updating the configs of gpt2-xl
Browse files- config.json +4 -3
- configs/BASIC.yaml +193 -193
config.json
CHANGED
|
@@ -15,10 +15,11 @@
|
|
| 15 |
"layer_norm_epsilon": 1e-05,
|
| 16 |
"model_type": "gpt2",
|
| 17 |
"n_ctx": 1024,
|
| 18 |
-
"n_embd":
|
| 19 |
-
"n_head":
|
| 20 |
-
"n_layer":
|
| 21 |
"n_positions": 1024,
|
|
|
|
| 22 |
"resid_pdrop": 0.1,
|
| 23 |
"summary_activation": null,
|
| 24 |
"summary_first_dropout": 0.1,
|
|
|
|
| 15 |
"layer_norm_epsilon": 1e-05,
|
| 16 |
"model_type": "gpt2",
|
| 17 |
"n_ctx": 1024,
|
| 18 |
+
"n_embd": 1600,
|
| 19 |
+
"n_head": 25,
|
| 20 |
+
"n_layer": 48,
|
| 21 |
"n_positions": 1024,
|
| 22 |
+
"output_past": true,
|
| 23 |
"resid_pdrop": 0.1,
|
| 24 |
"summary_activation": null,
|
| 25 |
"summary_first_dropout": 0.1,
|
configs/BASIC.yaml
CHANGED
|
@@ -39,26 +39,26 @@ model:
|
|
| 39 |
instance: Dropout
|
| 40 |
output_format: SAME
|
| 41 |
transformer.h.0.attn.softmax:
|
| 42 |
-
approximation_function:
|
| 43 |
input_format: SAME
|
| 44 |
instance: Softmax
|
| 45 |
output_format: SAME
|
| 46 |
transformer.h.0.ln_1:
|
| 47 |
-
approximation_function:
|
| 48 |
bias_format: SAME
|
| 49 |
input_format: SAME
|
| 50 |
instance: LayerNorm
|
| 51 |
output_format: SAME
|
| 52 |
weight_format: SAME
|
| 53 |
transformer.h.0.ln_2:
|
| 54 |
-
approximation_function:
|
| 55 |
bias_format: SAME
|
| 56 |
input_format: SAME
|
| 57 |
instance: LayerNorm
|
| 58 |
output_format: SAME
|
| 59 |
weight_format: SAME
|
| 60 |
transformer.h.0.mlp.act:
|
| 61 |
-
approximation_function:
|
| 62 |
input_format: SAME
|
| 63 |
instance: GELU
|
| 64 |
output_format: SAME
|
|
@@ -110,26 +110,26 @@ model:
|
|
| 110 |
instance: Dropout
|
| 111 |
output_format: SAME
|
| 112 |
transformer.h.1.attn.softmax:
|
| 113 |
-
approximation_function:
|
| 114 |
input_format: SAME
|
| 115 |
instance: Softmax
|
| 116 |
output_format: SAME
|
| 117 |
transformer.h.1.ln_1:
|
| 118 |
-
approximation_function:
|
| 119 |
bias_format: SAME
|
| 120 |
input_format: SAME
|
| 121 |
instance: LayerNorm
|
| 122 |
output_format: SAME
|
| 123 |
weight_format: SAME
|
| 124 |
transformer.h.1.ln_2:
|
| 125 |
-
approximation_function:
|
| 126 |
bias_format: SAME
|
| 127 |
input_format: SAME
|
| 128 |
instance: LayerNorm
|
| 129 |
output_format: SAME
|
| 130 |
weight_format: SAME
|
| 131 |
transformer.h.1.mlp.act:
|
| 132 |
-
approximation_function:
|
| 133 |
input_format: SAME
|
| 134 |
instance: GELU
|
| 135 |
output_format: SAME
|
|
@@ -181,26 +181,26 @@ model:
|
|
| 181 |
instance: Dropout
|
| 182 |
output_format: SAME
|
| 183 |
transformer.h.10.attn.softmax:
|
| 184 |
-
approximation_function:
|
| 185 |
input_format: SAME
|
| 186 |
instance: Softmax
|
| 187 |
output_format: SAME
|
| 188 |
transformer.h.10.ln_1:
|
| 189 |
-
approximation_function:
|
| 190 |
bias_format: SAME
|
| 191 |
input_format: SAME
|
| 192 |
instance: LayerNorm
|
| 193 |
output_format: SAME
|
| 194 |
weight_format: SAME
|
| 195 |
transformer.h.10.ln_2:
|
| 196 |
-
approximation_function:
|
| 197 |
bias_format: SAME
|
| 198 |
input_format: SAME
|
| 199 |
instance: LayerNorm
|
| 200 |
output_format: SAME
|
| 201 |
weight_format: SAME
|
| 202 |
transformer.h.10.mlp.act:
|
| 203 |
-
approximation_function:
|
| 204 |
input_format: SAME
|
| 205 |
instance: GELU
|
| 206 |
output_format: SAME
|
|
@@ -252,26 +252,26 @@ model:
|
|
| 252 |
instance: Dropout
|
| 253 |
output_format: SAME
|
| 254 |
transformer.h.11.attn.softmax:
|
| 255 |
-
approximation_function:
|
| 256 |
input_format: SAME
|
| 257 |
instance: Softmax
|
| 258 |
output_format: SAME
|
| 259 |
transformer.h.11.ln_1:
|
| 260 |
-
approximation_function:
|
| 261 |
bias_format: SAME
|
| 262 |
input_format: SAME
|
| 263 |
instance: LayerNorm
|
| 264 |
output_format: SAME
|
| 265 |
weight_format: SAME
|
| 266 |
transformer.h.11.ln_2:
|
| 267 |
-
approximation_function:
|
| 268 |
bias_format: SAME
|
| 269 |
input_format: SAME
|
| 270 |
instance: LayerNorm
|
| 271 |
output_format: SAME
|
| 272 |
weight_format: SAME
|
| 273 |
transformer.h.11.mlp.act:
|
| 274 |
-
approximation_function:
|
| 275 |
input_format: SAME
|
| 276 |
instance: GELU
|
| 277 |
output_format: SAME
|
|
@@ -323,26 +323,26 @@ model:
|
|
| 323 |
instance: Dropout
|
| 324 |
output_format: SAME
|
| 325 |
transformer.h.12.attn.softmax:
|
| 326 |
-
approximation_function:
|
| 327 |
input_format: SAME
|
| 328 |
instance: Softmax
|
| 329 |
output_format: SAME
|
| 330 |
transformer.h.12.ln_1:
|
| 331 |
-
approximation_function:
|
| 332 |
bias_format: SAME
|
| 333 |
input_format: SAME
|
| 334 |
instance: LayerNorm
|
| 335 |
output_format: SAME
|
| 336 |
weight_format: SAME
|
| 337 |
transformer.h.12.ln_2:
|
| 338 |
-
approximation_function:
|
| 339 |
bias_format: SAME
|
| 340 |
input_format: SAME
|
| 341 |
instance: LayerNorm
|
| 342 |
output_format: SAME
|
| 343 |
weight_format: SAME
|
| 344 |
transformer.h.12.mlp.act:
|
| 345 |
-
approximation_function:
|
| 346 |
input_format: SAME
|
| 347 |
instance: GELU
|
| 348 |
output_format: SAME
|
|
@@ -394,26 +394,26 @@ model:
|
|
| 394 |
instance: Dropout
|
| 395 |
output_format: SAME
|
| 396 |
transformer.h.13.attn.softmax:
|
| 397 |
-
approximation_function:
|
| 398 |
input_format: SAME
|
| 399 |
instance: Softmax
|
| 400 |
output_format: SAME
|
| 401 |
transformer.h.13.ln_1:
|
| 402 |
-
approximation_function:
|
| 403 |
bias_format: SAME
|
| 404 |
input_format: SAME
|
| 405 |
instance: LayerNorm
|
| 406 |
output_format: SAME
|
| 407 |
weight_format: SAME
|
| 408 |
transformer.h.13.ln_2:
|
| 409 |
-
approximation_function:
|
| 410 |
bias_format: SAME
|
| 411 |
input_format: SAME
|
| 412 |
instance: LayerNorm
|
| 413 |
output_format: SAME
|
| 414 |
weight_format: SAME
|
| 415 |
transformer.h.13.mlp.act:
|
| 416 |
-
approximation_function:
|
| 417 |
input_format: SAME
|
| 418 |
instance: GELU
|
| 419 |
output_format: SAME
|
|
@@ -465,26 +465,26 @@ model:
|
|
| 465 |
instance: Dropout
|
| 466 |
output_format: SAME
|
| 467 |
transformer.h.14.attn.softmax:
|
| 468 |
-
approximation_function:
|
| 469 |
input_format: SAME
|
| 470 |
instance: Softmax
|
| 471 |
output_format: SAME
|
| 472 |
transformer.h.14.ln_1:
|
| 473 |
-
approximation_function:
|
| 474 |
bias_format: SAME
|
| 475 |
input_format: SAME
|
| 476 |
instance: LayerNorm
|
| 477 |
output_format: SAME
|
| 478 |
weight_format: SAME
|
| 479 |
transformer.h.14.ln_2:
|
| 480 |
-
approximation_function:
|
| 481 |
bias_format: SAME
|
| 482 |
input_format: SAME
|
| 483 |
instance: LayerNorm
|
| 484 |
output_format: SAME
|
| 485 |
weight_format: SAME
|
| 486 |
transformer.h.14.mlp.act:
|
| 487 |
-
approximation_function:
|
| 488 |
input_format: SAME
|
| 489 |
instance: GELU
|
| 490 |
output_format: SAME
|
|
@@ -536,26 +536,26 @@ model:
|
|
| 536 |
instance: Dropout
|
| 537 |
output_format: SAME
|
| 538 |
transformer.h.15.attn.softmax:
|
| 539 |
-
approximation_function:
|
| 540 |
input_format: SAME
|
| 541 |
instance: Softmax
|
| 542 |
output_format: SAME
|
| 543 |
transformer.h.15.ln_1:
|
| 544 |
-
approximation_function:
|
| 545 |
bias_format: SAME
|
| 546 |
input_format: SAME
|
| 547 |
instance: LayerNorm
|
| 548 |
output_format: SAME
|
| 549 |
weight_format: SAME
|
| 550 |
transformer.h.15.ln_2:
|
| 551 |
-
approximation_function:
|
| 552 |
bias_format: SAME
|
| 553 |
input_format: SAME
|
| 554 |
instance: LayerNorm
|
| 555 |
output_format: SAME
|
| 556 |
weight_format: SAME
|
| 557 |
transformer.h.15.mlp.act:
|
| 558 |
-
approximation_function:
|
| 559 |
input_format: SAME
|
| 560 |
instance: GELU
|
| 561 |
output_format: SAME
|
|
@@ -607,26 +607,26 @@ model:
|
|
| 607 |
instance: Dropout
|
| 608 |
output_format: SAME
|
| 609 |
transformer.h.16.attn.softmax:
|
| 610 |
-
approximation_function:
|
| 611 |
input_format: SAME
|
| 612 |
instance: Softmax
|
| 613 |
output_format: SAME
|
| 614 |
transformer.h.16.ln_1:
|
| 615 |
-
approximation_function:
|
| 616 |
bias_format: SAME
|
| 617 |
input_format: SAME
|
| 618 |
instance: LayerNorm
|
| 619 |
output_format: SAME
|
| 620 |
weight_format: SAME
|
| 621 |
transformer.h.16.ln_2:
|
| 622 |
-
approximation_function:
|
| 623 |
bias_format: SAME
|
| 624 |
input_format: SAME
|
| 625 |
instance: LayerNorm
|
| 626 |
output_format: SAME
|
| 627 |
weight_format: SAME
|
| 628 |
transformer.h.16.mlp.act:
|
| 629 |
-
approximation_function:
|
| 630 |
input_format: SAME
|
| 631 |
instance: GELU
|
| 632 |
output_format: SAME
|
|
@@ -678,26 +678,26 @@ model:
|
|
| 678 |
instance: Dropout
|
| 679 |
output_format: SAME
|
| 680 |
transformer.h.17.attn.softmax:
|
| 681 |
-
approximation_function:
|
| 682 |
input_format: SAME
|
| 683 |
instance: Softmax
|
| 684 |
output_format: SAME
|
| 685 |
transformer.h.17.ln_1:
|
| 686 |
-
approximation_function:
|
| 687 |
bias_format: SAME
|
| 688 |
input_format: SAME
|
| 689 |
instance: LayerNorm
|
| 690 |
output_format: SAME
|
| 691 |
weight_format: SAME
|
| 692 |
transformer.h.17.ln_2:
|
| 693 |
-
approximation_function:
|
| 694 |
bias_format: SAME
|
| 695 |
input_format: SAME
|
| 696 |
instance: LayerNorm
|
| 697 |
output_format: SAME
|
| 698 |
weight_format: SAME
|
| 699 |
transformer.h.17.mlp.act:
|
| 700 |
-
approximation_function:
|
| 701 |
input_format: SAME
|
| 702 |
instance: GELU
|
| 703 |
output_format: SAME
|
|
@@ -749,26 +749,26 @@ model:
|
|
| 749 |
instance: Dropout
|
| 750 |
output_format: SAME
|
| 751 |
transformer.h.18.attn.softmax:
|
| 752 |
-
approximation_function:
|
| 753 |
input_format: SAME
|
| 754 |
instance: Softmax
|
| 755 |
output_format: SAME
|
| 756 |
transformer.h.18.ln_1:
|
| 757 |
-
approximation_function:
|
| 758 |
bias_format: SAME
|
| 759 |
input_format: SAME
|
| 760 |
instance: LayerNorm
|
| 761 |
output_format: SAME
|
| 762 |
weight_format: SAME
|
| 763 |
transformer.h.18.ln_2:
|
| 764 |
-
approximation_function:
|
| 765 |
bias_format: SAME
|
| 766 |
input_format: SAME
|
| 767 |
instance: LayerNorm
|
| 768 |
output_format: SAME
|
| 769 |
weight_format: SAME
|
| 770 |
transformer.h.18.mlp.act:
|
| 771 |
-
approximation_function:
|
| 772 |
input_format: SAME
|
| 773 |
instance: GELU
|
| 774 |
output_format: SAME
|
|
@@ -820,26 +820,26 @@ model:
|
|
| 820 |
instance: Dropout
|
| 821 |
output_format: SAME
|
| 822 |
transformer.h.19.attn.softmax:
|
| 823 |
-
approximation_function:
|
| 824 |
input_format: SAME
|
| 825 |
instance: Softmax
|
| 826 |
output_format: SAME
|
| 827 |
transformer.h.19.ln_1:
|
| 828 |
-
approximation_function:
|
| 829 |
bias_format: SAME
|
| 830 |
input_format: SAME
|
| 831 |
