Snider Cladius Maximus commited on
Commit ·
e24af92
1
Parent(s): 4d09a58
fix: exact configs from mlx-community/unsloth references
Browse filesEach quant level uses configs from the matching reference repo
(4bit from 4bit ref, 8bit from 8bit ref — not derived).
Fresh weights from full multimodal Google base model.
Co-Authored-By: Cladius Maximus <cladius@lethean.io>
- config.json +492 -492
config.json
CHANGED
|
@@ -23,23 +23,23 @@
|
|
| 23 |
"bits": 4,
|
| 24 |
"language_model.model.embed_tokens": {
|
| 25 |
"group_size": 64,
|
| 26 |
-
"bits":
|
| 27 |
},
|
| 28 |
"language_model.model.layers.0.self_attn.q_proj": {
|
| 29 |
"group_size": 64,
|
| 30 |
-
"bits":
|
| 31 |
},
|
| 32 |
"language_model.model.layers.0.self_attn.k_proj": {
|
| 33 |
"group_size": 64,
|
| 34 |
-
"bits":
|
| 35 |
},
|
| 36 |
"language_model.model.layers.0.self_attn.v_proj": {
|
| 37 |
"group_size": 64,
|
| 38 |
-
"bits":
|
| 39 |
},
|
| 40 |
"language_model.model.layers.0.self_attn.o_proj": {
|
| 41 |
"group_size": 64,
|
| 42 |
-
"bits":
|
| 43 |
},
|
| 44 |
"language_model.model.layers.0.mlp.gate_proj": {
|
| 45 |
"group_size": 64,
|
|
@@ -55,47 +55,47 @@
|
|
| 55 |
},
|
| 56 |
"language_model.model.layers.1.self_attn.q_proj": {
|
| 57 |
"group_size": 64,
|
| 58 |
-
"bits":
|
| 59 |
},
|
| 60 |
"language_model.model.layers.1.self_attn.k_proj": {
|
| 61 |
"group_size": 64,
|
| 62 |
-
"bits":
|
| 63 |
},
|
| 64 |
"language_model.model.layers.1.self_attn.v_proj": {
|
| 65 |
"group_size": 64,
|
| 66 |
-
"bits":
|
| 67 |
},
|
| 68 |
"language_model.model.layers.1.self_attn.o_proj": {
|
| 69 |
"group_size": 64,
|
| 70 |
-
"bits":
|
| 71 |
},
|
| 72 |
"language_model.model.layers.1.mlp.gate_proj": {
|
| 73 |
"group_size": 64,
|
| 74 |
-
"bits":
|
| 75 |
},
|
| 76 |
"language_model.model.layers.1.mlp.up_proj": {
|
| 77 |
"group_size": 64,
|
| 78 |
-
"bits":
|
| 79 |
},
|
| 80 |
"language_model.model.layers.1.mlp.down_proj": {
|
| 81 |
"group_size": 64,
|
| 82 |
-
"bits":
|
| 83 |
},
|
| 84 |
"language_model.model.layers.2.self_attn.q_proj": {
|
| 85 |
"group_size": 64,
|
| 86 |
-
"bits":
|
| 87 |
},
|
| 88 |
"language_model.model.layers.2.self_attn.k_proj": {
|
| 89 |
"group_size": 64,
|
| 90 |
-
"bits":
|
| 91 |
},
|
| 92 |
"language_model.model.layers.2.self_attn.v_proj": {
|
| 93 |
"group_size": 64,
|
| 94 |
-
"bits":
|
| 95 |
},
|
| 96 |
"language_model.model.layers.2.self_attn.o_proj": {
|
| 97 |
"group_size": 64,
|
| 98 |
-
"bits":
|
| 99 |
},
|
| 100 |
"language_model.model.layers.2.mlp.gate_proj": {
|
| 101 |
"group_size": 64,
|
|
@@ -111,19 +111,19 @@
|
|
| 111 |
},
|
| 112 |
"language_model.model.layers.3.self_attn.q_proj": {
|
| 113 |
"group_size": 64,
|
| 114 |
-
"bits":
|
| 115 |
},
|
| 116 |
"language_model.model.layers.3.self_attn.k_proj": {
|
| 117 |
"group_size": 64,
|
| 118 |
-
"bits":
|
| 119 |
},
|
| 120 |
"language_model.model.layers.3.self_attn.v_proj": {
|
| 121 |
"group_size": 64,
|
| 122 |
-
"bits":
|
| 123 |
},
|
| 124 |
"language_model.model.layers.3.self_attn.o_proj": {
|
| 125 |
"group_size": 64,
|
| 126 |
-
"bits":
|
| 127 |
},
|
| 128 |
"language_model.model.layers.3.mlp.gate_proj": {
|
| 129 |
"group_size": 64,
|
|
@@ -139,19 +139,19 @@
|
|
| 139 |
},
|
| 140 |
"language_model.model.layers.4.self_attn.q_proj": {
|
| 141 |
"group_size": 64,
|
| 142 |
-
"bits":
|
| 143 |
},
|
| 144 |
"language_model.model.layers.4.self_attn.k_proj": {
|
| 145 |
"group_size": 64,
|
| 146 |
-
"bits":
|
| 147 |
},
|
| 148 |
"language_model.model.layers.4.self_attn.v_proj": {
|
| 149 |
"group_size": 64,
|
| 150 |
-
"bits":
|
| 151 |
},
|
| 152 |
"language_model.model.layers.4.self_attn.o_proj": {
|
| 153 |
"group_size": 64,
|
| 154 |
-
"bits":
|
| 155 |
},
|
| 156 |
"language_model.model.layers.4.mlp.gate_proj": {
|
| 157 |
"group_size": 64,
|
|
@@ -167,15 +167,15 @@
|
|
| 167 |
},
|
| 168 |
"language_model.model.layers.5.self_attn.q_proj": {
|
| 169 |
"group_size": 64,
|
| 170 |
-
"bits":
|
| 171 |
},
|
| 172 |
"language_model.model.layers.5.self_attn.k_proj": {
|
| 173 |
"group_size": 64,
|
| 174 |
-
"bits":
|
| 175 |
},
|
| 176 |
"language_model.model.layers.5.self_attn.o_proj": {
|
| 177 |
"group_size": 64,
|
