeddings": 514,
|
| 342 |
+
"model_type": "roberta",
|
| 343 |
+
"num_attention_heads": 12,
|
| 344 |
+
"num_hidden_layers": 12,
|
| 345 |
+
"output_past": true,
|
| 346 |
+
"pad_token_id": 1,
|
| 347 |
+
"position_embedding_type": "absolute",
|
| 348 |
+
"torch_dtype": "float32",
|
| 349 |
+
"transformers_version": "4.55.0",
|
| 350 |
+
"type_vocab_size": 1,
|
| 351 |
+
"use_cache": true,
|
| 352 |
+
"vocab_size": 50265
|
| 353 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09ef9d417599e92481ac8724d6662760360060795114b908757979782abfbea3
|
| 3 |
+
size 499101908
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|