instance: LayerNorm
|
| 832 |
output_format: SAME
|
| 833 |
weight_format: SAME
|
| 834 |
transformer.h.19.ln_2:
|
| 835 |
-
approximation_function:
|
| 836 |
bias_format: SAME
|
| 837 |
input_format: SAME
|
| 838 |
instance: LayerNorm
|
| 839 |
output_format: SAME
|
| 840 |
weight_format: SAME
|
| 841 |
transformer.h.19.mlp.act:
|
| 842 |
-
approximation_function:
|
| 843 |
input_format: SAME
|
| 844 |
instance: GELU
|
| 845 |
output_format: SAME
|
|
@@ -891,26 +891,26 @@ model:
|
|
| 891 |
instance: Dropout
|
| 892 |
output_format: SAME
|
| 893 |
transformer.h.2.attn.softmax:
|
| 894 |
-
approximation_function:
|
| 895 |
input_format: SAME
|
| 896 |
instance: Softmax
|
| 897 |
output_format: SAME
|
| 898 |
transformer.h.2.ln_1:
|
| 899 |
-
approximation_function:
|
| 900 |
bias_format: SAME
|
| 901 |
input_format: SAME
|
| 902 |
instance: LayerNorm
|
| 903 |
output_format: SAME
|
| 904 |
weight_format: SAME
|
| 905 |
transformer.h.2.ln_2:
|
| 906 |
-
approximation_function:
|
| 907 |
bias_format: SAME
|
| 908 |
input_format: SAME
|
| 909 |
instance: LayerNorm
|
| 910 |
output_format: SAME
|
| 911 |
weight_format: SAME
|
| 912 |
transformer.h.2.mlp.act:
|
| 913 |
-
approximation_function:
|
| 914 |
input_format: SAME
|
| 915 |
instance: GELU
|
| 916 |
output_format: SAME
|
|
@@ -962,26 +962,26 @@ model:
|
|
| 962 |
instance: Dropout
|
| 963 |
output_format: SAME
|
| 964 |
transformer.h.20.attn.softmax:
|
| 965 |
-
approximation_function:
|
| 966 |
input_format: SAME
|
| 967 |
instance: Softmax
|
| 968 |
output_format: SAME
|
| 969 |
transformer.h.20.ln_1:
|
| 970 |
-
approximation_function:
|
| 971 |
bias_format: SAME
|
| 972 |
input_format: SAME
|
| 973 |
instance: LayerNorm
|
| 974 |
output_format: SAME
|
| 975 |
weight_format: SAME
|
| 976 |
transformer.h.20.ln_2:
|
| 977 |
-
approximation_function:
|
| 978 |
bias_format: SAME
|
| 979 |
input_format: SAME
|
| 980 |
instance: LayerNorm
|
| 981 |
output_format: SAME
|
| 982 |
weight_format: SAME
|
| 983 |
transformer.h.20.mlp.act:
|
| 984 |
-
approximation_function:
|
| 985 |
input_format: SAME
|
| 986 |
instance: GELU
|
| 987 |
output_format: SAME
|
|
@@ -1033,26 +1033,26 @@ model:
|
|
| 1033 |
instance: Dropout
|
| 1034 |
output_format: SAME
|
| 1035 |
transformer.h.21.attn.softmax:
|
| 1036 |
-
approximation_function:
|
| 1037 |
input_format: SAME
|
| 1038 |
instance: Softmax
|
| 1039 |
output_format: SAME
|
| 1040 |
transformer.h.21.ln_1:
|
| 1041 |
-
approximation_function:
|
| 1042 |
bias_format: SAME
|
| 1043 |
input_format: SAME
|
| 1044 |
instance: LayerNorm
|
| 1045 |
output_format: SAME
|
| 1046 |
weight_format: SAME
|
| 1047 |
transformer.h.21.ln_2:
|
| 1048 |
-
approximation_function:
|
| 1049 |
bias_format: SAME
|
| 1050 |
input_format: SAME
|
| 1051 |
instance: LayerNorm
|
| 1052 |
output_format: SAME
|
| 1053 |
weight_format: SAME
|
| 1054 |
transformer.h.21.mlp.act:
|
| 1055 |
-
approximation_function:
|
| 1056 |
input_format: SAME
|
| 1057 |
instance: GELU
|
| 1058 |
output_format: SAME
|
|
@@ -1104,26 +1104,26 @@ model:
|
|
| 1104 |
instance: Dropout
|
| 1105 |
output_format: SAME
|
| 1106 |
transformer.h.22.attn.softmax:
|
| 1107 |
-
approximation_function:
|
| 1108 |
input_format: SAME
|
| 1109 |
instance: Softmax
|
| 1110 |
output_format: SAME
|
| 1111 |
transformer.h.22.ln_1:
|
| 1112 |
-
approximation_function:
|
| 1113 |
bias_format: SAME
|
| 1114 |
input_format: SAME
|
| 1115 |
instance: LayerNorm
|
| 1116 |
output_format: SAME
|
| 1117 |
weight_format: SAME
|
| 1118 |
transformer.h.22.ln_2:
|
| 1119 |
-
approximation_function:
|
| 1120 |
bias_format: SAME
|
| 1121 |
input_format: SAME
|
| 1122 |
instance: LayerNorm
|
| 1123 |
output_format: SAME
|
| 1124 |
weight_format: SAME
|
| 1125 |
transformer.h.22.mlp.act:
|
| 1126 |
-
approximation_function:
|
| 1127 |
input_format: SAME
|
| 1128 |
instance: GELU
|
| 1129 |
output_format: SAME
|
|
@@ -1175,26 +1175,26 @@ model:
|
|
| 1175 |
instance: Dropout
|
| 1176 |
output_format: SAME
|
| 1177 |
transformer.h.23.attn.softmax:
|
| 1178 |
-
approximation_function:
|
| 1179 |
input_format: SAME
|
| 1180 |
instance: Softmax
|
| 1181 |
output_format: SAME
|
| 1182 |
transformer.h.23.ln_1:
|
| 1183 |
-
approximation_function:
|
| 1184 |
bias_format: SAME
|
| 1185 |
input_format: SAME
|
| 1186 |
instance: LayerNorm
|
| 1187 |
output_format: SAME
|
| 1188 |
weight_format: SAME
|
| 1189 |
transformer.h.23.ln_2:
|
| 1190 |
-
approximation_function:
|
| 1191 |
bias_format: SAME
|
| 1192 |
input_format: SAME
|
| 1193 |
instance: LayerNorm
|
| 1194 |
output_format: SAME
|
| 1195 |
weight_format: SAME
|
| 1196 |
transformer.h.23.mlp.act:
|
| 1197 |
-
approximation_function:
|
| 1198 |
input_format: SAME
|
| 1199 |
instance: GELU
|
| 1200 |
output_format: SAME
|
|
@@ -1246,26 +1246,26 @@ model:
|
|
| 1246 |
instance: Dropout
|
| 1247 |
output_format: SAME
|
| 1248 |
transformer.h.24.attn.softmax:
|
| 1249 |
-
approximation_function:
|
| 1250 |
input_format: SAME
|
| 1251 |
instance: Softmax
|
| 1252 |
output_format: SAME
|
| 1253 |
transformer.h.24.ln_1:
|
| 1254 |
-
approximation_function:
|
| 1255 |
bias_format: SAME
|
| 1256 |
input_format: SAME
|
| 1257 |
instance: LayerNorm
|
| 1258 |
output_format: SAME
|
| 1259 |
weight_format: SAME
|
| 1260 |
transformer.h.24.ln_2:
|
| 1261 |
-
approximation_function:
|
| 1262 |
bias_format: SAME
|
| 1263 |
input_format: SAME
|
| 1264 |
instance: LayerNorm
|
| 1265 |
output_format: SAME
|
| 1266 |
weight_format: SAME
|
| 1267 |
transformer.h.24.mlp.act:
|
| 1268 |
-
approximation_function:
|
| 1269 |
input_format: SAME
|
| 1270 |
instance: GELU
|
| 1271 |
output_format: SAME
|
|
@@ -1317,26 +1317,26 @@ model:
|
|
| 1317 |
instance: Dropout
|
| 1318 |
output_format: SAME
|
| 1319 |
transformer.h.25.attn.softmax:
|
| 1320 |
-
approximation_function:
|
| 1321 |
input_format: SAME
|
| 1322 |
instance: Softmax
|
| 1323 |
output_format: SAME
|
| 1324 |
transformer.h.25.ln_1:
|
| 1325 |
-
approximation_function:
|
| 1326 |
bias_format: SAME
|
| 1327 |
input_format: SAME
|
| 1328 |
instance: LayerNorm
|
| 1329 |
output_format: SAME
|
| 1330 |
weight_format: SAME
|
| 1331 |
transformer.h.25.ln_2:
|
| 1332 |
-
approximation_function:
|
| 1333 |
bias_format: SAME
|
| 1334 |
input_format: SAME
|
| 1335 |
instance: LayerNorm
|
| 1336 |
output_format: SAME
|
| 1337 |
weight_format: SAME
|
| 1338 |
transformer.h.25.mlp.act:
|
| 1339 |
-
approximation_function:
|
| 1340 |
input_format: SAME
|
| 1341 |
instance: GELU
|
| 1342 |
output_format: SAME
|
|
@@ -1388,26 +1388,26 @@ model:
|
|
| 1388 |
instance: Dropout
|
| 1389 |
output_format: SAME
|
| 1390 |
transformer.h.26.attn.softmax:
|
| 1391 |
-
approximation_function:
|
| 1392 |
input_format: SAME
|
| 1393 |
instance: Softmax
|
| 1394 |
output_format: SAME
|
| 1395 |
transformer.h.26.ln_1:
|
| 1396 |
-
approximation_function:
|
| 1397 |
bias_format: SAME
|
| 1398 |
input_format: SAME
|
| 1399 |
instance: LayerNorm
|
| 1400 |
output_format: SAME
|
| 1401 |
weight_format: SAME
|
| 1402 |
transformer.h.26.ln_2:
|
| 1403 |
-
approximation_function:
|
| 1404 |
bias_format: SAME
|
| 1405 |
input_format: SAME
|
| 1406 |
instance: LayerNorm
|
| 1407 |
output_format: SAME
|
| 1408 |
weight_format: SAME
|
| 1409 |
transformer.h.26.mlp.act:
|
| 1410 |
-
approximation_function:
|
| 1411 |
input_format: SAME
|
| 1412 |
instance: GELU
|
| 1413 |
output_format: SAME
|
|
@@ -1459,26 +1459,26 @@ model:
|
|
| 1459 |
instance: Dropout
|
| 1460 |
output_format: SAME
|
| 1461 |
transformer.h.27.attn.softmax:
|
| 1462 |
-
approximation_function:
|
| 1463 |
input_format: SAME
|
| 1464 |
instance: Softmax
|
| 1465 |
output_format: SAME
|
| 1466 |
transformer.h.27.ln_1:
|
| 1467 |
-
approximation_function:
|
| 1468 |
bias_format: SAME
|
| 1469 |
input_format: SAME
|
| 1470 |
instance: LayerNorm
|
| 1471 |
output_format: SAME
|
| 1472 |
weight_format: SAME
|
| 1473 |
transformer.h.27.ln_2:
|
| 1474 |
-
approximation_function:
|
| 1475 |
bias_format: SAME
|
| 1476 |
input_format: SAME
|
| 1477 |
instance: LayerNorm
|
| 1478 |
output_format: SAME
|
| 1479 |
weight_format: SAME
|
| 1480 |
transformer.h.27.mlp.act:
|
| 1481 |
-
approximation_function:
|
| 1482 |
input_format: SAME
|
| 1483 |
instance: GELU
|
| 1484 |
output_format: SAME
|
|
@@ -1530,26 +1530,26 @@ model:
|
|
| 1530 |
instance: Dropout
|
| 1531 |
output_format: SAME
|
| 1532 |
transformer.h.28.attn.softmax:
|
| 1533 |
-
approximation_function:
|
| 1534 |
input_format: SAME
|
| 1535 |
instance: Softmax
|
| 1536 |
output_format: SAME
|
| 1537 |
transformer.h.28.ln_1:
|
| 1538 |
-
approximation_function:
|
| 1539 |
bias_format: SAME
|
| 1540 |
input_format: SAME
|
| 1541 |
instance: LayerNorm
|
| 1542 |
output_format: SAME
|
| 1543 |
weight_format: SAME
|
| 1544 |
transformer.h.28.ln_2:
|
| 1545 |
-
approximation_function:
|
| 1546 |
bias_format: SAME
|
| 1547 |
input_format: SAME
|
| 1548 |
instance: LayerNorm
|
| 1549 |
output_format: SAME
|
| 1550 |
weight_format: SAME
|
| 1551 |
transformer.h.28.mlp.act:
|
| 1552 |
-
approximation_function:
|
| 1553 |
input_format: SAME
|
| 1554 |
instance: GELU
|
| 1555 |
output_format: SAME
|
|
@@ -1601,26 +1601,26 @@ model:
|
|
| 1601 |
instance: Dropout
|
| 1602 |
output_format: SAME
|
| 1603 |
transformer.h.29.attn.softmax:
|
| 1604 |
-
approximation_function:
|
| 1605 |
input_format: SAME
|
| 1606 |
instance: Softmax
|
| 1607 |
output_format: SAME
|
| 1608 |
transformer.h.29.ln_1:
|
| 1609 |
-
approximation_function:
|
| 1610 |
bias_format: SAME
|
| 1611 |
input_format: SAME
|
| 1612 |
instance: LayerNorm
|
| 1613 |
output_format: SAME
|
| 1614 |
weight_format: SAME
|
| 1615 |
transformer.h.29.ln_2:
|
| 1616 |
-
approximation_function:
|
| 1617 |
bias_format: SAME
|
| 1618 |
input_format: SAME
|
| 1619 |
instance: LayerNorm
|
| 1620 |
output_format: SAME
|
| 1621 |
weight_format: SAME
|
| 1622 |
transformer.h.29.mlp.act:
|
| 1623 |
-
approximation_function:
|
| 1624 |
input_format: SAME
|
| 1625 |
instance: GELU
|
| 1626 |
output_format: SAME
|
|
@@ -1672,26 +1672,26 @@ model:
|
|
| 1672 |
instance: Dropout
|
| 1673 |
output_format: SAME
|
| 1674 |
transformer.h.3.attn.softmax:
|
| 1675 |
-
approximation_function:
|
| 1676 |
input_format: SAME
|
| 1677 |
instance: Softmax
|
| 1678 |
output_format: SAME
|
| 1679 |
transformer.h.3.ln_1:
|
| 1680 |
-
approximation_function:
|
| 1681 |
bias_format: SAME
|
| 1682 |
input_format: SAME
|
| 1683 |
instance: LayerNorm
|
| 1684 |
output_format: SAME
|
| 1685 |