| 178 |
-
"bits":
|
| 179 |
},
|
| 180 |
"language_model.model.layers.5.mlp.gate_proj": {
|
| 181 |
"group_size": 64,
|
|
@@ -191,19 +191,19 @@
|
|
| 191 |
},
|
| 192 |
"language_model.model.layers.6.self_attn.q_proj": {
|
| 193 |
"group_size": 64,
|
| 194 |
-
"bits":
|
| 195 |
},
|
| 196 |
"language_model.model.layers.6.self_attn.k_proj": {
|
| 197 |
"group_size": 64,
|
| 198 |
-
"bits":
|
| 199 |
},
|
| 200 |
"language_model.model.layers.6.self_attn.v_proj": {
|
| 201 |
"group_size": 64,
|
| 202 |
-
"bits":
|
| 203 |
},
|
| 204 |
"language_model.model.layers.6.self_attn.o_proj": {
|
| 205 |
"group_size": 64,
|
| 206 |
-
"bits":
|
| 207 |
},
|
| 208 |
"language_model.model.layers.6.mlp.gate_proj": {
|
| 209 |
"group_size": 64,
|
|
@@ -219,19 +219,19 @@
|
|
| 219 |
},
|
| 220 |
"language_model.model.layers.7.self_attn.q_proj": {
|
| 221 |
"group_size": 64,
|
| 222 |
-
"bits":
|
| 223 |
},
|
| 224 |
"language_model.model.layers.7.self_attn.k_proj": {
|
| 225 |
"group_size": 64,
|
| 226 |
-
"bits":
|
| 227 |
},
|
| 228 |
"language_model.model.layers.7.self_attn.v_proj": {
|
| 229 |
"group_size": 64,
|
| 230 |
-
"bits":
|
| 231 |
},
|
| 232 |
"language_model.model.layers.7.self_attn.o_proj": {
|
| 233 |
"group_size": 64,
|
| 234 |
-
"bits":
|
| 235 |
},
|
| 236 |
"language_model.model.layers.7.mlp.gate_proj": {
|
| 237 |
"group_size": 64,
|
|
@@ -247,19 +247,19 @@
|
|
| 247 |
},
|
| 248 |
"language_model.model.layers.8.self_attn.q_proj": {
|
| 249 |
"group_size": 64,
|
| 250 |
-
"bits":
|
| 251 |
},
|
| 252 |
"language_model.model.layers.8.self_attn.k_proj": {
|
| 253 |
"group_size": 64,
|
| 254 |
-
"bits":
|
| 255 |
},
|
| 256 |
"language_model.model.layers.8.self_attn.v_proj": {
|
| 257 |
"group_size": 64,
|
| 258 |
-
"bits":
|
| 259 |
},
|
| 260 |
"language_model.model.layers.8.self_attn.o_proj": {
|
| 261 |
"group_size": 64,
|
| 262 |
-
"bits":
|
| 263 |
},
|
| 264 |
"language_model.model.layers.8.mlp.gate_proj": {
|
| 265 |
"group_size": 64,
|
|
@@ -275,19 +275,19 @@
|
|
| 275 |
},
|
| 276 |
"language_model.model.layers.9.self_attn.q_proj": {
|
| 277 |
"group_size": 64,
|
| 278 |
-
"bits":
|
| 279 |
},
|
| 280 |
"language_model.model.layers.9.self_attn.k_proj": {
|
| 281 |
"group_size": 64,
|
| 282 |
-
"bits":
|
| 283 |
},
|
| 284 |
"language_model.model.layers.9.self_attn.v_proj": {
|
| 285 |
"group_size": 64,
|
| 286 |
-
"bits":
|
| 287 |
},
|
| 288 |
"language_model.model.layers.9.self_attn.o_proj": {
|
| 289 |
"group_size": 64,
|
| 290 |
-
"bits":
|
| 291 |
},
|
| 292 |
"language_model.model.layers.9.mlp.gate_proj": {
|
| 293 |
"group_size": 64,
|
|
@@ -303,19 +303,19 @@
|
|
| 303 |
},
|
| 304 |
"language_model.model.layers.10.self_attn.q_proj": {
|
| 305 |
"group_size": 64,
|
| 306 |
-
"bits":
|
| 307 |
},
|
| 308 |
"language_model.model.layers.10.self_attn.k_proj": {
|
| 309 |
"group_size": 64,
|
| 310 |
-
"bits":
|
| 311 |
},
|
| 312 |
"language_model.model.layers.10.self_attn.v_proj": {
|
| 313 |
"group_size": 64,
|
| 314 |
-
"bits":
|
| 315 |
},
|
| 316 |
"language_model.model.layers.10.self_attn.o_proj": {
|
| 317 |
"group_size": 64,
|
| 318 |
-
"bits":
|
| 319 |
},
|
| 320 |
"language_model.model.layers.10.mlp.gate_proj": {
|
| 321 |
"group_size": 64,
|
|
@@ -331,15 +331,15 @@
|
|
| 331 |
},
|
| 332 |
"language_model.model.layers.11.self_attn.q_proj": {
|
| 333 |
"group_size": 64,
|
| 334 |
-
"bits":
|
| 335 |
},
|
| 336 |
"language_model.model.layers.11.self_attn.k_proj": {
|
| 337 |
"group_size": 64,
|
| 338 |
-
"bits":
|
| 339 |
},
|
| 340 |
"language_model.model.layers.11.self_attn.o_proj": {
|
| 341 |
"group_size": 64,
|
| 342 |
-
"bits":
|
| 343 |
},
|
| 344 |
"language_model.model.layers.11.mlp.gate_proj": {
|
| 345 |
"group_size": 64,
|
|
@@ -355,19 +355,19 @@
|
|
| 355 |
},
|
| 356 |
"language_model.model.layers.12.self_attn.q_proj": {
|
| 357 |
"group_size": 64,
|
| 358 |
-
"bits":
|
| 359 |
},
|
| 360 |
"language_model.model.layers.12.self_attn.k_proj": {
|
| 361 |
"group_size": 64,
|
| 362 |
-
"bits":
|
| 363 |
},
|
| 364 |
"language_model.model.layers.12.self_attn.v_proj": {
|
| 365 |
"group_size": 64,
|
| 366 |
-
"bits":
|
| 367 |
},
|
| 368 |
