Upload DebertaV2ForSequenceClassification
Browse files- config.json +339 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,339 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/mdeberta-v3-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifiers_size": [
|
| 8 |
+
3,
|
| 9 |
+
3,
|
| 10 |
+
3,
|
| 11 |
+
3,
|
| 12 |
+
3,
|
| 13 |
+
3,
|
| 14 |
+
3,
|
| 15 |
+
3,
|
| 16 |
+
3,
|
| 17 |
+
3,
|
| 18 |
+
3,
|
| 19 |
+
3,
|
| 20 |
+
3,
|
| 21 |
+
3,
|
| 22 |
+
3,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
1,
|
| 26 |
+
1,
|
| 27 |
+
1,
|
| 28 |
+
1,
|
| 29 |
+
1,
|
| 30 |
+
1,
|
| 31 |
+
1,
|
| 32 |
+
1,
|
| 33 |
+
1,
|
| 34 |
+
1,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
11,
|
| 43 |
+
15,
|
| 44 |
+
31,
|
| 45 |
+
12,
|
| 46 |
+
31,
|
| 47 |
+
2,
|
| 48 |
+
2,
|
| 49 |
+
3,
|
| 50 |
+
5,
|
| 51 |
+
6,
|
| 52 |
+
3,
|
| 53 |
+
3,
|
| 54 |
+
3,
|
| 55 |
+
3,
|
| 56 |
+
3,
|
| 57 |
+
3,
|
| 58 |
+
3,
|
| 59 |
+
3,
|
| 60 |
+
3,
|
| 61 |
+
2,
|
| 62 |
+
2,
|
| 63 |
+
2,
|
| 64 |
+
2,
|
| 65 |
+
6,
|
| 66 |
+
6,
|
| 67 |
+
6,
|
| 68 |
+
69,
|
| 69 |
+
2,
|
| 70 |
+
1,
|
| 71 |
+
8,
|
| 72 |
+
10,
|
| 73 |
+
2,
|
| 74 |
+
2,
|
| 75 |
+
5,
|
| 76 |
+
2,
|
| 77 |
+
2,
|
| 78 |
+
2,
|
| 79 |
+
2,
|
| 80 |
+
1,
|
| 81 |
+
1,
|
| 82 |
+
1,
|
| 83 |
+
20,
|
| 84 |
+
235,
|
| 85 |
+
1,
|
| 86 |
+
1,
|
| 87 |
+
1,
|
| 88 |
+
1,
|
| 89 |
+
1,
|
| 90 |
+
1,
|
| 91 |
+
1,
|
| 92 |
+
1,
|
| 93 |
+
1,
|
| 94 |
+
1,
|
| 95 |
+
1,
|
| 96 |
+
1,
|
| 97 |
+
1,
|
| 98 |
+
1,
|
| 99 |
+
1,
|
| 100 |
+
1,
|
| 101 |
+
1,
|
| 102 |
+
1,
|
| 103 |
+
1,
|
| 104 |
+
1,
|
| 105 |
+
1,
|
| 106 |
+
1,
|
| 107 |
+
1,
|
| 108 |
+
1,
|
| 109 |
+
1,
|
| 110 |
+
1,
|
| 111 |
+
1,
|
| 112 |
+
1,
|
| 113 |
+
1,
|
| 114 |
+
1,
|
| 115 |
+
1,
|
| 116 |
+
1,
|
| 117 |
+
1,
|
| 118 |
+
1,
|
| 119 |
+
1,
|
| 120 |
+
1,
|
| 121 |
+
1,
|
| 122 |
+
1,
|
| 123 |
+
1,
|
| 124 |
+
1,
|
| 125 |
+
1,
|
| 126 |
+
1,
|
| 127 |
+
1,
|
| 128 |
+
1,
|
| 129 |
+
1,
|
| 130 |
+
1,
|
| 131 |
+
1,
|
| 132 |
+
1,
|
| 133 |
+
1,
|
| 134 |
+
1,
|
| 135 |
+
1,
|
| 136 |
+
1,
|
| 137 |
+
1,
|
| 138 |
+
1,
|
| 139 |
+
1,
|
| 140 |
+
1,
|
| 141 |
+
1,
|
| 142 |
+
1,
|
| 143 |
+
1,
|
| 144 |
+
1,
|
| 145 |
+
1,
|
| 146 |
+
1,
|
| 147 |
+
1,
|
| 148 |
+
1,
|
| 149 |
+
1,
|
| 150 |
+
1,
|
| 151 |
+
9,
|
| 152 |
+
17
|
| 153 |
+
],
|
| 154 |
+
"hidden_act": "gelu",
|
| 155 |
+
"hidden_dropout_prob": 0.1,
|
| 156 |
+
"hidden_size": 768,
|
| 157 |
+
"id2label": {
|
| 158 |
+
"0": "entailment",
|
| 159 |
+
"1": "neutral",
|
| 160 |
+
"2": "contradiction"
|
| 161 |
+
},
|
| 162 |
+
"initializer_range": 0.02,
|
| 163 |
+
"intermediate_size": 3072,
|
| 164 |
+
"label2id": {
|
| 165 |
+
"contradiction": 2,
|
| 166 |
+
"entailment": 0,
|
| 167 |
+
"neutral": 1
|
| 168 |
+
},
|