weight_format: SAME
|
| 1686 |
transformer.h.3.ln_2:
|
| 1687 |
-
approximation_function:
|
| 1688 |
bias_format: SAME
|
| 1689 |
input_format: SAME
|
| 1690 |
instance: LayerNorm
|
| 1691 |
output_format: SAME
|
| 1692 |
weight_format: SAME
|
| 1693 |
transformer.h.3.mlp.act:
|
| 1694 |
-
approximation_function:
|
| 1695 |
input_format: SAME
|
| 1696 |
instance: GELU
|
| 1697 |
output_format: SAME
|
|
@@ -1743,26 +1743,26 @@ model:
|
|
| 1743 |
instance: Dropout
|
| 1744 |
output_format: SAME
|
| 1745 |
transformer.h.30.attn.softmax:
|
| 1746 |
-
approximation_function:
|
| 1747 |
input_format: SAME
|
| 1748 |
instance: Softmax
|
| 1749 |
output_format: SAME
|
| 1750 |
transformer.h.30.ln_1:
|
| 1751 |
-
approximation_function:
|
| 1752 |
bias_format: SAME
|
| 1753 |
input_format: SAME
|
| 1754 |
instance: LayerNorm
|
| 1755 |
output_format: SAME
|
| 1756 |
weight_format: SAME
|
| 1757 |
transformer.h.30.ln_2:
|
| 1758 |
-
approximation_function:
|
| 1759 |
bias_format: SAME
|
| 1760 |
input_format: SAME
|
| 1761 |
instance: LayerNorm
|
| 1762 |
output_format: SAME
|
| 1763 |
weight_format: SAME
|
| 1764 |
transformer.h.30.mlp.act:
|
| 1765 |
-
approximation_function:
|
| 1766 |
input_format: SAME
|
| 1767 |
instance: GELU
|
| 1768 |
output_format: SAME
|
|
@@ -1814,26 +1814,26 @@ model:
|
|
| 1814 |
instance: Dropout
|
| 1815 |
output_format: SAME
|
| 1816 |
transformer.h.31.attn.softmax:
|
| 1817 |
-
approximation_function:
|
| 1818 |
input_format: SAME
|
| 1819 |
instance: Softmax
|
| 1820 |
output_format: SAME
|
| 1821 |
transformer.h.31.ln_1:
|
| 1822 |
-
approximation_function:
|
| 1823 |
bias_format: SAME
|
| 1824 |
input_format: SAME
|
| 1825 |
instance: LayerNorm
|
| 1826 |
output_format: SAME
|
| 1827 |
weight_format: SAME
|
| 1828 |
transformer.h.31.ln_2:
|
| 1829 |
-
approximation_function:
|
| 1830 |
bias_format: SAME
|
| 1831 |
input_format: SAME
|
| 1832 |
instance: LayerNorm
|
| 1833 |
output_format: SAME
|
| 1834 |
weight_format: SAME
|
| 1835 |
transformer.h.31.mlp.act:
|
| 1836 |
-
approximation_function:
|
| 1837 |
input_format: SAME
|
| 1838 |
instance: GELU
|
| 1839 |
output_format: SAME
|
|
@@ -1885,26 +1885,26 @@ model:
|
|
| 1885 |
instance: Dropout
|
| 1886 |
output_format: SAME
|
| 1887 |
transformer.h.32.attn.softmax:
|
| 1888 |
-
approximation_function:
|
| 1889 |
input_format: SAME
|
| 1890 |
instance: Softmax
|
| 1891 |
output_format: SAME
|
| 1892 |
transformer.h.32.ln_1:
|
| 1893 |
-
approximation_function:
|
| 1894 |
bias_format: SAME
|
| 1895 |
input_format: SAME
|
| 1896 |
instance: LayerNorm
|
| 1897 |
output_format: SAME
|
| 1898 |
weight_format: SAME
|
| 1899 |
transformer.h.32.ln_2:
|
| 1900 |
-
approximation_function:
|
| 1901 |
bias_format: SAME
|
| 1902 |
input_format: SAME
|
| 1903 |
instance: LayerNorm
|
| 1904 |
output_format: SAME
|
| 1905 |
weight_format: SAME
|
| 1906 |
transformer.h.32.mlp.act:
|
| 1907 |
-
approximation_function:
|
| 1908 |
input_format: SAME
|
| 1909 |
instance: GELU
|
| 1910 |
output_format: SAME
|
|
@@ -1956,26 +1956,26 @@ model:
|
|
| 1956 |
instance: Dropout
|
| 1957 |
output_format: SAME
|
| 1958 |
transformer.h.33.attn.softmax:
|
| 1959 |
-
approximation_function:
|
| 1960 |
input_format: SAME
|
| 1961 |
instance: Softmax
|
| 1962 |
output_format: SAME
|
| 1963 |
transformer.h.33.ln_1:
|
| 1964 |
-
approximation_function:
|
| 1965 |
bias_format: SAME
|
| 1966 |
input_format: SAME
|
| 1967 |
instance: LayerNorm
|
| 1968 |
output_format: SAME
|
| 1969 |
weight_format: SAME
|
| 1970 |
transformer.h.33.ln_2:
|
| 1971 |
-
approximation_function:
|
| 1972 |
bias_format: SAME
|
| 1973 |
input_format: SAME
|
| 1974 |
instance: LayerNorm
|
| 1975 |
output_format: SAME
|
| 1976 |
weight_format: SAME
|
| 1977 |
transformer.h.33.mlp.act:
|
| 1978 |
-
approximation_function:
|
| 1979 |
input_format: SAME
|
| 1980 |
instance: GELU
|
| 1981 |
output_format: SAME
|
|
@@ -2027,26 +2027,26 @@ model:
|
|
| 2027 |
instance: Dropout
|
| 2028 |
output_format: SAME
|
| 2029 |
transformer.h.34.attn.softmax:
|
| 2030 |
-
approximation_function:
|
| 2031 |
input_format: SAME
|
| 2032 |
instance: Softmax
|
| 2033 |
output_format: SAME
|
| 2034 |
transformer.h.34.ln_1:
|
| 2035 |
-
approximation_function:
|
| 2036 |
bias_format: SAME
|
| 2037 |
input_format: SAME
|
| 2038 |
instance: LayerNorm
|
| 2039 |
output_format: SAME
|
| 2040 |
weight_format: SAME
|
| 2041 |
transformer.h.34.ln_2:
|
| 2042 |
-
approximation_function:
|
| 2043 |
bias_format: SAME
|
| 2044 |
input_format: SAME
|
| 2045 |
instance: LayerNorm
|
| 2046 |
output_format: SAME
|
| 2047 |
weight_format: SAME
|
| 2048 |
transformer.h.34.mlp.act:
|
| 2049 |
-
approximation_function:
|
| 2050 |
input_format: SAME
|
| 2051 |
instance: GELU
|
| 2052 |
output_format: SAME
|
|
@@ -2098,26 +2098,26 @@ model:
|
|
| 2098 |
instance: Dropout
|
| 2099 |
output_format: SAME
|
| 2100 |
transformer.h.35.attn.softmax:
|
| 2101 |
-
approximation_function:
|
| 2102 |
input_format: SAME
|
| 2103 |
instance: Softmax
|
| 2104 |
output_format: SAME
|
| 2105 |
transformer.h.35.ln_1:
|
| 2106 |
-
approximation_function:
|
| 2107 |
bias_format: SAME
|
| 2108 |
input_format: SAME
|
| 2109 |
instance: LayerNorm
|
| 2110 |
output_format: SAME
|
| 2111 |
weight_format: SAME
|
| 2112 |
transformer.h.35.ln_2:
|
| 2113 |
-
approximation_function:
|
| 2114 |
bias_format: SAME
|
| 2115 |
input_format: SAME
|
| 2116 |
instance: LayerNorm
|
| 2117 |
output_format: SAME
|
| 2118 |
weight_format: SAME
|
| 2119 |
transformer.h.35.mlp.act:
|
| 2120 |
-
approximation_function:
|
| 2121 |
input_format: SAME
|
| 2122 |
instance: GELU
|
| 2123 |
output_format: SAME
|
|
@@ -2169,26 +2169,26 @@ model:
|
|
| 2169 |
instance: Dropout
|
| 2170 |
output_format: SAME
|
| 2171 |
transformer.h.36.attn.softmax:
|
| 2172 |
-
approximation_function:
|
| 2173 |
input_format: SAME
|
| 2174 |
instance: Softmax
|
| 2175 |
output_format: SAME
|
| 2176 |
transformer.h.36.ln_1:
|
| 2177 |
-
approximation_function:
|
| 2178 |
bias_format: SAME
|
| 2179 |
input_format: SAME
|
| 2180 |
instance: LayerNorm
|
| 2181 |
output_format: SAME
|
| 2182 |
weight_format: SAME
|
| 2183 |
transformer.h.36.ln_2:
|
| 2184 |
-
approximation_function:
|
| 2185 |
bias_format: SAME
|
| 2186 |
input_format: SAME
|
| 2187 |
instance: LayerNorm
|
| 2188 |
output_format: SAME
|
| 2189 |
weight_format: SAME
|
| 2190 |
transformer.h.36.mlp.act:
|
| 2191 |
-
approximation_function:
|
| 2192 |
input_format: SAME
|
| 2193 |
instance: GELU
|
| 2194 |
output_format: SAME
|
|
@@ -2240,26 +2240,26 @@ model:
|
|
| 2240 |
instance: Dropout
|
| 2241 |
output_format: SAME
|
| 2242 |
transformer.h.37.attn.softmax:
|
| 2243 |
-
approximation_function:
|
| 2244 |
input_format: SAME
|
| 2245 |
instance: Softmax
|
| 2246 |
output_format: SAME
|
| 2247 |
transformer.h.37.ln_1:
|
| 2248 |
-
approximation_function:
|
| 2249 |
bias_format: SAME
|
| 2250 |
input_format: SAME
|
| 2251 |
instance: LayerNorm
|
| 2252 |
output_format: SAME
|
| 2253 |
weight_format: SAME
|
| 2254 |
transformer.h.37.ln_2:
|
| 2255 |
-
approximation_function:
|
| 2256 |
bias_format: SAME
|
| 2257 |
input_format: SAME
|
| 2258 |
instance: LayerNorm
|
| 2259 |
output_format: SAME
|
| 2260 |
weight_format: SAME
|
| 2261 |
transformer.h.37.mlp.act:
|
| 2262 |
-
approximation_function:
|
| 2263 |
input_format: SAME
|
| 2264 |
instance: GELU
|
| 2265 |
output_format: SAME
|
|
@@ -2311,26 +2311,26 @@ model:
|
|
| 2311 |
instance: Dropout
|
| 2312 |
output_format: SAME
|
| 2313 |
transformer.h.38.attn.softmax:
|
| 2314 |
-
approximation_function:
|
| 2315 |
input_format: SAME
|
| 2316 |
instance: Softmax
|
| 2317 |
output_format: SAME
|
| 2318 |
transformer.h.38.ln_1:
|
| 2319 |
-
approximation_function:
|
| 2320 |
bias_format: SAME
|
| 2321 |
input_format: SAME
|
| 2322 |
instance: LayerNorm
|
| 2323 |
output_format: SAME
|
| 2324 |
weight_format: SAME
|
| 2325 |
transformer.h.38.ln_2:
|
| 2326 |
-
approximation_function:
|
| 2327 |
bias_format: SAME
|
| 2328 |
input_format: SAME
|
| 2329 |
instance: LayerNorm
|
| 2330 |
output_format: SAME
|
| 2331 |
weight_format: SAME
|
| 2332 |
transformer.h.38.mlp.act:
|
| 2333 |
-
approximation_function:
|
| 2334 |
input_format: SAME
|
| 2335 |
instance: GELU
|
| 2336 |
output_format: SAME
|
|
@@ -2382,26 +2382,26 @@ model:
|
|
| 2382 |
instance: Dropout
|
| 2383 |
output_format: SAME
|
| 2384 |
transformer.h.39.attn.softmax:
|
| 2385 |
-
approximation_function:
|
| 2386 |
input_format: SAME
|
| 2387 |
instance: Softmax
|
| 2388 |
output_format: SAME
|
| 2389 |
transformer.h.39.ln_1:
|
| 2390 |
-
approximation_function:
|
| 2391 |
bias_format: SAME
|
| 2392 |
input_format: SAME
|
| 2393 |
instance: LayerNorm
|
| 2394 |
output_format: SAME
|
| 2395 |
weight_format: SAME
|
| 2396 |
transformer.h.39.ln_2:
|
| 2397 |
-
approximation_function:
|
| 2398 |
bias_format: SAME
|
| 2399 |
input_format: SAME
|
| 2400 |
instance: LayerNorm
|
| 2401 |
output_format: SAME
|
| 2402 |
weight_format: SAME
|
| 2403 |
transformer.h.39.mlp.act:
|
| 2404 |
-
approximation_function:
|
| 2405 |
input_format: SAME
|
| 2406 |
instance: GELU
|
| 2407 |
output_format: SAME
|
|
@@ -2453,26 +2453,26 @@ model:
|
|
| 2453 |
instance: Dropout
|
| 2454 |
output_format: SAME
|
| 2455 |
transformer.h.4.attn.softmax:
|
| 2456 |
-
approximation_function:
|
| 2457 |
input_format: SAME
|
| 2458 |
instance: Softmax
|
| 2459 |
output_format: SAME
|
| 2460 |
transformer.h.4.ln_1:
|
| 2461 |
-
approximation_function:
|
| 2462 |
bias_format: SAME
|
| 2463 |
input_format: SAME
|
| 2464 |
instance: LayerNorm
|
| 2465 |
output_format: SAME
|
| 2466 |
weight_format: SAME
|
| 2467 |
transformer.h.4.ln_2:
|
| 2468 |
-
approximation_function:
|
| 2469 |
bias_format: SAME
|
| 2470 |
input_format: SAME
|
| 2471 |
instance: LayerNorm
|
| 2472 |
output_format: SAME
|
| 2473 |
weight_format: SAME
|
| 2474 |
transformer.h.4.mlp.act:
|
| 2475 |
-
approximation_function:
|
| 2476 |
input_format: SAME
|
| 2477 |
instance: GELU
|
| 2478 |
output_format: SAME
|
|
@@ -2524,26 +2524,26 @@ model:
|
|
| 2524 |
instance: Dropout
|
| 2525 |
output_format: SAME
|
| 2526 |
transformer.h.40.attn.softmax:
|
| 2527 |
-
approximation_function:
|
| 2528 |
input_format: SAME
|
| 2529 |
instance: Softmax
|
| 2530 |
output_format: SAME
|
| 2531 |
transformer.h.40.ln_1:
|
| 2532 |
-
approximation_function:
|
| 2533 |
bias_format: SAME
|
| 2534 |
input_format: SAME
|
| 2535 |
instance: LayerNorm
|
| 2536 |
output_format: SAME