"language_model.model.layers.12.self_attn.o_proj": {
|
| 369 |
"group_size": 64,
|
| 370 |
-
"bits":
|
| 371 |
},
|
| 372 |
"language_model.model.layers.12.mlp.gate_proj": {
|
| 373 |
"group_size": 64,
|
|
@@ -383,19 +383,19 @@
|
|
| 383 |
},
|
| 384 |
"language_model.model.layers.13.self_attn.q_proj": {
|
| 385 |
"group_size": 64,
|
| 386 |
-
"bits":
|
| 387 |
},
|
| 388 |
"language_model.model.layers.13.self_attn.k_proj": {
|
| 389 |
"group_size": 64,
|
| 390 |
-
"bits":
|
| 391 |
},
|
| 392 |
"language_model.model.layers.13.self_attn.v_proj": {
|
| 393 |
"group_size": 64,
|
| 394 |
-
"bits":
|
| 395 |
},
|
| 396 |
"language_model.model.layers.13.self_attn.o_proj": {
|
| 397 |
"group_size": 64,
|
| 398 |
-
"bits":
|
| 399 |
},
|
| 400 |
"language_model.model.layers.13.mlp.gate_proj": {
|
| 401 |
"group_size": 64,
|
|
@@ -411,19 +411,19 @@
|
|
| 411 |
},
|
| 412 |
"language_model.model.layers.14.self_attn.q_proj": {
|
| 413 |
"group_size": 64,
|
| 414 |
-
"bits":
|
| 415 |
},
|
| 416 |
"language_model.model.layers.14.self_attn.k_proj": {
|
| 417 |
"group_size": 64,
|
| 418 |
-
"bits":
|
| 419 |
},
|
| 420 |
"language_model.model.layers.14.self_attn.v_proj": {
|
| 421 |
"group_size": 64,
|
| 422 |
-
"bits":
|
| 423 |
},
|
| 424 |
"language_model.model.layers.14.self_attn.o_proj": {
|
| 425 |
"group_size": 64,
|
| 426 |
-
"bits":
|
| 427 |
},
|
| 428 |
"language_model.model.layers.14.mlp.gate_proj": {
|
| 429 |
"group_size": 64,
|
|
@@ -439,19 +439,19 @@
|
|
| 439 |
},
|
| 440 |
"language_model.model.layers.15.self_attn.q_proj": {
|
| 441 |
"group_size": 64,
|
| 442 |
-
"bits":
|
| 443 |
},
|
| 444 |
"language_model.model.layers.15.self_attn.k_proj": {
|
| 445 |
"group_size": 64,
|
| 446 |
-
"bits":
|
| 447 |
},
|
| 448 |
"language_model.model.layers.15.self_attn.v_proj": {
|
| 449 |
"group_size": 64,
|
| 450 |
-
"bits":
|
| 451 |
},
|
| 452 |
"language_model.model.layers.15.self_attn.o_proj": {
|
| 453 |
"group_size": 64,
|
| 454 |
-
"bits":
|
| 455 |
},
|
| 456 |
"language_model.model.layers.15.mlp.gate_proj": {
|
| 457 |
"group_size": 64,
|
|
@@ -467,19 +467,19 @@
|
|
| 467 |
},
|
| 468 |
"language_model.model.layers.16.self_attn.q_proj": {
|
| 469 |
"group_size": 64,
|
| 470 |
-
"bits":
|
| 471 |
},
|
| 472 |
"language_model.model.layers.16.self_attn.k_proj": {
|
| 473 |
"group_size": 64,
|
| 474 |
-
"bits":
|
| 475 |
},
|
| 476 |
"language_model.model.layers.16.self_attn.v_proj": {
|
| 477 |
"group_size": 64,
|
| 478 |
-
"bits":
|
| 479 |
},
|
| 480 |
"language_model.model.layers.16.self_attn.o_proj": {
|
| 481 |
"group_size": 64,
|
| 482 |
-
"bits":
|
| 483 |
},
|
| 484 |
"language_model.model.layers.16.mlp.gate_proj": {
|
| 485 |
"group_size": 64,
|
|
@@ -495,15 +495,15 @@
|
|
| 495 |
},
|
| 496 |
"language_model.model.layers.17.self_attn.q_proj": {
|
| 497 |
"group_size": 64,
|
| 498 |
-
"bits":
|
| 499 |
},
|
| 500 |
"language_model.model.layers.17.self_attn.k_proj": {
|
| 501 |
"group_size": 64,
|
| 502 |
-
"bits":
|
| 503 |
},
|
| 504 |
"language_model.model.layers.17.self_attn.o_proj": {
|
| 505 |
"group_size": 64,
|
| 506 |
-
"bits":
|
| 507 |
},
|
| 508 |
"language_model.model.layers.17.mlp.gate_proj": {
|
| 509 |
"group_size": 64,
|
|
@@ -519,19 +519,19 @@
|
|
| 519 |
},
|
| 520 |
"language_model.model.layers.18.self_attn.q_proj": {
|
| 521 |
"group_size": 64,
|
| 522 |
-
"bits":
|
| 523 |
},
|
| 524 |
"language_model.model.layers.18.self_attn.k_proj": {
|
| 525 |
"group_size": 64,
|
| 526 |
-
"bits":
|
| 527 |
},
|
| 528 |
"language_model.model.layers.18.self_attn.v_proj": {
|
| 529 |
"group_size": 64,
|
| 530 |
-
"bits":
|
| 531 |
},
|
| 532 |
"language_model.model.layers.18.self_attn.o_proj": {
|
| 533 |
"group_size": 64,
|
| 534 |
-
"bits":
|
| 535 |
},
|
| 536 |
"language_model.model.layers.18.mlp.gate_proj": {
|
| 537 |
"group_size": 64,
|
|
@@ -547,19 +547,19 @@
|
|
| 547 |
},
|
| 548 |
"language_model.model.layers.19.self_attn.q_proj": {
|
| 549 |
"group_size": 64,
|
| 550 |
-
"bits":
|
| 551 |
},
|
| 552 |
"language_model.model.layers.19.self_attn.k_proj": {
|
| 553 |
"group_size": 64,
|
| 554 |
-
"bits":
|
| 555 |
},
|
| 556 |
"language_model.model.layers.19.self_attn.v_proj": {
|