| 169 |
+
"layer_norm_eps": 1e-07,
|
| 170 |
+
"max_position_embeddings": 512,
|
| 171 |
+
"max_relative_positions": -1,
|
| 172 |
+
"model_type": "deberta-v2",
|
| 173 |
+
"norm_rel_ebd": "layer_norm",
|
| 174 |
+
"num_attention_heads": 12,
|
| 175 |
+
"num_hidden_layers": 12,
|
| 176 |
+
"pad_token_id": 0,
|
| 177 |
+
"pooler_dropout": 0,
|
| 178 |
+
"pooler_hidden_act": "gelu",
|
| 179 |
+
"pooler_hidden_size": 768,
|
| 180 |
+
"pos_att_type": [
|
| 181 |
+
"p2c",
|
| 182 |
+
"c2p"
|
| 183 |
+
],
|
| 184 |
+
"position_biased_input": false,
|
| 185 |
+
"position_buckets": 256,
|
| 186 |
+
"relative_attention": true,
|
| 187 |
+
"share_att_key": true,
|
| 188 |
+
"tasks": [
|
| 189 |
+
"xnli/ur",
|
| 190 |
+
"xnli/ar",
|
| 191 |
+
"xnli/bg",
|
| 192 |
+
"xnli/de",
|
| 193 |
+
"xnli/el",
|
| 194 |
+
"xnli/zh",
|
| 195 |
+
"xnli/vi",
|
| 196 |
+
"xnli/tr",
|
| 197 |
+
"xnli/th",
|
| 198 |
+
"xnli/sw",
|
| 199 |
+
"xnli/en",
|
| 200 |
+
"xnli/ru",
|
| 201 |
+
"xnli/es",
|
| 202 |
+
"xnli/fr",
|
| 203 |
+
"xnli/hi",
|
| 204 |
+
"americas_nli/all_languages",
|
| 205 |
+
"multilingual-NLI-26lang-2mil7",
|
| 206 |
+
"stsb_multi_mt/pl",
|
| 207 |
+
"stsb_multi_mt/it",
|
| 208 |
+
"stsb_multi_mt/fr",
|
| 209 |
+
"stsb_multi_mt/nl",
|
| 210 |
+
"stsb_multi_mt/es",
|
| 211 |
+
"stsb_multi_mt/de",
|
| 212 |
+
"stsb_multi_mt/en",
|
| 213 |
+
"stsb_multi_mt/pt",
|
| 214 |
+
"stsb_multi_mt/ru",
|
| 215 |
+
"stsb_multi_mt/zh",
|
| 216 |
+
"paws-x/zh",
|
| 217 |
+
"paws-x/ko",
|
| 218 |
+
"paws-x/fr",
|
| 219 |
+
"paws-x/es",
|
| 220 |
+
"paws-x/de",
|
| 221 |
+
"paws-x/en",
|
| 222 |
+
"paws-x/ja",
|
| 223 |
+
"miam/dihana",
|
| 224 |
+
"miam/ilisten",
|
| 225 |
+
"miam/loria",
|
| 226 |
+
"miam/maptask",
|
| 227 |
+
"miam/vm2",
|
| 228 |
+
"x-stance/de",
|
| 229 |
+
"x-stance/fr",
|
| 230 |
+
"multilingual-sentiments/all",
|
| 231 |
+
"universal-joy",
|
| 232 |
+
"amazon_reviews_multi/all_languages",
|
| 233 |
+
"tweet_sentiment_multilingual/all",
|
| 234 |
+
"tweet_sentiment_multilingual/portuguese",
|
| 235 |
+
"tweet_sentiment_multilingual/italian",
|
| 236 |
+
"tweet_sentiment_multilingual/spanish",
|
| 237 |
+
"tweet_sentiment_multilingual/german",
|
| 238 |
+
"tweet_sentiment_multilingual/french",
|
| 239 |
+
"tweet_sentiment_multilingual/english",
|
| 240 |
+
"tweet_sentiment_multilingual/arabic",
|
| 241 |
+
"tweet_sentiment_multilingual/hindi",
|
| 242 |
+
"offenseval_2020/tr",
|
| 243 |
+
"offenseval_2020/ar",
|
| 244 |
+
"offenseval_2020/da",
|
| 245 |
+
"offenseval_2020/gr",
|
| 246 |
+
"offenseval_dravidian/tamil",
|
| 247 |
+
"offenseval_dravidian/malayalam",
|
| 248 |
+
"offenseval_dravidian/kannada",
|
| 249 |
+
"MLMA_hate_speech",
|
| 250 |
+
"xglue/qam",
|
| 251 |
+
"xsum_factuality",
|
| 252 |
+
"x-fact",
|
| 253 |
+