|
| 2537 |
weight_format: SAME
|
| 2538 |
transformer.h.40.ln_2:
|
| 2539 |
-
approximation_function:
|
| 2540 |
bias_format: SAME
|
| 2541 |
input_format: SAME
|
| 2542 |
instance: LayerNorm
|
| 2543 |
output_format: SAME
|
| 2544 |
weight_format: SAME
|
| 2545 |
transformer.h.40.mlp.act:
|
| 2546 |
-
approximation_function:
|
| 2547 |
input_format: SAME
|
| 2548 |
instance: GELU
|
| 2549 |
output_format: SAME
|
|
@@ -2595,26 +2595,26 @@ model:
|
|
| 2595 |
instance: Dropout
|
| 2596 |
output_format: SAME
|
| 2597 |
transformer.h.41.attn.softmax:
|
| 2598 |
-
approximation_function:
|
| 2599 |
input_format: SAME
|
| 2600 |
instance: Softmax
|
| 2601 |
output_format: SAME
|
| 2602 |
transformer.h.41.ln_1:
|
| 2603 |
-
approximation_function:
|
| 2604 |
bias_format: SAME
|
| 2605 |
input_format: SAME
|
| 2606 |
instance: LayerNorm
|
| 2607 |
output_format: SAME
|
| 2608 |
weight_format: SAME
|
| 2609 |
transformer.h.41.ln_2:
|
| 2610 |
-
approximation_function:
|
| 2611 |
bias_format: SAME
|
| 2612 |
input_format: SAME
|
| 2613 |
instance: LayerNorm
|
| 2614 |
output_format: SAME
|
| 2615 |
weight_format: SAME
|
| 2616 |
transformer.h.41.mlp.act:
|
| 2617 |
-
approximation_function:
|
| 2618 |
input_format: SAME
|
| 2619 |
instance: GELU
|
| 2620 |
output_format: SAME
|
|
@@ -2666,26 +2666,26 @@ model:
|
|
| 2666 |
instance: Dropout
|
| 2667 |
output_format: SAME
|
| 2668 |
transformer.h.42.attn.softmax:
|
| 2669 |
-
approximation_function:
|
| 2670 |
input_format: SAME
|
| 2671 |
instance: Softmax
|
| 2672 |
output_format: SAME
|
| 2673 |
transformer.h.42.ln_1:
|
| 2674 |
-
approximation_function:
|
| 2675 |
bias_format: SAME
|
| 2676 |
input_format: SAME
|
| 2677 |
instance: LayerNorm
|
| 2678 |
output_format: SAME
|
| 2679 |
weight_format: SAME
|
| 2680 |
transformer.h.42.ln_2:
|
| 2681 |
-
approximation_function:
|
| 2682 |
bias_format: SAME
|
| 2683 |
input_format: SAME
|
| 2684 |
instance: LayerNorm
|
| 2685 |
output_format: SAME
|
| 2686 |
weight_format: SAME
|
| 2687 |
transformer.h.42.mlp.act:
|
| 2688 |
-
approximation_function:
|
| 2689 |
input_format: SAME
|
| 2690 |
instance: GELU
|
| 2691 |
output_format: SAME
|
|
@@ -2737,26 +2737,26 @@ model:
|
|
| 2737 |
instance: Dropout
|
| 2738 |
output_format: SAME
|
| 2739 |
transformer.h.43.attn.softmax:
|
| 2740 |
-
approximation_function:
|
| 2741 |
input_format: SAME
|
| 2742 |
instance: Softmax
|
| 2743 |
output_format: SAME
|
| 2744 |
transformer.h.43.ln_1:
|
| 2745 |
-
approximation_function:
|
| 2746 |
bias_format: SAME
|
| 2747 |
input_format: SAME
|
| 2748 |
instance: LayerNorm
|
| 2749 |
output_format: SAME
|
| 2750 |
weight_format: SAME
|
| 2751 |
transformer.h.43.ln_2:
|
| 2752 |
-
approximation_function:
|
| 2753 |
bias_format: SAME
|
| 2754 |
input_format: SAME
|
| 2755 |
instance: LayerNorm
|
| 2756 |
output_format: SAME
|
| 2757 |
weight_format: SAME
|
| 2758 |
transformer.h.43.mlp.act:
|
| 2759 |
-
approximation_function:
|
| 2760 |
input_format: SAME
|
| 2761 |
instance: GELU
|
| 2762 |
output_format: SAME
|
|
@@ -2808,26 +2808,26 @@ model:
|
|
| 2808 |
instance: Dropout
|
| 2809 |
output_format: SAME
|
| 2810 |
transformer.h.44.attn.softmax:
|
| 2811 |
-
approximation_function:
|
| 2812 |
input_format: SAME
|
| 2813 |
instance: Softmax
|
| 2814 |
output_format: SAME
|
| 2815 |
transformer.h.44.ln_1:
|
| 2816 |
-
approximation_function:
|
| 2817 |
bias_format: SAME
|
| 2818 |
input_format: SAME
|
| 2819 |
instance: LayerNorm
|
| 2820 |
output_format: SAME
|
| 2821 |
weight_format: SAME
|
| 2822 |
transformer.h.44.ln_2:
|
| 2823 |
-
approximation_function:
|
| 2824 |
bias_format: SAME
|
| 2825 |
input_format: SAME
|
| 2826 |
instance: LayerNorm
|
| 2827 |
output_format: SAME
|
| 2828 |
weight_format: SAME
|
| 2829 |
transformer.h.44.mlp.act:
|
| 2830 |
-
approximation_function:
|
| 2831 |
input_format: SAME
|
| 2832 |
instance: GELU
|
| 2833 |
output_format: SAME
|
|
@@ -2879,26 +2879,26 @@ model:
|
|
| 2879 |
instance: Dropout
|
| 2880 |
output_format: SAME
|
| 2881 |
transformer.h.45.attn.softmax:
|
| 2882 |
-
approximation_function:
|
| 2883 |
input_format: SAME
|
| 2884 |
instance: Softmax
|
| 2885 |
output_format: SAME
|
| 2886 |
transformer.h.45.ln_1:
|
| 2887 |
-
approximation_function:
|
| 2888 |
bias_format: SAME
|
| 2889 |
input_format: SAME
|
| 2890 |
instance: LayerNorm
|
| 2891 |
output_format: SAME
|
| 2892 |
weight_format: SAME
|
| 2893 |
transformer.h.45.ln_2:
|
| 2894 |
-
approximation_function:
|
| 2895 |
bias_format: SAME
|
| 2896 |
input_format: SAME
|
| 2897 |
instance: LayerNorm
|
| 2898 |
output_format: SAME
|
| 2899 |
weight_format: SAME
|
| 2900 |
transformer.h.45.mlp.act:
|
| 2901 |
-
approximation_function:
|
| 2902 |
input_format: SAME
|
| 2903 |
instance: GELU
|
| 2904 |
output_format: SAME
|
|
@@ -2950,26 +2950,26 @@ model:
|
|
| 2950 |
instance: Dropout
|
| 2951 |
output_format: SAME
|
| 2952 |
transformer.h.46.attn.softmax:
|
| 2953 |
-
approximation_function:
|
| 2954 |
input_format: SAME
|
| 2955 |
instance: Softmax
|
| 2956 |
output_format: SAME
|
| 2957 |
transformer.h.46.ln_1:
|
| 2958 |
-
approximation_function:
|
| 2959 |
bias_format: SAME
|
| 2960 |
input_format: SAME
|
| 2961 |
instance: LayerNorm
|
| 2962 |
output_format: SAME
|
| 2963 |
weight_format: SAME
|
| 2964 |
transformer.h.46.ln_2:
|
| 2965 |
-
approximation_function:
|
| 2966 |
bias_format: SAME
|
| 2967 |
input_format: SAME
|
| 2968 |
instance: LayerNorm
|
| 2969 |
output_format: SAME
|
| 2970 |
weight_format: SAME
|
| 2971 |
transformer.h.46.mlp.act:
|
| 2972 |
-
approximation_function:
|
| 2973 |
input_format: SAME
|
| 2974 |
instance: GELU
|
| 2975 |
output_format: SAME
|
|
@@ -3021,26 +3021,26 @@ model:
|
|
| 3021 |
instance: Dropout
|
| 3022 |
output_format: SAME
|
| 3023 |
transformer.h.47.attn.softmax:
|
| 3024 |
-
approximation_function:
|
| 3025 |
input_format: SAME
|
| 3026 |
instance: Softmax
|
| 3027 |
output_format: SAME
|
| 3028 |
transformer.h.47.ln_1:
|
| 3029 |
-
approximation_function:
|
| 3030 |
bias_format: SAME
|
| 3031 |
input_format: SAME
|
| 3032 |
instance: LayerNorm
|
| 3033 |
output_format: SAME
|
| 3034 |
weight_format: SAME
|
| 3035 |
transformer.h.47.ln_2:
|
| 3036 |
-
approximation_function:
|
| 3037 |
bias_format: SAME
|
| 3038 |
input_format: SAME
|
| 3039 |
instance: LayerNorm
|
| 3040 |
output_format: SAME
|
| 3041 |
weight_format: SAME
|
| 3042 |
transformer.h.47.mlp.act:
|
| 3043 |
-
approximation_function:
|
| 3044 |
input_format: SAME
|
| 3045 |
instance: GELU
|
| 3046 |
output_format: SAME
|
|
@@ -3092,26 +3092,26 @@ model:
|
|
| 3092 |
instance: Dropout
|
| 3093 |
output_format: SAME
|
| 3094 |
transformer.h.5.attn.softmax:
|
| 3095 |
-
approximation_function:
|
| 3096 |
input_format: SAME
|
| 3097 |
instance: Softmax
|
| 3098 |
output_format: SAME
|
| 3099 |
transformer.h.5.ln_1:
|
| 3100 |
-
approximation_function:
|
| 3101 |
bias_format: SAME
|
| 3102 |
input_format: SAME
|
| 3103 |
instance: LayerNorm
|
| 3104 |
output_format: SAME
|
| 3105 |
weight_format: SAME
|
| 3106 |
transformer.h.5.ln_2:
|
| 3107 |
-
approximation_function:
|
| 3108 |
bias_format: SAME
|
| 3109 |
input_format: SAME
|
| 3110 |
instance: LayerNorm
|
| 3111 |
output_format: SAME
|
| 3112 |
weight_format: SAME
|
| 3113 |
transformer.h.5.mlp.act:
|
| 3114 |
-
approximation_function:
|
| 3115 |
input_format: SAME
|
| 3116 |
instance: GELU
|
| 3117 |
output_format: SAME
|
|
@@ -3163,26 +3163,26 @@ model:
|
|
| 3163 |
instance: Dropout
|
| 3164 |
output_format: SAME
|
| 3165 |
transformer.h.6.attn.softmax:
|
| 3166 |
-
approximation_function:
|
| 3167 |
input_format: SAME
|
| 3168 |
instance: Softmax
|
| 3169 |
output_format: SAME
|
| 3170 |
transformer.h.6.ln_1:
|
| 3171 |
-
approximation_function:
|
| 3172 |
bias_format: SAME
|
| 3173 |
input_format: SAME
|
| 3174 |
instance: LayerNorm
|
| 3175 |
output_format: SAME
|
| 3176 |
weight_format: SAME
|
| 3177 |
transformer.h.6.ln_2:
|
| 3178 |
-
approximation_function:
|
| 3179 |
bias_format: SAME
|
| 3180 |
input_format: SAME
|
| 3181 |
instance: LayerNorm
|
| 3182 |
output_format: SAME
|
| 3183 |
weight_format: SAME
|
| 3184 |
transformer.h.6.mlp.act:
|
| 3185 |
-
approximation_function:
|
| 3186 |
input_format: SAME
|
| 3187 |
instance: GELU
|
| 3188 |
output_format: SAME
|
|
@@ -3234,26 +3234,26 @@ model:
|
|
| 3234 |
instance: Dropout
|
| 3235 |
output_format: SAME
|
| 3236 |
transformer.h.7.attn.softmax:
|
| 3237 |
-
approximation_function:
|
| 3238 |
input_format: SAME
|
| 3239 |
instance: Softmax
|
| 3240 |
output_format: SAME
|
| 3241 |
transformer.h.7.ln_1:
|
| 3242 |
-
approximation_function:
|
| 3243 |
bias_format: SAME
|
| 3244 |
input_format: SAME
|
| 3245 |
instance: LayerNorm
|
| 3246 |
output_format: SAME
|
| 3247 |
weight_format: SAME
|
| 3248 |
transformer.h.7.ln_2:
|
| 3249 |
-
approximation_function:
|
| 3250 |
bias_format: SAME
|
| 3251 |
input_format: SAME
|
| 3252 |
instance: LayerNorm
|
| 3253 |
output_format: SAME
|
| 3254 |
weight_format: SAME
|
| 3255 |
transformer.h.7.mlp.act:
|
| 3256 |
-
approximation_function:
|
| 3257 |
input_format: SAME
|
| 3258 |
instance: GELU
|
| 3259 |
output_format: SAME
|
|
@@ -3305,26 +3305,26 @@ model:
|
|
| 3305 |
instance: Dropout
|
| 3306 |
output_format: SAME
|
| 3307 |
transformer.h.8.attn.softmax:
|
| 3308 |
-
approximation_function:
|
| 3309 |
input_format: SAME
|
| 3310 |
instance: Softmax
|
| 3311 |
output_format: SAME
|
| 3312 |
transformer.h.8.ln_1:
|
| 3313 |
-
approximation_function:
|
| 3314 |
bias_format: SAME
|
| 3315 |
input_format: SAME
|
| 3316 |
instance: LayerNorm
|
| 3317 |
output_format: SAME
|
| 3318 |
weight_format: SAME
|
| 3319 |
transformer.h.8.ln_2:
|
| 3320 |
-
approximation_function:
|
| 3321 |
bias_format: SAME
|
| 3322 |
input_format: SAME
|
| 3323 |
instance: LayerNorm
|
| 3324 |
output_format: SAME
|
| 3325 |
weight_format: SAME
|
| 3326 |
transformer.h.8.mlp.act:
|
| 3327 |
-
approximation_function:
|
| 3328 |
input_format: SAME
|
| 3329 |
instance: GELU
|
| 3330 |
output_format: SAME
|
|
@@ -3376,26 +3376,26 @@ model:
|
|
| 3376 |
instance: Dropout
|
| 3377 |
output_format: SAME
|
| 3378 |
transformer.h.9.attn.softmax:
|
| 3379 |
-
approximation_function:
|
| 3380 |
input_format: SAME
|
| 3381 |
instance: Softmax
|
| 3382 |
output_format: SAME
|
| 3383 |
transformer.h.9.ln_1:
|
| 3384 |
-
approximation_function:
|
| 3385 |
bias_format: SAME
|
| 3386 |
input_format: SAME
|
| 3387 |
instance: LayerNorm
|
| 3388 |
output_format: SAME