| 557 |
"group_size": 64,
|
| 558 |
-
"bits":
|
| 559 |
},
|
| 560 |
"language_model.model.layers.19.self_attn.o_proj": {
|
| 561 |
"group_size": 64,
|
| 562 |
-
"bits":
|
| 563 |
},
|
| 564 |
"language_model.model.layers.19.mlp.gate_proj": {
|
| 565 |
"group_size": 64,
|
|
@@ -575,19 +575,19 @@
|
|
| 575 |
},
|
| 576 |
"language_model.model.layers.20.self_attn.q_proj": {
|
| 577 |
"group_size": 64,
|
| 578 |
-
"bits":
|
| 579 |
},
|
| 580 |
"language_model.model.layers.20.self_attn.k_proj": {
|
| 581 |
"group_size": 64,
|
| 582 |
-
"bits":
|
| 583 |
},
|
| 584 |
"language_model.model.layers.20.self_attn.v_proj": {
|
| 585 |
"group_size": 64,
|
| 586 |
-
"bits":
|
| 587 |
},
|
| 588 |
"language_model.model.layers.20.self_attn.o_proj": {
|
| 589 |
"group_size": 64,
|
| 590 |
-
"bits":
|
| 591 |
},
|
| 592 |
"language_model.model.layers.20.mlp.gate_proj": {
|
| 593 |
"group_size": 64,
|
|
@@ -603,19 +603,19 @@
|
|
| 603 |
},
|
| 604 |
"language_model.model.layers.21.self_attn.q_proj": {
|
| 605 |
"group_size": 64,
|
| 606 |
-
"bits":
|
| 607 |
},
|
| 608 |
"language_model.model.layers.21.self_attn.k_proj": {
|
| 609 |
"group_size": 64,
|
| 610 |
-
"bits":
|
| 611 |
},
|
| 612 |
"language_model.model.layers.21.self_attn.v_proj": {
|
| 613 |
"group_size": 64,
|
| 614 |
-
"bits":
|
| 615 |
},
|
| 616 |
"language_model.model.layers.21.self_attn.o_proj": {
|
| 617 |
"group_size": 64,
|
| 618 |
-
"bits":
|
| 619 |
},
|
| 620 |
"language_model.model.layers.21.mlp.gate_proj": {
|
| 621 |
"group_size": 64,
|
|
@@ -631,19 +631,19 @@
|
|
| 631 |
},
|
| 632 |
"language_model.model.layers.22.self_attn.q_proj": {
|
| 633 |
"group_size": 64,
|
| 634 |
-
"bits":
|
| 635 |
},
|
| 636 |
"language_model.model.layers.22.self_attn.k_proj": {
|
| 637 |
"group_size": 64,
|
| 638 |
-
"bits":
|
| 639 |
},
|
| 640 |
"language_model.model.layers.22.self_attn.v_proj": {
|
| 641 |
"group_size": 64,
|
| 642 |
-
"bits":
|
| 643 |
},
|
| 644 |
"language_model.model.layers.22.self_attn.o_proj": {
|
| 645 |
"group_size": 64,
|
| 646 |
-
"bits":
|
| 647 |
},
|
| 648 |
"language_model.model.layers.22.mlp.gate_proj": {
|
| 649 |
"group_size": 64,
|
|
@@ -659,15 +659,15 @@
|
|
| 659 |
},
|
| 660 |
"language_model.model.layers.23.self_attn.q_proj": {
|
| 661 |
"group_size": 64,
|
| 662 |
-
"bits":
|
| 663 |
},
|
| 664 |
"language_model.model.layers.23.self_attn.k_proj": {
|
| 665 |
"group_size": 64,
|
| 666 |
-
"bits":
|
| 667 |
},
|
| 668 |
"language_model.model.layers.23.self_attn.o_proj": {
|
| 669 |
"group_size": 64,
|
| 670 |
-
"bits":
|
| 671 |
},
|
| 672 |
"language_model.model.layers.23.mlp.gate_proj": {
|
| 673 |
"group_size": 64,
|
|
@@ -683,19 +683,19 @@
|
|
| 683 |
},
|
| 684 |
"language_model.model.layers.24.self_attn.q_proj": {
|
| 685 |
"group_size": 64,
|
| 686 |
-
"bits":
|
| 687 |
},
|
| 688 |
"language_model.model.layers.24.self_attn.k_proj": {
|
| 689 |
"group_size": 64,
|
| 690 |
-
"bits":
|
| 691 |
},
|
| 692 |
"language_model.model.layers.24.self_attn.v_proj": {
|
| 693 |
"group_size": 64,
|
| 694 |
-
"bits":
|
| 695 |
},
|
| 696 |
"language_model.model.layers.24.self_attn.o_proj": {
|
| 697 |
"group_size": 64,
|
| 698 |
-
"bits":
|
| 699 |
},
|
| 700 |
"language_model.model.layers.24.mlp.gate_proj": {
|
| 701 |
"group_size": 64,
|
|
@@ -711,19 +711,19 @@
|
|
| 711 |
},
|
| 712 |
"language_model.model.layers.25.self_attn.q_proj": {
|
| 713 |
"group_size": 64,
|
| 714 |
-
"bits":
|
| 715 |
},
|
| 716 |
"language_model.model.layers.25.self_attn.k_proj": {
|
| 717 |
"group_size": 64,
|
| 718 |
-
"bits":
|
| 719 |
},
|
| 720 |
"language_model.model.layers.25.self_attn.v_proj": {
|
| 721 |
"group_size": 64,
|
| 722 |
-
"bits":
|
| 723 |
},
|
| 724 |
"language_model.model.layers.25.self_attn.o_proj": {
|
| 725 |
"group_size": 64,
|
| 726 |
-
"bits":
|
| 727 |
},
|
| 728 |
"language_model.model.layers.25.mlp.gate_proj": {
|
| 729 |
"group_size": 64,
|
|
@@ -739,19 +739,19 @@
|
|
| 739 |
},
|
| 740 |
"language_model.model.layers.26.self_attn.q_proj": {
|
| 741 |
"group_size": 64,
|
| 742 |
-
"bits":
|
| 743 |
},
|
| 744 |
"language_model.model.layers.26.self_attn.k_proj": {
|
| 745 |
"group_size": 64,
|
| 746 |
-