"xglue/nc",
|
| 254 |
+
"xglue/qadsm",
|
| 255 |
+
"xglue/qam",
|
| 256 |
+
"xglue/wpr",
|
| 257 |
+
"xlwic/xlwic_en_ko",
|
| 258 |
+
"xlwic/xlwic_fr_fr",
|
| 259 |
+
"xlwic/xlwic_it_it",
|
| 260 |
+
"xlwic/xlwic_de_de",
|
| 261 |
+
"oasst1_dense_flat/quality",
|
| 262 |
+
"oasst1_dense_flat/toxicity",
|
| 263 |
+
"oasst1_dense_flat/helpfulness",
|
| 264 |
+
"language-identification",
|
| 265 |
+
"wili_2018",
|
| 266 |
+
"exams/multilingual",
|
| 267 |
+
"xcsr/X-CSQA-es",
|
| 268 |
+
"xcsr/X-CSQA-en",
|
| 269 |
+
"xcsr/X-CODAH-ur",
|
| 270 |
+
"xcsr/X-CODAH-sw",
|
| 271 |
+
"xcsr/X-CODAH-hi",
|
| 272 |
+
"xcsr/X-CODAH-vi",
|
| 273 |
+
"xcsr/X-CODAH-ar",
|
| 274 |
+
"xcsr/X-CODAH-ru",
|
| 275 |
+
"xcsr/X-CODAH-pt",
|
| 276 |
+
"xcsr/X-CODAH-pl",
|
| 277 |
+
"xcsr/X-CODAH-nl",
|
| 278 |
+
"xcsr/X-CODAH-jap",
|
| 279 |
+
"xcsr/X-CODAH-it",
|
| 280 |
+
"xcsr/X-CODAH-fr",
|
| 281 |
+
"xcsr/X-CODAH-es",
|
| 282 |
+
"xcsr/X-CODAH-de",
|
| 283 |
+
"xcsr/X-CODAH-zh",
|
| 284 |
+
"xcsr/X-CODAH-en",
|
| 285 |
+
"xcsr/X-CSQA-ur",
|
| 286 |
+
"xcsr/X-CSQA-sw",
|
| 287 |
+
"xcsr/X-CSQA-zh",
|
| 288 |
+
"xcsr/X-CSQA-vi",
|
| 289 |
+
"xcsr/X-CSQA-ar",
|
| 290 |
+
"xcsr/X-CSQA-ru",
|
| 291 |
+
"xcsr/X-CSQA-pt",
|
| 292 |
+
"xcsr/X-CSQA-pl",
|
| 293 |
+
"xcsr/X-CSQA-nl",
|
| 294 |
+
"xcsr/X-CSQA-jap",
|
| 295 |
+
"xcsr/X-CSQA-it",
|
| 296 |
+
"xcsr/X-CSQA-fr",
|
| 297 |
+
"xcsr/X-CSQA-de",
|
| 298 |
+
"xcsr/X-CSQA-hi",
|
| 299 |
+
"xcopa/translation-th",
|
| 300 |
+
"xcopa/sw",
|
| 301 |
+
"xcopa/et",
|
| 302 |
+
"xcopa/ht",
|
| 303 |
+
"xcopa/it",
|
| 304 |
+
"xcopa/id",
|
| 305 |
+
"xcopa/qu",
|
| 306 |
+
"xcopa/translation-vi",
|
| 307 |
+
"xcopa/zh",
|
| 308 |
+
"xcopa/ta",
|
| 309 |
+
"xcopa/th",
|
| 310 |
+
"xcopa/vi",
|
| 311 |
+
"xcopa/tr",
|
| 312 |
+
"xcopa/translation-ht",
|
| 313 |
+
"xcopa/translation-it",
|
| 314 |
+
"xcopa/translation-id",
|
| 315 |
+
"xcopa/translation-sw",
|
| 316 |
+
"xcopa/translation-zh",
|
| 317 |
+
"xcopa/translation-ta",
|
| 318 |
+
"xcopa/translation-tr",
|
| 319 |
+
"xcopa/translation-et",
|
| 320 |
+
"xstory_cloze/ar",
|
| 321 |
+
"xstory_cloze/te",
|
| 322 |
+
"xstory_cloze/sw",
|
| 323 |
+
"xstory_cloze/id",
|
| 324 |
+
"xstory_cloze/hi",
|
| 325 |
+
"xstory_cloze/es",
|
| 326 |
+
"xstory_cloze/my",
|
| 327 |
+
"xstory_cloze/ru",
|
| 328 |
+
"xstory_cloze/en",
|
| 329 |
+
"xstory_cloze/eu",
|
| 330 |
+
"xstory_cloze/zh",
|
| 331 |
+
"hh-rlhf",
|
| 332 |
+
"xglue/ner",
|
| 333 |
+
"xglue/pos"
|
| 334 |
+
],
|
| 335 |
+
"torch_dtype": "float32",
|
| 336 |
+
"transformers_version": "4.26.1",
|
| 337 |
+
"type_vocab_size": 0,
|
| 338 |
+
"vocab_size": 251000
|
| 339 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83bbe159cce4136ff46b1815812d57de7cb8cf6bff95445d04eadc0bedd70c62
|
| 3 |
+
size 1115320633
|