|
| 3389 |
weight_format: SAME
|
| 3390 |
transformer.h.9.ln_2:
|
| 3391 |
-
approximation_function:
|
| 3392 |
bias_format: SAME
|
| 3393 |
input_format: SAME
|
| 3394 |
instance: LayerNorm
|
| 3395 |
output_format: SAME
|
| 3396 |
weight_format: SAME
|
| 3397 |
transformer.h.9.mlp.act:
|
| 3398 |
-
approximation_function:
|
| 3399 |
input_format: SAME
|
| 3400 |
instance: GELU
|
| 3401 |
output_format: SAME
|
|
@@ -3421,7 +3421,7 @@ model:
|
|
| 3421 |
instance: Dropout
|
| 3422 |
output_format: SAME
|
| 3423 |
transformer.ln_f:
|
| 3424 |
-
approximation_function:
|
| 3425 |
bias_format: SAME
|
| 3426 |
input_format: SAME
|
| 3427 |
instance: LayerNorm
|
|
|
|
| 39 |
instance: Dropout
|
| 40 |
output_format: SAME
|
| 41 |
transformer.h.0.attn.softmax:
|
| 42 |
+
approximation_function: NONE
|
| 43 |
input_format: SAME
|
| 44 |
instance: Softmax
|
| 45 |
output_format: SAME
|
| 46 |
transformer.h.0.ln_1:
|
| 47 |
+
approximation_function: NONE
|
| 48 |
bias_format: SAME
|
| 49 |
input_format: SAME
|
| 50 |
instance: LayerNorm
|
| 51 |
output_format: SAME
|
| 52 |
weight_format: SAME
|
| 53 |
transformer.h.0.ln_2:
|
| 54 |
+
approximation_function: NONE
|
| 55 |
bias_format: SAME
|
| 56 |
input_format: SAME
|
| 57 |
instance: LayerNorm
|
| 58 |
output_format: SAME
|
| 59 |
weight_format: SAME
|
| 60 |
transformer.h.0.mlp.act:
|
| 61 |
+
approximation_function: NONE
|
| 62 |
input_format: SAME
|
| 63 |
instance: GELU
|
| 64 |
output_format: SAME
|
|
|
|
| 110 |
instance: Dropout
|
| 111 |
output_format: SAME
|
| 112 |
transformer.h.1.attn.softmax:
|
| 113 |
+
approximation_function: NONE
|
| 114 |
input_format: SAME
|
| 115 |
instance: Softmax
|
| 116 |
output_format: SAME
|
| 117 |
transformer.h.1.ln_1:
|
| 118 |
+
approximation_function: NONE
|
| 119 |
bias_format: SAME
|
| 120 |
input_format: SAME
|
| 121 |
instance: LayerNorm
|
| 122 |
output_format: SAME
|
| 123 |
weight_format: SAME
|
| 124 |
transformer.h.1.ln_2:
|
| 125 |
+
approximation_function: NONE
|
| 126 |
bias_format: SAME
|
| 127 |
input_format: SAME
|
| 128 |
instance: LayerNorm
|
| 129 |
output_format: SAME
|
| 130 |
weight_format: SAME
|
| 131 |
transformer.h.1.mlp.act:
|
| 132 |
+
approximation_function: NONE
|
| 133 |
input_format: SAME
|
| 134 |
instance: GELU
|
| 135 |
output_format: SAME
|
|
|
|
| 181 |
instance: Dropout
|
| 182 |
output_format: SAME
|
| 183 |
transformer.h.10.attn.softmax:
|
| 184 |
+
approximation_function: NONE
|
| 185 |
input_format: SAME
|
| 186 |
instance: Softmax
|
| 187 |
output_format: SAME
|
| 188 |
transformer.h.10.ln_1:
|
| 189 |
+
approximation_function: NONE
|
| 190 |
bias_format: SAME
|
| 191 |
input_format: SAME
|
| 192 |
instance: LayerNorm
|
| 193 |
output_format: SAME
|
| 194 |
weight_format: SAME
|
| 195 |
transformer.h.10.ln_2:
|
| 196 |
+
approximation_function: NONE
|
| 197 |
bias_format: SAME
|
| 198 |
input_format: SAME
|
| 199 |
instance: LayerNorm
|
| 200 |
output_format: SAME
|
| 201 |
weight_format: SAME
|
| 202 |
transformer.h.10.mlp.act:
|
| 203 |
+
approximation_function: NONE
|
| 204 |
input_format: SAME
|
| 205 |
instance: GELU
|
| 206 |
output_format: SAME
|
|
|
|
| 252 |
instance: Dropout
|
| 253 |
output_format: SAME
|
| 254 |
transformer.h.11.attn.softmax:
|
| 255 |
+
approximation_function: NONE
|
| 256 |
input_format: SAME
|
| 257 |
instance: Softmax
|
| 258 |
output_format: SAME
|
| 259 |
transformer.h.11.ln_1:
|
| 260 |
+
approximation_function: NONE
|
| 261 |
bias_format: SAME
|
| 262 |
input_format: SAME
|
| 263 |
instance: LayerNorm
|
| 264 |
output_format: SAME
|
| 265 |
weight_format: SAME
|
| 266 |
transformer.h.11.ln_2:
|
| 267 |
+
approximation_function: NONE
|
| 268 |
bias_format: SAME
|
| 269 |
input_format: SAME
|
| 270 |
instance: LayerNorm
|
| 271 |
output_format: SAME
|
| 272 |
weight_format: SAME
|
| 273 |
transformer.h.11.mlp.act:
|
| 274 |
+
approximation_function: NONE
|
| 275 |
input_format: SAME
|
| 276 |
instance: GELU
|
| 277 |
output_format: SAME
|
|
|
|
| 323 |
instance: Dropout
|
| 324 |
output_format: SAME
|
| 325 |
transformer.h.12.attn.softmax:
|
| 326 |
+
approximation_function: NONE
|
| 327 |
input_format: SAME
|
| 328 |
instance: Softmax
|
| 329 |
output_format: SAME
|
| 330 |
transformer.h.12.ln_1:
|
| 331 |
+
approximation_function: NONE
|
| 332 |
bias_format: SAME
|
| 333 |
input_format: SAME
|
| 334 |
instance: LayerNorm
|
| 335 |
output_format: SAME
|
| 336 |
weight_format: SAME
|
| 337 |
transformer.h.12.ln_2:
|
| 338 |
+
approximation_function: NONE
|
| 339 |
bias_format: SAME
|
| 340 |
input_format: SAME
|
| 341 |
instance: LayerNorm
|
| 342 |
output_format: SAME
|
| 343 |
weight_format: SAME
|
| 344 |
transformer.h.12.mlp.act:
|
| 345 |
+
approximation_function: NONE
|
| 346 |
input_format: SAME
|
| 347 |
instance: GELU
|
| 348 |
output_format: SAME
|
|
|
|
| 394 |
instance: Dropout
|
| 395 |
output_format: SAME
|
| 396 |
transformer.h.13.attn.softmax:
|
| 397 |
+
approximation_function: NONE
|
| 398 |
input_format: SAME
|
| 399 |
instance: Softmax
|
| 400 |
output_format: SAME
|
| 401 |
transformer.h.13.ln_1:
|
| 402 |
+
approximation_function: NONE
|
| 403 |
bias_format: SAME
|
| 404 |
input_format: SAME
|
| 405 |
instance: LayerNorm
|
| 406 |
output_format: SAME
|
| 407 |
weight_format: SAME
|
| 408 |
transformer.h.13.ln_2:
|
| 409 |
+
approximation_function: NONE
|
| 410 |
bias_format: SAME
|
| 411 |
input_format: SAME
|
| 412 |
instance: LayerNorm
|
| 413 |
output_format: SAME
|
| 414 |
weight_format: SAME
|
| 415 |
transformer.h.13.mlp.act:
|
| 416 |
+
approximation_function: NONE
|
| 417 |
input_format: SAME
|
| 418 |
instance: GELU
|
| 419 |
output_format: SAME
|
|
|
|
| 465 |
instance: Dropout
|
| 466 |
output_format: SAME
|
| 467 |
transformer.h.14.attn.softmax:
|
| 468 |
+
approximation_function: NONE
|
| 469 |
input_format: SAME
|
| 470 |
instance: Softmax
|
| 471 |
output_format: SAME
|
| 472 |
transformer.h.14.ln_1:
|
| 473 |
+
approximation_function: NONE
|
| 474 |
bias_format: SAME
|
| 475 |
input_format: SAME
|
| 476 |
instance: LayerNorm
|
| 477 |
output_format: SAME
|
| 478 |
weight_format: SAME
|
| 479 |
transformer.h.14.ln_2:
|
| 480 |
+
approximation_function: NONE
|
| 481 |
bias_format: SAME
|
| 482 |
input_format: SAME
|
| 483 |
instance: LayerNorm
|
| 484 |
output_format: SAME
|
| 485 |
weight_format: SAME
|
| 486 |
transformer.h.14.mlp.act:
|
| 487 |
+
approximation_function: NONE
|
| 488 |
input_format: SAME
|
| 489 |
instance: GELU
|
| 490 |
output_format: SAME
|
|
|
|
| 536 |
instance: Dropout
|
| 537 |
output_format: SAME
|
| 538 |
transformer.h.15.attn.softmax:
|
| 539 |
+
approximation_function: NONE
|
| 540 |
input_format: SAME
|
| 541 |
instance: Softmax
|
| 542 |
output_format: SAME
|
| 543 |
transformer.h.15.ln_1:
|
| 544 |
+
approximation_function: NONE
|
| 545 |
bias_format: SAME
|
| 546 |
input_format: SAME
|
| 547 |
instance: LayerNorm
|
| 548 |
output_format: SAME
|
| 549 |
weight_format: SAME
|
| 550 |
transformer.h.15.ln_2:
|
| 551 |
+
approximation_function: NONE
|
| 552 |
bias_format: SAME
|
| 553 |
input_format: SAME
|
| 554 |
instance: LayerNorm
|
| 555 |
output_format: SAME
|
| 556 |
weight_format: SAME
|
| 557 |
transformer.h.15.mlp.act:
|
| 558 |
+
approximation_function: NONE
|
| 559 |
input_format: SAME
|
| 560 |
instance: GELU
|
| 561 |
output_format: SAME
|
|
|
|
| 607 |
instance: Dropout
|
| 608 |
output_format: SAME
|
| 609 |
transformer.h.16.attn.softmax:
|
| 610 |
+
approximation_function: NONE
|
| 611 |
input_format: SAME
|
| 612 |
instance: Softmax
|
| 613 |
output_format: SAME
|
| 614 |
transformer.h.16.ln_1:
|
| 615 |
+
approximation_function: NONE
|
| 616 |
bias_format: SAME
|
| 617 |
input_format: SAME
|
| 618 |
instance: LayerNorm
|
| 619 |
output_format: SAME
|
| 620 |
weight_format: SAME
|
| 621 |
transformer.h.16.ln_2:
|
| 622 |
+
approximation_function: NONE
|
| 623 |
bias_format: SAME
|
| 624 |
input_format: SAME
|
| 625 |
instance: LayerNorm
|
| 626 |
output_format: SAME
|
| 627 |
weight_format: SAME
|
| 628 |
transformer.h.16.mlp.act:
|
| 629 |
+
approximation_function: NONE
|
| 630 |
input_format: SAME
|
| 631 |
instance: GELU
|
| 632 |
output_format: SAME
|
|
|
|
| 678 |
instance: Dropout
|
| 679 |
output_format: SAME
|
| 680 |
transformer.h.17.attn.softmax:
|
| 681 |
+
approximation_function: NONE
|
| 682 |
input_format: SAME
|
| 683 |
instance: Softmax
|
| 684 |
output_format: SAME
|
| 685 |
transformer.h.17.ln_1:
|
| 686 |
+
approximation_function: NONE
|
| 687 |
bias_format: SAME
|
| 688 |
input_format: SAME
|
| 689 |
instance: LayerNorm
|
| 690 |
output_format: SAME
|
| 691 |
weight_format: SAME
|
| 692 |
transformer.h.17.ln_2:
|
| 693 |
+
approximation_function: NONE
|
| 694 |
bias_format: SAME
|
| 695 |
input_format: SAME
|
| 696 |
instance: LayerNorm
|
| 697 |
output_format: SAME
|
| 698 |
weight_format: SAME
|
| 699 |
transformer.h.17.mlp.act:
|
| 700 |
+
approximation_function: NONE
|
| 701 |
input_format: SAME
|
| 702 |
instance: GELU
|
| 703 |
output_format: SAME
|
|
|
|
| 749 |
instance: Dropout
|
| 750 |
output_format: SAME
|
| 751 |
transformer.h.18.attn.softmax:
|
| 752 |
+
approximation_function: NONE
|
| 753 |
input_format: SAME
|
| 754 |
instance: Softmax
|
| 755 |
output_format: SAME
|
| 756 |
transformer.h.18.ln_1:
|
| 757 |
+
approximation_function: NONE
|
| 758 |
bias_format: SAME
|
| 759 |
input_format: SAME
|
| 760 |
instance: LayerNorm
|
| 761 |
output_format: SAME
|
| 762 |
weight_format: SAME
|
| 763 |
transformer.h.18.ln_2:
|
| 764 |
+
approximation_function: NONE
|
| 765 |
bias_format: SAME
|
| 766 |
input_format: SAME
|
| 767 |
instance: LayerNorm
|
| 768 |
output_format: SAME
|
| 769 |
weight_format: SAME
|
| 770 |
transformer.h.18.mlp.act:
|
| 771 |
+
approximation_function: NONE
|
| 772 |
input_format: SAME
|
| 773 |
instance: GELU
|
| 774 |
output_format: SAME
|
|
|
|
| 820 |
instance: Dropout
|
| 821 |
output_format: SAME
|
| 822 |
transformer.h.19.attn.softmax:
|
| 823 |
+
approximation_function: NONE
|
| 824 |
input_format: SAME
|
| 825 |
instance: Softmax
|
| 826 |
output_format: SAME
|
| 827 |
transformer.h.19.ln_1:
|
| 828 |
+
approximation_function: NONE
|
| 829 |
bias_format: SAME
|
| 830 |
input_format: SAME
|
| 831 |
instance: LayerNorm
|
| 832 |
output_format: SAME
|
| 833 |
weight_format: SAME
|
| 834 |
transformer.h.19.ln_2:
|
| 835 |
+
approximation_function: NONE
|
| 836 |
bias_format: SAME
|
| 837 |
input_format: SAME
|
| 838 |
instance: LayerNorm
|
| 839 |