"bits":
|
| 747 |
},
|
| 748 |
"language_model.model.layers.26.self_attn.v_proj": {
|
| 749 |
"group_size": 64,
|
| 750 |
-
"bits":
|
| 751 |
},
|
| 752 |
"language_model.model.layers.26.self_attn.o_proj": {
|
| 753 |
"group_size": 64,
|
| 754 |
-
"bits":
|
| 755 |
},
|
| 756 |
"language_model.model.layers.26.mlp.gate_proj": {
|
| 757 |
"group_size": 64,
|
|
@@ -767,19 +767,19 @@
|
|
| 767 |
},
|
| 768 |
"language_model.model.layers.27.self_attn.q_proj": {
|
| 769 |
"group_size": 64,
|
| 770 |
-
"bits":
|
| 771 |
},
|
| 772 |
"language_model.model.layers.27.self_attn.k_proj": {
|
| 773 |
"group_size": 64,
|
| 774 |
-
"bits":
|
| 775 |
},
|
| 776 |
"language_model.model.layers.27.self_attn.v_proj": {
|
| 777 |
"group_size": 64,
|
| 778 |
-
"bits":
|
| 779 |
},
|
| 780 |
"language_model.model.layers.27.self_attn.o_proj": {
|
| 781 |
"group_size": 64,
|
| 782 |
-
"bits":
|
| 783 |
},
|
| 784 |
"language_model.model.layers.27.mlp.gate_proj": {
|
| 785 |
"group_size": 64,
|
|
@@ -795,19 +795,19 @@
|
|
| 795 |
},
|
| 796 |
"language_model.model.layers.28.self_attn.q_proj": {
|
| 797 |
"group_size": 64,
|
| 798 |
-
"bits":
|
| 799 |
},
|
| 800 |
"language_model.model.layers.28.self_attn.k_proj": {
|
| 801 |
"group_size": 64,
|
| 802 |
-
"bits":
|
| 803 |
},
|
| 804 |
"language_model.model.layers.28.self_attn.v_proj": {
|
| 805 |
"group_size": 64,
|
| 806 |
-
"bits":
|
| 807 |
},
|
| 808 |
"language_model.model.layers.28.self_attn.o_proj": {
|
| 809 |
"group_size": 64,
|
| 810 |
-
"bits":
|
| 811 |
},
|
| 812 |
"language_model.model.layers.28.mlp.gate_proj": {
|
| 813 |
"group_size": 64,
|
|
@@ -823,15 +823,15 @@
|
|
| 823 |
},
|
| 824 |
"language_model.model.layers.29.self_attn.q_proj": {
|
| 825 |
"group_size": 64,
|
| 826 |
-
"bits":
|
| 827 |
},
|
| 828 |
"language_model.model.layers.29.self_attn.k_proj": {
|
| 829 |
"group_size": 64,
|
| 830 |
-
"bits":
|
| 831 |
},
|
| 832 |
"language_model.model.layers.29.self_attn.o_proj": {
|
| 833 |
"group_size": 64,
|
| 834 |
-
"bits":
|
| 835 |
},
|
| 836 |
"language_model.model.layers.29.mlp.gate_proj": {
|
| 837 |
"group_size": 64,
|
|
@@ -847,19 +847,19 @@
|
|
| 847 |
},
|
| 848 |
"language_model.model.layers.30.self_attn.q_proj": {
|
| 849 |
"group_size": 64,
|
| 850 |
-
"bits":
|
| 851 |
},
|
| 852 |
"language_model.model.layers.30.self_attn.k_proj": {
|
| 853 |
"group_size": 64,
|
| 854 |
-
"bits":
|
| 855 |
},
|
| 856 |
"language_model.model.layers.30.self_attn.v_proj": {
|
| 857 |
"group_size": 64,
|
| 858 |
-
"bits":
|
| 859 |
},
|
| 860 |
"language_model.model.layers.30.self_attn.o_proj": {
|
| 861 |
"group_size": 64,
|
| 862 |
-
"bits":
|
| 863 |
},
|
| 864 |
"language_model.model.layers.30.mlp.gate_proj": {
|
| 865 |
"group_size": 64,
|
|
@@ -875,19 +875,19 @@
|
|
| 875 |
},
|
| 876 |
"language_model.model.layers.31.self_attn.q_proj": {
|
| 877 |
"group_size": 64,
|
| 878 |
-
"bits":
|
| 879 |
},
|
| 880 |
"language_model.model.layers.31.self_attn.k_proj": {
|
| 881 |
"group_size": 64,
|
| 882 |
-
"bits":
|
| 883 |
},
|
| 884 |
"language_model.model.layers.31.self_attn.v_proj": {
|
| 885 |
"group_size": 64,
|
| 886 |
-
"bits":
|
| 887 |
},
|
| 888 |
"language_model.model.layers.31.self_attn.o_proj": {
|
| 889 |
"group_size": 64,
|
| 890 |
-
"bits":
|
| 891 |
},
|
| 892 |
"language_model.model.layers.31.mlp.gate_proj": {
|
| 893 |
"group_size": 64,
|
|
@@ -903,19 +903,19 @@
|
|
| 903 |
},
|
| 904 |
"language_model.model.layers.32.self_attn.q_proj": {
|
| 905 |
"group_size": 64,
|
| 906 |
-
"bits":
|
| 907 |
},
|
| 908 |
"language_model.model.layers.32.self_attn.k_proj": {
|
| 909 |
"group_size": 64,
|
| 910 |
-
"bits":
|
| 911 |
},
|
| 912 |
"language_model.model.layers.32.self_attn.v_proj": {
|
| 913 |
"group_size": 64,
|
| 914 |
-
"bits":
|
| 915 |
},
|
| 916 |
"language_model.model.layers.32.self_attn.o_proj": {
|
| 917 |
"group_size": 64,
|
| 918 |
-
"bits":
|
| 919 |
},
|
| 920 |
"language_model.model.layers.32.mlp.gate_proj": {
|
| 921 |
"group_size": 64,
|
|
@@ -931,19 +931,19 @@
|
|
| 931 |
},
|
| 932 |
"language_model.model.layers.33.self_attn.q_proj": {
|
| 933 |
"group_size": 64,
|
| 934 |
-
"bits":
|
| 935 |
},
|
| 936 |
"language_model.model.layers.33.self_attn.k_proj": {
|
| 937 |
"group_size": 64,
|
| 938 |
-
"bits":
|
| 939 |
},
|
| 940 |
"language_model.model.layers.33.self_attn.v_proj": {
|
| 941 |
"group_size": 64,
|
| 942 |
-
"bits":
|
| 943 |
},
|
| 944 |
"language_model.model.layers.33.self_attn.o_proj": {
|
| 945 |
"group_size": 64,
|
| 946 |
-
"bits":
|
| 947 |
},
|
| 948 |
"language_model.model.layers.33.mlp.gate_proj": {
|
| 949 |
"group_size": 64,
|
|
@@ -959,19 +959,19 @@
|
|
| 959 |
},
|
| 960 |
"language_model.model.layers.34.self_attn.q_proj": {
|
| 961 |
"group_size": 64,
|
| 962 |
-
"bits":
|
| 963 |
},
|
| 964 |
"language_model.model.layers.34.self_attn.k_proj": {
|
| 965 |
"group_size": 64,
|
| 966 |
-
"bits":
|
| 967 |
},
|
| 968 |
"language_model.model.layers.34.self_attn.v_proj": {
|
| 969 |
"group_size": 64,
|
| 970 |
-
"bits":
|
| 971 |
},
|
| 972 |
"language_model.model.layers.34.self_attn.o_proj": {
|
| 973 |
"group_size": 64,
|
| 974 |
-
"bits":
|
| 975 |
},
|
| 976 |
"language_model.model.layers.34.mlp.gate_proj": {
|
| 977 |
"group_size": 64,
|
|
@@ -987,15 +987,15 @@
|
|
| 987 |
},
|
| 988 |
"language_model.model.layers.35.self_attn.q_proj": {
|
| 989 |
"group_size": 64,
|
| 990 |
-
"bits":
|
| 991 |
},
|
| 992 |
"language_model.model.layers.35.self_attn.k_proj": {
|
| 993 |
"group_size": 64,
|
| 994 |
-
"bits":
|
| 995 |
},
|
| 996 |
"language_model.model.layers.35.self_attn.o_proj": {
|
| 997 |
"group_size": 64,
|
| 998 |
-
"bits":
|
| 999 |
},
|
| 1000 |
"language_model.model.layers.35.mlp.gate_proj": {
|
| 1001 |
"group_size": 64,
|
|
@@ -1011,19 +1011,19 @@
|
|
| 1011 |
},
|
| 1012 |
"language_model.model.layers.36.self_attn.q_proj": {
|
| 1013 |
"group_size": 64,
|
| 1014 |
-
"bits":
|
| 1015 |
},
|
| 1016 |
"language_model.model.layers.36.self_attn.k_proj": {
|
| 1017 |
"group_size": 64,
|
| 1018 |
-
"bits":
|
| 1019 |
},
|
| 1020 |
"language_model.model.layers.36.self_attn.v_proj": {
|
| 1021 |
"group_size": 64,
|
| 1022 |
-
"bits":
|
| 1023 |
},
|
| 1024 |
"language_model.model.layers.36.self_attn.o_proj": {
|
| 1025 |
"group_size": 64,
|
| 1026 |
-
"bits":
|
| 1027 |
},
|
| 1028 |
"language_model.model.layers.36.mlp.gate_proj": {
|
| 1029 |
"group_size": 64,
|
|
@@ -1039,19 +1039,19 @@
|
|
| 1039 |
},
|
| 1040 |
"language_model.model.layers.37.self_attn.q_proj": {
|
| 1041 |
"group_size": 64,
|
| 1042 |
-
"bits":
|
| 1043 |
},
|
| 1044 |
"language_model.model.layers.37.self_attn.k_proj": {
|
| 1045 |
"group_size": 64,
|
| 1046 |
-
"bits":
|
| 1047 |
},
|
| 1048 |
"language_model.model.layers.37.self_attn.v_proj": {
|
| 1049 |
"group_size": 64,
|
| 1050 |
-
"bits":
|
| 1051 |
},
|
| 1052 |
"language_model.model.layers.37.self_attn.o_proj": {
|
| 1053 |
"group_size": 64,
|
| 1054 |
-
"bits":
|
| 1055 |
},
|
| 1056 |
"language_model.model.layers.37.mlp.gate_proj": {
|
| 1057 |
"group_size": 64,
|
|
@@ -1067,19 +1067,19 @@
|
|
| 1067 |
},
|
| 1068 |
"language_model.model.layers.38.self_attn.q_proj": {
|
| 1069 |
"group_size": 64,
|
| 1070 |
-
"bits":
|
| 1071 |
},
|
| 1072 |
"language_model.model.layers.38.self_attn.k_proj": {
|
| 1073 |
"group_size": 64,
|
| 1074 |
-
"bits":
|
| 1075 |
},
|
| 1076 |
"language_model.model.layers.38.self_attn.v_proj": {
|
| 1077 |
"group_size": 64,
|
| 1078 |
-
"bits":
|
| 1079 |
},
|
| 1080 |
"language_model.model.layers.38.self_attn.o_proj": {
|
| 1081 |
"group_size": 64,
|
| 1082 |
-
"bits":
|
| 1083 |
},
|
| 1084 |
"language_model.model.layers.38.mlp.gate_proj": {
|
| 1085 |
"group_size": 64,
|
|
@@ -1095,19 +1095,19 @@
|
|
| 1095 |
},
|
| 1096 |
"language_model.model.layers.39.self_attn.q_proj": {
|
| 1097 |
"group_size": 64,
|
| 1098 |
-
"bits":
|
| 1099 |
},
|
| 1100 |
"language_model.model.layers.39.self_attn.k_proj": {
|
| 1101 |
"group_size": 64,
|
| 1102 |
-
"bits":
|
| 1103 |
},
|
| 1104 |
"language_model.model.layers.39.self_attn.v_proj": {
|
| 1105 |
"group_size": 64,
|
| 1106 |
-
"bits":
|
| 1107 |
},
|
| 1108 |
"language_model.model.layers.39.self_attn.o_proj": {
|
| 1109 |
"group_size": 64,
|
| 1110 |
-
"bits":
|
| 1111 |
},
|
| 1112 |
"language_model.model.layers.39.mlp.gate_proj": {
|
| 1113 |
"group_size": 64,
|
|