output_format: SAME
|
| 840 |
weight_format: SAME
|
| 841 |
transformer.h.19.mlp.act:
|
| 842 |
+
approximation_function: NONE
|
| 843 |
input_format: SAME
|
| 844 |
instance: GELU
|
| 845 |
output_format: SAME
|
|
|
|
| 891 |
instance: Dropout
|
| 892 |
output_format: SAME
|
| 893 |
transformer.h.2.attn.softmax:
|
| 894 |
+
approximation_function: NONE
|
| 895 |
input_format: SAME
|
| 896 |
instance: Softmax
|
| 897 |
output_format: SAME
|
| 898 |
transformer.h.2.ln_1:
|
| 899 |
+
approximation_function: NONE
|
| 900 |
bias_format: SAME
|
| 901 |
input_format: SAME
|
| 902 |
instance: LayerNorm
|
| 903 |
output_format: SAME
|
| 904 |
weight_format: SAME
|
| 905 |
transformer.h.2.ln_2:
|
| 906 |
+
approximation_function: NONE
|
| 907 |
bias_format: SAME
|
| 908 |
input_format: SAME
|
| 909 |
instance: LayerNorm
|
| 910 |
output_format: SAME
|
| 911 |
weight_format: SAME
|
| 912 |
transformer.h.2.mlp.act:
|
| 913 |
+
approximation_function: NONE
|
| 914 |
input_format: SAME
|
| 915 |
instance: GELU
|
| 916 |
output_format: SAME
|
|
|
|
| 962 |
instance: Dropout
|
| 963 |
output_format: SAME
|
| 964 |
transformer.h.20.attn.softmax:
|
| 965 |
+
approximation_function: NONE
|
| 966 |
input_format: SAME
|
| 967 |
instance: Softmax
|
| 968 |
output_format: SAME
|
| 969 |
transformer.h.20.ln_1:
|
| 970 |
+
approximation_function: NONE
|
| 971 |
bias_format: SAME
|
| 972 |
input_format: SAME
|
| 973 |
instance: LayerNorm
|
| 974 |
output_format: SAME
|
| 975 |
weight_format: SAME
|
| 976 |
transformer.h.20.ln_2:
|
| 977 |
+
approximation_function: NONE
|
| 978 |
bias_format: SAME
|
| 979 |
input_format: SAME
|
| 980 |
instance: LayerNorm
|
| 981 |
output_format: SAME
|
| 982 |
weight_format: SAME
|
| 983 |
transformer.h.20.mlp.act:
|
| 984 |
+
approximation_function: NONE
|
| 985 |
input_format: SAME
|
| 986 |
instance: GELU
|
| 987 |
output_format: SAME
|
|
|
|
| 1033 |
instance: Dropout
|
| 1034 |
output_format: SAME
|
| 1035 |
transformer.h.21.attn.softmax:
|
| 1036 |
+
approximation_function: NONE
|
| 1037 |
input_format: SAME
|
| 1038 |
instance: Softmax
|
| 1039 |
output_format: SAME
|
| 1040 |
transformer.h.21.ln_1:
|
| 1041 |
+
approximation_function: NONE
|
| 1042 |
bias_format: SAME
|
| 1043 |
input_format: SAME
|
| 1044 |
instance: LayerNorm
|
| 1045 |
output_format: SAME
|
| 1046 |
weight_format: SAME
|
| 1047 |
transformer.h.21.ln_2:
|
| 1048 |
+
approximation_function: NONE
|
| 1049 |
bias_format: SAME
|
| 1050 |
input_format: SAME
|
| 1051 |
instance: LayerNorm
|
| 1052 |
output_format: SAME
|
| 1053 |
weight_format: SAME
|
| 1054 |
transformer.h.21.mlp.act:
|
| 1055 |
+
approximation_function: NONE
|
| 1056 |
input_format: SAME
|
| 1057 |
instance: GELU
|
| 1058 |
output_format: SAME
|
|
|
|
| 1104 |
instance: Dropout
|
| 1105 |
output_format: SAME
|
| 1106 |
transformer.h.22.attn.softmax:
|
| 1107 |
+
approximation_function: NONE
|
| 1108 |
input_format: SAME
|
| 1109 |
instance: Softmax
|
| 1110 |
output_format: SAME
|
| 1111 |
transformer.h.22.ln_1:
|
| 1112 |
+
approximation_function: NONE
|
| 1113 |
bias_format: SAME
|
| 1114 |
input_format: SAME
|
| 1115 |
instance: LayerNorm
|
| 1116 |
output_format: SAME
|
| 1117 |
weight_format: SAME
|
| 1118 |
transformer.h.22.ln_2:
|
| 1119 |
+
approximation_function: NONE
|
| 1120 |
bias_format: SAME
|
| 1121 |
input_format: SAME
|
| 1122 |
instance: LayerNorm
|
| 1123 |
output_format: SAME
|
| 1124 |
weight_format: SAME
|
| 1125 |
transformer.h.22.mlp.act:
|
| 1126 |
+
approximation_function: NONE
|
| 1127 |
input_format: SAME
|
| 1128 |
instance: GELU
|
| 1129 |
output_format: SAME
|
|
|
|
| 1175 |
instance: Dropout
|
| 1176 |
output_format: SAME
|
| 1177 |
transformer.h.23.attn.softmax:
|
| 1178 |
+
approximation_function: NONE
|
| 1179 |
input_format: SAME
|
| 1180 |
instance: Softmax
|
| 1181 |
output_format: SAME
|
| 1182 |
transformer.h.23.ln_1:
|
| 1183 |
+
approximation_function: NONE
|
| 1184 |
bias_format: SAME
|
| 1185 |
input_format: SAME
|
| 1186 |
instance: LayerNorm
|
| 1187 |
output_format: SAME
|
| 1188 |
weight_format: SAME
|
| 1189 |
transformer.h.23.ln_2:
|
| 1190 |
+
approximation_function: NONE
|
| 1191 |
bias_format: SAME
|
| 1192 |
input_format: SAME
|
| 1193 |
instance: LayerNorm
|
| 1194 |
output_format: SAME
|
| 1195 |
weight_format: SAME
|
| 1196 |
transformer.h.23.mlp.act:
|
| 1197 |
+
approximation_function: NONE
|
| 1198 |
input_format: SAME
|
| 1199 |
instance: GELU
|
| 1200 |
output_format: SAME
|
|
|
|
| 1246 |
instance: Dropout
|
| 1247 |
output_format: SAME
|
| 1248 |
transformer.h.24.attn.softmax:
|
| 1249 |
+
approximation_function: NONE
|
| 1250 |
input_format: SAME
|
| 1251 |
instance: Softmax
|
| 1252 |
output_format: SAME
|
| 1253 |
transformer.h.24.ln_1:
|
| 1254 |
+
approximation_function: NONE
|
| 1255 |
bias_format: SAME
|
| 1256 |
input_format: SAME
|
| 1257 |
instance: LayerNorm
|
| 1258 |
output_format: SAME
|
| 1259 |
weight_format: SAME
|
| 1260 |
transformer.h.24.ln_2:
|
| 1261 |
+
approximation_function: NONE
|
| 1262 |
bias_format: SAME
|
| 1263 |
input_format: SAME
|
| 1264 |
instance: LayerNorm
|
| 1265 |
output_format: SAME
|
| 1266 |
weight_format: SAME
|
| 1267 |
transformer.h.24.mlp.act:
|
| 1268 |
+
approximation_function: NONE
|
| 1269 |
input_format: SAME
|
| 1270 |
instance: GELU
|
| 1271 |
output_format: SAME
|
|
|
|
| 1317 |
instance: Dropout
|
| 1318 |
output_format: SAME
|
| 1319 |
transformer.h.25.attn.softmax:
|
| 1320 |
+
approximation_function: NONE
|
| 1321 |
input_format: SAME
|
| 1322 |
instance: Softmax
|
| 1323 |
output_format: SAME
|
| 1324 |
transformer.h.25.ln_1:
|
| 1325 |
+
approximation_function: NONE
|
| 1326 |
bias_format: SAME
|
| 1327 |
input_format: SAME
|
| 1328 |
instance: LayerNorm
|
| 1329 |
output_format: SAME
|
| 1330 |
weight_format: SAME
|
| 1331 |
transformer.h.25.ln_2:
|
| 1332 |
+
approximation_function: NONE
|
| 1333 |
bias_format: SAME
|
| 1334 |
input_format: SAME
|
| 1335 |
instance: LayerNorm
|
| 1336 |
output_format: SAME
|
| 1337 |
weight_format: SAME
|
| 1338 |
transformer.h.25.mlp.act:
|
| 1339 |
+
approximation_function: NONE
|
| 1340 |
input_format: SAME
|
| 1341 |
instance: GELU
|
| 1342 |
output_format: SAME
|
|
|
|
| 1388 |
instance: Dropout
|
| 1389 |
output_format: SAME
|
| 1390 |
transformer.h.26.attn.softmax:
|
| 1391 |
+
approximation_function: NONE
|
| 1392 |
input_format: SAME
|
| 1393 |
instance: Softmax
|
| 1394 |
output_format: SAME
|
| 1395 |
transformer.h.26.ln_1:
|
| 1396 |
+
approximation_function: NONE
|
| 1397 |
bias_format: SAME
|
| 1398 |
input_format: SAME
|
| 1399 |
instance: LayerNorm
|
| 1400 |
output_format: SAME
|
| 1401 |
weight_format: SAME
|
| 1402 |
transformer.h.26.ln_2:
|
| 1403 |
+
approximation_function: NONE
|
| 1404 |
bias_format: SAME
|
| 1405 |
input_format: SAME
|
| 1406 |
instance: LayerNorm
|
| 1407 |
output_format: SAME
|
| 1408 |
weight_format: SAME
|
| 1409 |
transformer.h.26.mlp.act:
|
| 1410 |
+
approximation_function: NONE
|
| 1411 |
input_format: SAME
|
| 1412 |
instance: GELU
|
| 1413 |
output_format: SAME
|
|
|
|
| 1459 |
instance: Dropout
|
| 1460 |
output_format: SAME
|
| 1461 |
transformer.h.27.attn.softmax:
|
| 1462 |
+
approximation_function: NONE
|
| 1463 |
input_format: SAME
|
| 1464 |
instance: Softmax
|
| 1465 |
output_format: SAME
|
| 1466 |
transformer.h.27.ln_1:
|
| 1467 |
+
approximation_function: NONE
|
| 1468 |
bias_format: SAME
|
| 1469 |
input_format: SAME
|
| 1470 |
instance: LayerNorm
|
| 1471 |
output_format: SAME
|
| 1472 |
weight_format: SAME
|
| 1473 |
transformer.h.27.ln_2:
|
| 1474 |
+
approximation_function: NONE
|
| 1475 |
bias_format: SAME
|
| 1476 |
input_format: SAME
|
| 1477 |
instance: LayerNorm
|
| 1478 |
output_format: SAME
|
| 1479 |
weight_format: SAME
|
| 1480 |
transformer.h.27.mlp.act:
|
| 1481 |
+
approximation_function: NONE
|
| 1482 |
input_format: SAME
|
| 1483 |
instance: GELU
|
| 1484 |
output_format: SAME
|
|
|
|
| 1530 |
instance: Dropout
|
| 1531 |
output_format: SAME
|
| 1532 |
transformer.h.28.attn.softmax:
|
| 1533 |
+
approximation_function: NONE
|
| 1534 |
input_format: SAME
|
| 1535 |
instance: Softmax
|
| 1536 |
output_format: SAME
|
| 1537 |
transformer.h.28.ln_1:
|
| 1538 |
+
approximation_function: NONE
|
| 1539 |
bias_format: SAME
|
| 1540 |
input_format: SAME
|
| 1541 |
instance: LayerNorm
|
| 1542 |
output_format: SAME
|
| 1543 |
weight_format: SAME
|
| 1544 |
transformer.h.28.ln_2:
|
| 1545 |
+
approximation_function: NONE
|
| 1546 |
bias_format: SAME
|
| 1547 |
input_format: SAME
|
| 1548 |
instance: LayerNorm
|
| 1549 |
output_format: SAME
|
| 1550 |
weight_format: SAME
|
| 1551 |
transformer.h.28.mlp.act:
|
| 1552 |
+
approximation_function: NONE
|
| 1553 |
input_format: SAME
|
| 1554 |
instance: GELU
|
| 1555 |
output_format: SAME
|
|
|
|
| 1601 |
instance: Dropout
|
| 1602 |
output_format: SAME
|
| 1603 |
transformer.h.29.attn.softmax:
|
| 1604 |
+
approximation_function: NONE
|
| 1605 |
input_format: SAME
|
| 1606 |
instance: Softmax
|
| 1607 |
output_format: SAME
|
| 1608 |
transformer.h.29.ln_1:
|
| 1609 |
+
approximation_function: NONE
|
| 1610 |
bias_format: SAME
|
| 1611 |
input_format: SAME
|
| 1612 |
instance: LayerNorm
|
| 1613 |
output_format: SAME
|
| 1614 |
weight_format: SAME
|
| 1615 |
transformer.h.29.ln_2:
|
| 1616 |
+
approximation_function: NONE
|
| 1617 |
bias_format: SAME
|
| 1618 |
input_format: SAME
|
| 1619 |
instance: LayerNorm
|
| 1620 |
output_format: SAME
|
| 1621 |
weight_format: SAME
|
| 1622 |
transformer.h.29.mlp.act:
|
| 1623 |
+
approximation_function: NONE
|
| 1624 |
input_format: SAME
|
| 1625 |
instance: GELU
|
| 1626 |
output_format: SAME
|
|
|
|
| 1672 |
instance: Dropout
|
| 1673 |
output_format: SAME
|
| 1674 |
transformer.h.3.attn.softmax:
|
| 1675 |
+
approximation_function: NONE
|
| 1676 |
input_format: SAME
|
| 1677 |
instance: Softmax
|
| 1678 |
output_format: SAME
|
| 1679 |
transformer.h.3.ln_1:
|
| 1680 |
+
approximation_function: NONE
|
| 1681 |
bias_format: SAME
|
| 1682 |
input_format: SAME
|
| 1683 |
instance: LayerNorm
|
| 1684 |
output_format: SAME
|
| 1685 |
weight_format: SAME
|
| 1686 |
transformer.h.3.ln_2:
|
| 1687 |
+
approximation_function: NONE
|
| 1688 |
bias_format: SAME
|
| 1689 |
input_format: SAME
|
| 1690 |
instance: LayerNorm
|
| 1691 |
output_format: SAME
|
| 1692 |
weight_format: SAME
|
| 1693 |
transformer.h.3.mlp.act:
|
| 1694 |
+
approximation_function: NONE
|
| 1695 |
input_format: SAME
|
| 1696 |
instance: GELU
|