@@ -1123,19 +1123,19 @@
|
|
| 1123 |
},
|
| 1124 |
"language_model.model.layers.40.self_attn.q_proj": {
|
| 1125 |
"group_size": 64,
|
| 1126 |
-
"bits":
|
| 1127 |
},
|
| 1128 |
"language_model.model.layers.40.self_attn.k_proj": {
|
| 1129 |
"group_size": 64,
|
| 1130 |
-
"bits":
|
| 1131 |
},
|
| 1132 |
"language_model.model.layers.40.self_attn.v_proj": {
|
| 1133 |
"group_size": 64,
|
| 1134 |
-
"bits":
|
| 1135 |
},
|
| 1136 |
"language_model.model.layers.40.self_attn.o_proj": {
|
| 1137 |
"group_size": 64,
|
| 1138 |
-
"bits":
|
| 1139 |
},
|
| 1140 |
"language_model.model.layers.40.mlp.gate_proj": {
|
| 1141 |
"group_size": 64,
|
|
@@ -1151,15 +1151,15 @@
|
|
| 1151 |
},
|
| 1152 |
"language_model.model.layers.41.self_attn.q_proj": {
|
| 1153 |
"group_size": 64,
|
| 1154 |
-
"bits":
|
| 1155 |
},
|
| 1156 |
"language_model.model.layers.41.self_attn.k_proj": {
|
| 1157 |
"group_size": 64,
|
| 1158 |
-
"bits":
|
| 1159 |
},
|
| 1160 |
"language_model.model.layers.41.self_attn.o_proj": {
|
| 1161 |
"group_size": 64,
|
| 1162 |
-
"bits":
|
| 1163 |
},
|
| 1164 |
"language_model.model.layers.41.mlp.gate_proj": {
|
| 1165 |
"group_size": 64,
|
|
@@ -1175,19 +1175,19 @@
|
|
| 1175 |
},
|
| 1176 |
"language_model.model.layers.42.self_attn.q_proj": {
|
| 1177 |
"group_size": 64,
|
| 1178 |
-
"bits":
|
| 1179 |
},
|
| 1180 |
"language_model.model.layers.42.self_attn.k_proj": {
|
| 1181 |
"group_size": 64,
|
| 1182 |
-
"bits":
|
| 1183 |
},
|
| 1184 |
"language_model.model.layers.42.self_attn.v_proj": {
|
| 1185 |
"group_size": 64,
|
| 1186 |
-
"bits":
|
| 1187 |
},
|
| 1188 |
"language_model.model.layers.42.self_attn.o_proj": {
|
| 1189 |
"group_size": 64,
|
| 1190 |
-
"bits":
|
| 1191 |
},
|
| 1192 |
"language_model.model.layers.42.mlp.gate_proj": {
|
| 1193 |
"group_size": 64,
|
|
@@ -1203,19 +1203,19 @@
|
|
| 1203 |
},
|
| 1204 |
"language_model.model.layers.43.self_attn.q_proj": {
|
| 1205 |
"group_size": 64,
|
| 1206 |
-
"bits":
|
| 1207 |
},
|
| 1208 |
"language_model.model.layers.43.self_attn.k_proj": {
|
| 1209 |
"group_size": 64,
|
| 1210 |
-
"bits":
|
| 1211 |
},
|
| 1212 |
"language_model.model.layers.43.self_attn.v_proj": {
|
| 1213 |
"group_size": 64,
|
| 1214 |
-
"bits":
|
| 1215 |
},
|
| 1216 |
"language_model.model.layers.43.self_attn.o_proj": {
|
| 1217 |
"group_size": 64,
|
| 1218 |
-
"bits":
|
| 1219 |
},
|
| 1220 |
"language_model.model.layers.43.mlp.gate_proj": {
|
| 1221 |
"group_size": 64,
|
|
@@ -1231,19 +1231,19 @@
|
|
| 1231 |
},
|
| 1232 |
"language_model.model.layers.44.self_attn.q_proj": {
|
| 1233 |
"group_size": 64,
|
| 1234 |
-
"bits":
|
| 1235 |
},
|
| 1236 |
"language_model.model.layers.44.self_attn.k_proj": {
|
| 1237 |
"group_size": 64,
|
| 1238 |
-
"bits":
|
| 1239 |
},
|
| 1240 |
"language_model.model.layers.44.self_attn.v_proj": {
|
| 1241 |
"group_size": 64,
|
| 1242 |
-
"bits":
|
| 1243 |
},
|
| 1244 |
"language_model.model.layers.44.self_attn.o_proj": {
|
| 1245 |
"group_size": 64,
|
| 1246 |
-
"bits":
|
| 1247 |
},
|
| 1248 |
"language_model.model.layers.44.mlp.gate_proj": {
|
| 1249 |
"group_size": 64,
|
|
@@ -1259,19 +1259,19 @@
|
|
| 1259 |
},
|
| 1260 |
"language_model.model.layers.45.self_attn.q_proj": {
|
| 1261 |
"group_size": 64,
|
| 1262 |
-
"bits":
|
| 1263 |
},
|
| 1264 |
"language_model.model.layers.45.self_attn.k_proj": {
|
| 1265 |
"group_size": 64,
|
| 1266 |
-
"bits":
|
| 1267 |
},
|
| 1268 |
"language_model.model.layers.45.self_attn.v_proj": {
|
| 1269 |
"group_size": 64,
|
| 1270 |
-
"bits":
|
| 1271 |
},
|
| 1272 |
"language_model.model.layers.45.self_attn.o_proj": {
|
| 1273 |
"group_size": 64,
|
| 1274 |
-
"bits":
|
| 1275 |
},
|
| 1276 |
"language_model.model.layers.45.mlp.gate_proj": {
|
| 1277 |
"group_size": 64,
|
|
@@ -1287,19 +1287,19 @@
|
|
| 1287 |
},
|
| 1288 |
"language_model.model.layers.46.self_attn.q_proj": {
|
| 1289 |
"group_size": 64,
|
| 1290 |
-
"bits":
|
| 1291 |
},
|
| 1292 |
"language_model.model.layers.46.self_attn.k_proj": {
|
| 1293 |
"group_size": 64,
|
| 1294 |
-
"bits":
|
| 1295 |
},
|
| 1296 |
"language_model.model.layers.46.self_attn.v_proj": {
|
| 1297 |
"group_size": 64,
|
| 1298 |
-
"bits":
|
| 1299 |
},
|