| 1697 |
output_format: SAME
|
|
|
|
| 1743 |
instance: Dropout
|
| 1744 |
output_format: SAME
|
| 1745 |
transformer.h.30.attn.softmax:
|
| 1746 |
+
approximation_function: NONE
|
| 1747 |
input_format: SAME
|
| 1748 |
instance: Softmax
|
| 1749 |
output_format: SAME
|
| 1750 |
transformer.h.30.ln_1:
|
| 1751 |
+
approximation_function: NONE
|
| 1752 |
bias_format: SAME
|
| 1753 |
input_format: SAME
|
| 1754 |
instance: LayerNorm
|
| 1755 |
output_format: SAME
|
| 1756 |
weight_format: SAME
|
| 1757 |
transformer.h.30.ln_2:
|
| 1758 |
+
approximation_function: NONE
|
| 1759 |
bias_format: SAME
|
| 1760 |
input_format: SAME
|
| 1761 |
instance: LayerNorm
|
| 1762 |
output_format: SAME
|
| 1763 |
weight_format: SAME
|
| 1764 |
transformer.h.30.mlp.act:
|
| 1765 |
+
approximation_function: NONE
|
| 1766 |
input_format: SAME
|
| 1767 |
instance: GELU
|
| 1768 |
output_format: SAME
|
|
|
|
| 1814 |
instance: Dropout
|
| 1815 |
output_format: SAME
|
| 1816 |
transformer.h.31.attn.softmax:
|
| 1817 |
+
approximation_function: NONE
|
| 1818 |
input_format: SAME
|
| 1819 |
instance: Softmax
|
| 1820 |
output_format: SAME
|
| 1821 |
transformer.h.31.ln_1:
|
| 1822 |
+
approximation_function: NONE
|
| 1823 |
bias_format: SAME
|
| 1824 |
input_format: SAME
|
| 1825 |
instance: LayerNorm
|
| 1826 |
output_format: SAME
|
| 1827 |
weight_format: SAME
|
| 1828 |
transformer.h.31.ln_2:
|
| 1829 |
+
approximation_function: NONE
|
| 1830 |
bias_format: SAME
|
| 1831 |
input_format: SAME
|
| 1832 |
instance: LayerNorm
|
| 1833 |
output_format: SAME
|
| 1834 |
weight_format: SAME
|
| 1835 |
transformer.h.31.mlp.act:
|
| 1836 |
+
approximation_function: NONE
|
| 1837 |
input_format: SAME
|
| 1838 |
instance: GELU
|
| 1839 |
output_format: SAME
|
|
|
|
| 1885 |
instance: Dropout
|
| 1886 |
output_format: SAME
|
| 1887 |
transformer.h.32.attn.softmax:
|
| 1888 |
+
approximation_function: NONE
|
| 1889 |
input_format: SAME
|
| 1890 |
instance: Softmax
|
| 1891 |
output_format: SAME
|
| 1892 |
transformer.h.32.ln_1:
|
| 1893 |
+
approximation_function: NONE
|
| 1894 |
bias_format: SAME
|
| 1895 |
input_format: SAME
|
| 1896 |
instance: LayerNorm
|
| 1897 |
output_format: SAME
|
| 1898 |
weight_format: SAME
|
| 1899 |
transformer.h.32.ln_2:
|
| 1900 |
+
approximation_function: NONE
|
| 1901 |
bias_format: SAME
|
| 1902 |
input_format: SAME
|
| 1903 |
instance: LayerNorm
|
| 1904 |
output_format: SAME
|
| 1905 |
weight_format: SAME
|
| 1906 |
transformer.h.32.mlp.act:
|
| 1907 |
+
approximation_function: NONE
|
| 1908 |
input_format: SAME
|
| 1909 |
instance: GELU
|
| 1910 |
output_format: SAME
|
|
|
|
| 1956 |
instance: Dropout
|
| 1957 |
output_format: SAME
|
| 1958 |
transformer.h.33.attn.softmax:
|
| 1959 |
+
approximation_function: NONE
|
| 1960 |
input_format: SAME
|
| 1961 |
instance: Softmax
|
| 1962 |
output_format: SAME
|
| 1963 |
transformer.h.33.ln_1:
|
| 1964 |
+
approximation_function: NONE
|
| 1965 |
bias_format: SAME
|
| 1966 |
input_format: SAME
|
| 1967 |
instance: LayerNorm
|
| 1968 |
output_format: SAME
|
| 1969 |
weight_format: SAME
|
| 1970 |
transformer.h.33.ln_2:
|
| 1971 |
+
approximation_function: NONE
|
| 1972 |
bias_format: SAME
|
| 1973 |
input_format: SAME
|
| 1974 |
instance: LayerNorm
|
| 1975 |
output_format: SAME
|
| 1976 |
weight_format: SAME
|
| 1977 |
transformer.h.33.mlp.act:
|
| 1978 |
+
approximation_function: NONE
|
| 1979 |
input_format: SAME
|
| 1980 |
instance: GELU
|
| 1981 |
output_format: SAME
|
|
|
|
| 2027 |
instance: Dropout
|
| 2028 |
output_format: SAME
|
| 2029 |
transformer.h.34.attn.softmax:
|
| 2030 |
+
approximation_function: NONE
|
| 2031 |
input_format: SAME
|
| 2032 |
instance: Softmax
|
| 2033 |
output_format: SAME
|
| 2034 |
transformer.h.34.ln_1:
|
| 2035 |
+
approximation_function: NONE
|
| 2036 |
bias_format: SAME
|
| 2037 |
input_format: SAME
|
| 2038 |
instance: LayerNorm
|
| 2039 |
output_format: SAME
|
| 2040 |
weight_format: SAME
|
| 2041 |
transformer.h.34.ln_2:
|
| 2042 |
+
approximation_function: NONE
|
| 2043 |
bias_format: SAME
|
| 2044 |
input_format: SAME
|
| 2045 |
instance: LayerNorm
|
| 2046 |
output_format: SAME
|
| 2047 |
weight_format: SAME
|
| 2048 |
transformer.h.34.mlp.act:
|
| 2049 |
+
approximation_function: NONE
|
| 2050 |
input_format: SAME
|
| 2051 |
instance: GELU
|
| 2052 |
output_format: SAME
|
|
|
|
| 2098 |
instance: Dropout
|
| 2099 |
output_format: SAME
|
| 2100 |
transformer.h.35.attn.softmax:
|
| 2101 |
+
approximation_function: NONE
|
| 2102 |
input_format: SAME
|
| 2103 |
instance: Softmax
|
| 2104 |
output_format: SAME
|
| 2105 |
transformer.h.35.ln_1:
|
| 2106 |
+
approximation_function: NONE
|
| 2107 |
bias_format: SAME
|
| 2108 |
input_format: SAME
|
| 2109 |
instance: LayerNorm
|
| 2110 |
output_format: SAME
|
| 2111 |
weight_format: SAME
|
| 2112 |
transformer.h.35.ln_2:
|
| 2113 |
+
approximation_function: NONE
|
| 2114 |
bias_format: SAME
|
| 2115 |
input_format: SAME
|
| 2116 |
instance: LayerNorm
|
| 2117 |
output_format: SAME
|
| 2118 |
weight_format: SAME
|
| 2119 |
transformer.h.35.mlp.act:
|
| 2120 |
+
approximation_function: NONE
|
| 2121 |
input_format: SAME
|
| 2122 |
instance: GELU
|
| 2123 |
output_format: SAME
|
|
|
|
| 2169 |
instance: Dropout
|
| 2170 |
output_format: SAME
|
| 2171 |
transformer.h.36.attn.softmax:
|
| 2172 |
+
approximation_function: NONE
|
| 2173 |
input_format: SAME
|
| 2174 |
instance: Softmax
|
| 2175 |
output_format: SAME
|
| 2176 |
transformer.h.36.ln_1:
|
| 2177 |
+
approximation_function: NONE
|
| 2178 |
bias_format: SAME
|
| 2179 |
input_format: SAME
|
| 2180 |
instance: LayerNorm
|
| 2181 |
output_format: SAME
|
| 2182 |
weight_format: SAME
|
| 2183 |
transformer.h.36.ln_2:
|
| 2184 |
+
approximation_function: NONE
|
| 2185 |
bias_format: SAME
|
| 2186 |
input_format: SAME
|
| 2187 |
instance: LayerNorm
|
| 2188 |
output_format: SAME
|
| 2189 |
weight_format: SAME
|
| 2190 |
transformer.h.36.mlp.act:
|
| 2191 |
+
approximation_function: NONE
|
| 2192 |
input_format: SAME
|
| 2193 |
instance: GELU
|
| 2194 |
output_format: SAME
|
|
|
|
| 2240 |
instance: Dropout
|
| 2241 |
output_format: SAME
|
| 2242 |
transformer.h.37.attn.softmax:
|
| 2243 |
+
approximation_function: NONE
|
| 2244 |
input_format: SAME
|
| 2245 |
instance: Softmax
|
| 2246 |
output_format: SAME
|
| 2247 |
transformer.h.37.ln_1:
|
| 2248 |
+
approximation_function: NONE
|
| 2249 |
bias_format: SAME
|
| 2250 |
input_format: SAME
|
| 2251 |
instance: LayerNorm
|
| 2252 |
output_format: SAME
|
| 2253 |
weight_format: SAME
|
| 2254 |
transformer.h.37.ln_2:
|
| 2255 |
+
approximation_function: NONE
|
| 2256 |
bias_format: SAME
|
| 2257 |
input_format: SAME
|
| 2258 |
instance: LayerNorm
|
| 2259 |
output_format: SAME
|
| 2260 |
weight_format: SAME
|
| 2261 |
transformer.h.37.mlp.act:
|
| 2262 |
+
approximation_function: NONE
|
| 2263 |
input_format: SAME
|
| 2264 |
instance: GELU
|
| 2265 |
output_format: SAME
|
|
|
|
| 2311 |
instance: Dropout
|
| 2312 |
output_format: SAME
|
| 2313 |
transformer.h.38.attn.softmax:
|
| 2314 |
+
approximation_function: NONE
|
| 2315 |
input_format: SAME
|
| 2316 |
instance: Softmax
|
| 2317 |
output_format: SAME
|
| 2318 |
transformer.h.38.ln_1:
|
| 2319 |
+
approximation_function: NONE
|
| 2320 |
bias_format: SAME
|
| 2321 |
input_format: SAME
|
| 2322 |
instance: LayerNorm
|
| 2323 |
output_format: SAME
|
| 2324 |
weight_format: SAME
|
| 2325 |
transformer.h.38.ln_2:
|
| 2326 |
+
approximation_function: NONE
|
| 2327 |
bias_format: SAME
|
| 2328 |
input_format: SAME
|
| 2329 |
instance: LayerNorm
|
| 2330 |
output_format: SAME
|
| 2331 |
weight_format: SAME
|
| 2332 |
transformer.h.38.mlp.act:
|
| 2333 |
+
approximation_function: NONE
|
| 2334 |
input_format: SAME
|
| 2335 |
instance: GELU
|
| 2336 |
output_format: SAME
|
|
|
|
| 2382 |
instance: Dropout
|
| 2383 |
output_format: SAME
|
| 2384 |
transformer.h.39.attn.softmax:
|
| 2385 |
+
approximation_function: NONE
|
| 2386 |
input_format: SAME
|
| 2387 |
instance: Softmax
|
| 2388 |
output_format: SAME
|
| 2389 |
transformer.h.39.ln_1:
|
| 2390 |
+
approximation_function: NONE
|
| 2391 |
bias_format: SAME
|
| 2392 |
input_format: SAME
|
| 2393 |
instance: LayerNorm
|
| 2394 |
output_format: SAME
|
| 2395 |
weight_format: SAME
|
| 2396 |
transformer.h.39.ln_2:
|
| 2397 |
+
approximation_function: NONE
|
| 2398 |
bias_format: SAME
|
| 2399 |
input_format: SAME
|
| 2400 |
instance: LayerNorm
|
| 2401 |
output_format: SAME
|
| 2402 |
weight_format: SAME
|
| 2403 |
transformer.h.39.mlp.act:
|
| 2404 |
+
approximation_function: NONE
|
| 2405 |
input_format: SAME
|
| 2406 |
instance: GELU
|
| 2407 |
output_format: SAME
|
|
|
|
| 2453 |
instance: Dropout
|
| 2454 |
output_format: SAME
|
| 2455 |
transformer.h.4.attn.softmax:
|
| 2456 |
+
approximation_function: NONE
|
| 2457 |
input_format: SAME
|
| 2458 |
instance: Softmax
|
| 2459 |
output_format: SAME
|
| 2460 |
transformer.h.4.ln_1:
|
| 2461 |
+
approximation_function: NONE
|
| 2462 |
bias_format: SAME
|
| 2463 |
input_format: SAME
|
| 2464 |
instance: LayerNorm
|
| 2465 |
output_format: SAME
|
| 2466 |
weight_format: SAME
|
| 2467 |
transformer.h.4.ln_2:
|
| 2468 |
+
approximation_function: NONE
|
| 2469 |
bias_format: SAME
|
| 2470 |
input_format: SAME
|
| 2471 |
instance: LayerNorm
|
| 2472 |
output_format: SAME
|
| 2473 |
weight_format: SAME
|
| 2474 |
transformer.h.4.mlp.act:
|
| 2475 |
+
approximation_function: NONE
|
| 2476 |
input_format: SAME
|
| 2477 |
instance: GELU
|
| 2478 |
output_format: SAME
|
|
|
|
| 2524 |
instance: Dropout
|
| 2525 |
output_format: SAME
|
| 2526 |
transformer.h.40.attn.softmax:
|
| 2527 |
+
approximation_function: NONE
|
| 2528 |
input_format: SAME
|
| 2529 |
instance: Softmax
|
| 2530 |
output_format: SAME
|
| 2531 |
transformer.h.40.ln_1:
|
| 2532 |
+
approximation_function: NONE
|
| 2533 |
bias_format: SAME
|
| 2534 |
input_format: SAME
|
| 2535 |
instance: LayerNorm
|
| 2536 |
output_format: SAME
|
| 2537 |
weight_format: SAME
|
| 2538 |
transformer.h.40.ln_2:
|
| 2539 |
+
approximation_function: NONE
|
| 2540 |
bias_format: SAME
|
| 2541 |
input_format: SAME
|
| 2542 |
instance: LayerNorm
|
| 2543 |
output_format: SAME
|
| 2544 |
weight_format: SAME
|
| 2545 |
transformer.h.40.mlp.act:
|
| 2546 |
+
approximation_function: NONE
|
| 2547 |
input_format: SAME
|
| 2548 |
instance: GELU
|
| 2549 |
output_format: SAME
|
|
|
|
| 2595 |
instance: Dropout
|
| 2596 |
output_format: SAME
|
| 2597 |
transformer.h.41.attn.softmax:
|
| 2598 |
+
approximation_function: NONE
|
| 2599 |
input_format: SAME
|
| 2600 |
instance: Softmax
|
| 2601 |
output_format: SAME
|
| 2602 |
transformer.h.41.ln_1:
|
| 2603 |
+
approximation_function: NONE
|
| 2604 |
bias_format: SAME
|
| 2605 |
input_format: SAME
|
| 2606 |
instance: LayerNorm
|
| 2607 |
output_format: SAME
|
| 2608 |
weight_format: SAME
|
| 2609 |
transformer.h.41.ln_2:
|
| 2610 |
+
approximation_function: NONE
|
| 2611 |
bias_format: SAME
|
| 2612 |
input_format: SAME
|
| 2613 |
instance: LayerNorm
|
| 2614 |
output_format: SAME
|
| 2615 |
weight_format: SAME
|
| 2616 |
transformer.h.41.mlp.act:
|
| 2617 |
+
approximation_function: NONE
|
| 2618 |
input_format: SAME
|
| 2619 |
instance: GELU
|
| 2620 |
output_format: SAME
|
|
|
|
| 2666 |
instance: Dropout
|
| 2667 |
output_format: SAME
|
| 2668 |
transformer.h.42.attn.softmax:
|
| 2669 |
+
approximation_function: NONE
|
| 2670 |
input_format: SAME
|
| 2671 |
instance: Softmax
|
| 2672 |
output_format: SAME
|
| 2673 |
transformer.h.42.ln_1:
|
| 2674 |
+
approximation_function: NONE
|
| 2675 |
bias_format: SAME
|
| 2676 |
input_format: SAME
|
| 2677 |
instance: LayerNorm
|
| 2678 |
output_format: SAME
|
| 2679 |
weight_format: SAME
|
| 2680 |
transformer.h.42.ln_2:
|
| 2681 |
+
approximation_function: NONE
|
| 2682 |
bias_format: SAME
|
| 2683 |
input_format: SAME
|
| 2684 |
instance: LayerNorm
|
| 2685 |
output_format: SAME
|
| 2686 |
weight_format: SAME
|
| 2687 |
transformer.h.42.mlp.act:
|
| 2688 |
+
approximation_function: NONE
|
| 2689 |
input_format: SAME
|
| 2690 |
instance: GELU
|
| 2691 |
output_format: SAME
|
|
|
|
| 2737 |
instance: Dropout
|
| 2738 |
output_format: SAME
|
| 2739 |
transformer.h.43.attn.softmax:
|
| 2740 |
+
approximation_function: NONE
|
| 2741 |
input_format: SAME
|
| 2742 |
instance: Softmax
|
| 2743 |
output_format: SAME
|
| 2744 |
transformer.h.43.ln_1:
|
| 2745 |
+
approximation_function: NONE
|
| 2746 |
bias_format: SAME
|
| 2747 |
input_format: SAME
|
| 2748 |
instance: LayerNorm
|
| 2749 |
output_format: SAME
|
| 2750 |
weight_format: SAME
|
| 2751 |
transformer.h.43.ln_2:
|
| 2752 |
+
approximation_function: NONE
|
| 2753 |
bias_format: SAME
|
| 2754 |
input_format: SAME
|
| 2755 |
instance: LayerNorm
|
| 2756 |
output_format: SAME
|
| 2757 |
weight_format: SAME
|
| 2758 |
transformer.h.43.mlp.act:
|
| 2759 |
+
approximation_function: NONE
|
| 2760 |
input_format: SAME
|
| 2761 |
instance: GELU
|
| 2762 |
output_format: SAME
|
|
|
|
| 2808 |
instance: Dropout
|
| 2809 |
output_format: SAME
|
| 2810 |
transformer.h.44.attn.softmax:
|
| 2811 |
+
approximation_function: NONE
|
| 2812 |
input_format: SAME
|
| 2813 |
instance: Softmax
|
| 2814 |
output_format: SAME
|
| 2815 |
transformer.h.44.ln_1:
|
| 2816 |
+
approximation_function: NONE
|
| 2817 |
bias_format: SAME
|
| 2818 |
input_format: SAME
|
| 2819 |
instance: LayerNorm
|
| 2820 |
output_format: SAME
|
| 2821 |
weight_format: SAME
|
| 2822 |
transformer.h.44.ln_2:
|
| 2823 |
+
approximation_function: NONE
|
| 2824 |
bias_format: SAME
|
| 2825 |
input_format: SAME
|
| 2826 |
instance: LayerNorm
|
| 2827 |
output_format: SAME
|
| 2828 |
weight_format: SAME
|
| 2829 |
transformer.h.44.mlp.act:
|
| 2830 |
+
approximation_function: NONE
|
| 2831 |
input_format: SAME
|
| 2832 |
instance: GELU
|
| 2833 |
output_format: SAME
|
|
|
|
| 2879 |
instance: Dropout
|
| 2880 |
output_format: SAME
|
| 2881 |
transformer.h.45.attn.softmax:
|
| 2882 |
+
approximation_function: NONE
|
| 2883 |
input_format: SAME
|
| 2884 |
instance: Softmax
|
| 2885 |
output_format: SAME
|
| 2886 |
transformer.h.45.ln_1:
|
| 2887 |
+
approximation_function: NONE
|
| 2888 |
bias_format: SAME
|
| 2889 |
input_format: SAME
|
| 2890 |
instance: LayerNorm
|
| 2891 |
output_format: SAME
|
| 2892 |
weight_format: SAME
|
| 2893 |
transformer.h.45.ln_2:
|
| 2894 |
+
approximation_function: NONE
|
| 2895 |
bias_format: SAME
|
| 2896 |
input_format: SAME
|
| 2897 |
instance: LayerNorm
|
| 2898 |
output_format: SAME
|
| 2899 |
weight_format: SAME
|
| 2900 |
transformer.h.45.mlp.act:
|
| 2901 |
+
approximation_function: NONE
|
| 2902 |
input_format: SAME
|
| 2903 |
instance: GELU
|
| 2904 |
output_format: SAME
|
|
|
|
| 2950 |
instance: Dropout
|
| 2951 |
output_format: SAME
|
| 2952 |
transformer.h.46.attn.softmax:
|
| 2953 |
+
approximation_function: NONE
|
| 2954 |
input_format: SAME
|
| 2955 |
instance: Softmax
|
| 2956 |
output_format: SAME
|
| 2957 |
transformer.h.46.ln_1:
|
| 2958 |
+
approximation_function: NONE
|
| 2959 |
bias_format: SAME
|
| 2960 |
input_format: SAME
|
| 2961 |
instance: LayerNorm
|
| 2962 |
output_format: SAME
|
| 2963 |
weight_format: SAME
|
| 2964 |
transformer.h.46.ln_2:
|
| 2965 |
+
approximation_function: NONE
|
| 2966 |
bias_format: SAME
|
| 2967 |
input_format: SAME
|
| 2968 |
instance: LayerNorm
|
| 2969 |
output_format: SAME
|
| 2970 |
weight_format: SAME
|
| 2971 |
transformer.h.46.mlp.act:
|
| 2972 |
+
approximation_function: NONE
|
| 2973 |
input_format: SAME
|
| 2974 |
instance: GELU
|
| 2975 |
output_format: SAME
|
|
|
|
| 3021 |
instance: Dropout
|
| 3022 |
output_format: SAME
|
| 3023 |
transformer.h.47.attn.softmax:
|
| 3024 |
+
approximation_function: NONE
|
| 3025 |
input_format: SAME
|
| 3026 |
instance: Softmax
|
| 3027 |
output_format: SAME
|
| 3028 |
transformer.h.47.ln_1:
|
| 3029 |
+
approximation_function: NONE
|
| 3030 |
bias_format: SAME
|
| 3031 |
input_format: SAME
|
| 3032 |
instance: LayerNorm
|
| 3033 |
output_format: SAME
|
| 3034 |
weight_format: SAME
|
| 3035 |
transformer.h.47.ln_2:
|
| 3036 |
+
approximation_function: NONE
|
| 3037 |
bias_format: SAME
|
| 3038 |
input_format: SAME
|
| 3039 |
instance: LayerNorm
|
| 3040 |
output_format: SAME
|
| 3041 |
weight_format: SAME
|
| 3042 |
transformer.h.47.mlp.act:
|
| 3043 |
+
approximation_function: NONE
|
| 3044 |
input_format: SAME
|
| 3045 |
instance: GELU
|
| 3046 |
output_format: SAME
|
|
|
|
| 3092 |
instance: Dropout
|
| 3093 |
output_format: SAME
|
| 3094 |
transformer.h.5.attn.softmax:
|
| 3095 |
+
approximation_function: NONE
|
| 3096 |
input_format: SAME
|
| 3097 |
instance: Softmax
|
| 3098 |
output_format: SAME
|
| 3099 |
transformer.h.5.ln_1:
|
| 3100 |
+
approximation_function: NONE
|
| 3101 |
bias_format: SAME
|
| 3102 |
input_format: SAME
|
| 3103 |
instance: LayerNorm
|
| 3104 |
output_format: SAME
|
| 3105 |
weight_format: SAME
|
| 3106 |
transformer.h.5.ln_2:
|
| 3107 |
+
approximation_function: NONE
|
| 3108 |
bias_format: SAME
|
| 3109 |
input_format: SAME
|
| 3110 |
instance: LayerNorm
|
| 3111 |
output_format: SAME
|
| 3112 |
weight_format: SAME
|
| 3113 |
transformer.h.5.mlp.act:
|
| 3114 |
+
approximation_function: NONE
|
| 3115 |
input_format: SAME
|
| 3116 |
instance: GELU
|
| 3117 |
output_format: SAME
|
|
|
|
| 3163 |
instance: Dropout
|
| 3164 |
output_format: SAME
|
| 3165 |
transformer.h.6.attn.softmax:
|
| 3166 |
+
approximation_function: NONE
|
| 3167 |
input_format: SAME
|
| 3168 |
instance: Softmax
|
| 3169 |
output_format: SAME
|
| 3170 |
transformer.h.6.ln_1:
|
| 3171 |
+
approximation_function: NONE
|
| 3172 |
bias_format: SAME
|
| 3173 |
input_format: SAME
|
| 3174 |
instance: LayerNorm
|
| 3175 |
output_format: SAME
|
| 3176 |
weight_format: SAME
|
| 3177 |
transformer.h.6.ln_2:
|
| 3178 |
+
approximation_function: NONE
|
| 3179 |
bias_format: SAME
|
| 3180 |
input_format: SAME
|
| 3181 |
instance: LayerNorm
|
| 3182 |
output_format: SAME
|
| 3183 |
weight_format: SAME
|
| 3184 |
transformer.h.6.mlp.act:
|
| 3185 |
+
approximation_function: NONE
|
| 3186 |
input_format: SAME
|
| 3187 |
instance: GELU
|
| 3188 |
output_format: SAME
|
|
|
|
| 3234 |
instance: Dropout
|
| 3235 |
output_format: SAME
|
| 3236 |
transformer.h.7.attn.softmax:
|
| 3237 |
+
approximation_function: NONE
|
| 3238 |
input_format: SAME
|
| 3239 |
instance: Softmax
|
| 3240 |
output_format: SAME
|
| 3241 |
transformer.h.7.ln_1:
|
| 3242 |
+
approximation_function: NONE
|
| 3243 |
bias_format: SAME
|
| 3244 |
input_format: SAME
|
| 3245 |
instance: LayerNorm
|
| 3246 |
output_format: SAME
|
| 3247 |
weight_format: SAME
|
| 3248 |
transformer.h.7.ln_2:
|
| 3249 |
+
approximation_function: NONE
|
| 3250 |
bias_format: SAME
|
| 3251 |
input_format: SAME
|
| 3252 |
instance: LayerNorm
|
| 3253 |
output_format: SAME
|
| 3254 |
weight_format: SAME
|
| 3255 |
transformer.h.7.mlp.act:
|
| 3256 |
+
approximation_function: NONE
|
| 3257 |
input_format: SAME
|
| 3258 |
instance: GELU
|
| 3259 |
output_format: SAME
|
|
|
|
| 3305 |
instance: Dropout
|
| 3306 |
output_format: SAME
|
| 3307 |
transformer.h.8.attn.softmax:
|
| 3308 |
+
approximation_function: NONE
|
| 3309 |
input_format: SAME
|
| 3310 |
instance: Softmax
|
| 3311 |
output_format: SAME
|
| 3312 |
transformer.h.8.ln_1:
|
| 3313 |
+
approximation_function: NONE
|
| 3314 |
bias_format: SAME
|
| 3315 |
input_format: SAME
|
| 3316 |
instance: LayerNorm
|
| 3317 |
output_format: SAME
|
| 3318 |
weight_format: SAME
|
| 3319 |
transformer.h.8.ln_2:
|
| 3320 |
+
approximation_function: NONE
|
| 3321 |
bias_format: SAME
|
| 3322 |
input_format: SAME
|
| 3323 |
instance: LayerNorm
|
| 3324 |
output_format: SAME
|
| 3325 |
weight_format: SAME
|
| 3326 |
transformer.h.8.mlp.act:
|
| 3327 |
+
approximation_function: NONE
|
| 3328 |
input_format: SAME
|
| 3329 |
instance: GELU
|
| 3330 |
output_format: SAME
|
|
|
|
| 3376 |
instance: Dropout
|
| 3377 |
output_format: SAME
|
| 3378 |
transformer.h.9.attn.softmax:
|
| 3379 |
+
approximation_function: NONE
|
| 3380 |
input_format: SAME
|
| 3381 |
instance: Softmax
|
| 3382 |
output_format: SAME
|
| 3383 |
transformer.h.9.ln_1:
|
| 3384 |
+
approximation_function: NONE
|
| 3385 |
bias_format: SAME
|
| 3386 |
input_format: SAME
|
| 3387 |
instance: LayerNorm
|
| 3388 |
output_format: SAME
|
| 3389 |
weight_format: SAME
|
| 3390 |
transformer.h.9.ln_2:
|
| 3391 |
+
approximation_function: NONE
|
| 3392 |
bias_format: SAME
|
| 3393 |
input_format: SAME
|
| 3394 |
instance: LayerNorm
|
| 3395 |
output_format: SAME
|
| 3396 |
weight_format: SAME
|
| 3397 |
transformer.h.9.mlp.act:
|
| 3398 |
+
approximation_function: NONE
|
| 3399 |
input_format: SAME
|
| 3400 |
instance: GELU
|
| 3401 |
output_format: SAME
|
|
|
|
| 3421 |
instance: Dropout
|
| 3422 |
output_format: SAME
|
| 3423 |
transformer.ln_f:
|
| 3424 |
+
approximation_function: NONE
|
| 3425 |
bias_format: SAME
|
| 3426 |
input_format: SAME
|
| 3427 |
instance: LayerNorm
|