| 1300 |
"language_model.model.layers.46.self_attn.o_proj": {
|
| 1301 |
"group_size": 64,
|
| 1302 |
-
"bits":
|
| 1303 |
},
|
| 1304 |
"language_model.model.layers.46.mlp.gate_proj": {
|
| 1305 |
"group_size": 64,
|
|
@@ -1315,15 +1315,15 @@
|
|
| 1315 |
},
|
| 1316 |
"language_model.model.layers.47.self_attn.q_proj": {
|
| 1317 |
"group_size": 64,
|
| 1318 |
-
"bits":
|
| 1319 |
},
|
| 1320 |
"language_model.model.layers.47.self_attn.k_proj": {
|
| 1321 |
"group_size": 64,
|
| 1322 |
-
"bits":
|
| 1323 |
},
|
| 1324 |
"language_model.model.layers.47.self_attn.o_proj": {
|
| 1325 |
"group_size": 64,
|
| 1326 |
-
"bits":
|
| 1327 |
},
|
| 1328 |
"language_model.model.layers.47.mlp.gate_proj": {
|
| 1329 |
"group_size": 64,
|
|
@@ -1339,19 +1339,19 @@
|
|
| 1339 |
},
|
| 1340 |
"language_model.model.layers.48.self_attn.q_proj": {
|
| 1341 |
"group_size": 64,
|
| 1342 |
-
"bits":
|
| 1343 |
},
|
| 1344 |
"language_model.model.layers.48.self_attn.k_proj": {
|
| 1345 |
"group_size": 64,
|
| 1346 |
-
"bits":
|
| 1347 |
},
|
| 1348 |
"language_model.model.layers.48.self_attn.v_proj": {
|
| 1349 |
"group_size": 64,
|
| 1350 |
-
"bits":
|
| 1351 |
},
|
| 1352 |
"language_model.model.layers.48.self_attn.o_proj": {
|
| 1353 |
"group_size": 64,
|
| 1354 |
-
"bits":
|
| 1355 |
},
|
| 1356 |
"language_model.model.layers.48.mlp.gate_proj": {
|
| 1357 |
"group_size": 64,
|
|
@@ -1367,19 +1367,19 @@
|
|
| 1367 |
},
|
| 1368 |
"language_model.model.layers.49.self_attn.q_proj": {
|
| 1369 |
"group_size": 64,
|
| 1370 |
-
"bits":
|
| 1371 |
},
|
| 1372 |
"language_model.model.layers.49.self_attn.k_proj": {
|
| 1373 |
"group_size": 64,
|
| 1374 |
-
"bits":
|
| 1375 |
},
|
| 1376 |
"language_model.model.layers.49.self_attn.v_proj": {
|
| 1377 |
"group_size": 64,
|
| 1378 |
-
"bits":
|
| 1379 |
},
|
| 1380 |
"language_model.model.layers.49.self_attn.o_proj": {
|
| 1381 |
"group_size": 64,
|
| 1382 |
-
"bits":
|
| 1383 |
},
|
| 1384 |
"language_model.model.layers.49.mlp.gate_proj": {
|
| 1385 |
"group_size": 64,
|
|
@@ -1395,19 +1395,19 @@
|
|
| 1395 |
},
|
| 1396 |
"language_model.model.layers.50.self_attn.q_proj": {
|
| 1397 |
"group_size": 64,
|
| 1398 |
-
"bits":
|
| 1399 |
},
|
| 1400 |
"language_model.model.layers.50.self_attn.k_proj": {
|
| 1401 |
"group_size": 64,
|
| 1402 |
-
"bits":
|
| 1403 |
},
|
| 1404 |
"language_model.model.layers.50.self_attn.v_proj": {
|
| 1405 |
"group_size": 64,
|
| 1406 |
-
"bits":
|
| 1407 |
},
|
| 1408 |
"language_model.model.layers.50.self_attn.o_proj": {
|
| 1409 |
"group_size": 64,
|
| 1410 |
-
"bits":
|
| 1411 |
},
|
| 1412 |
"language_model.model.layers.50.mlp.gate_proj": {
|
| 1413 |
"group_size": 64,
|
|
@@ -1423,19 +1423,19 @@
|
|
| 1423 |
},
|
| 1424 |
"language_model.model.layers.51.self_attn.q_proj": {
|
| 1425 |
"group_size": 64,
|
| 1426 |
-
"bits":
|
| 1427 |
},
|
| 1428 |
"language_model.model.layers.51.self_attn.k_proj": {
|
| 1429 |
"group_size": 64,
|
| 1430 |
-
"bits":
|
| 1431 |
},
|
| 1432 |
"language_model.model.layers.51.self_attn.v_proj": {
|
| 1433 |
"group_size": 64,
|
| 1434 |
-
"bits":
|
| 1435 |
},
|
| 1436 |
"language_model.model.layers.51.self_attn.o_proj": {
|
| 1437 |
"group_size": 64,
|
| 1438 |
-
"bits":
|
| 1439 |
},
|
| 1440 |
"language_model.model.layers.51.mlp.gate_proj": {
|
| 1441 |
"group_size": 64,
|
|
@@ -1451,43 +1451,43 @@
|
|
| 1451 |
},
|
| 1452 |
"language_model.model.layers.52.self_attn.q_proj": {
|
| 1453 |
"group_size": 64,
|
| 1454 |
-
"bits":
|
| 1455 |
},
|
| 1456 |
"language_model.model.layers.52.self_attn.k_proj": {
|
| 1457 |
"group_size": 64,
|
| 1458 |
-
"bits":
|
| 1459 |
},
|
| 1460 |
"language_model.model.layers.52.self_attn.v_proj": {
|
| 1461 |
"group_size": 64,
|
| 1462 |
-
"bits":
|
| 1463 |
},
|
| 1464 |
"language_model.model.layers.52.self_attn.o_proj": {
|
| 1465 |
"group_size": 64,
|
| 1466 |
-
"bits":
|
| 1467 |
},
|
| 1468 |
"language_model.model.layers.52.mlp.gate_proj": {
|
| 1469 |
"group_size": 64,
|
| 1470 |
-
"bits":
|
| 1471 |
},
|
| 1472 |
"language_model.model.layers.52.mlp.up_proj": {
|
| 1473 |
"group_size": 64,
|
| 1474 |
-
"bits":
|
| 1475 |
},
|
| 1476 |
"language_model.model.layers.52.mlp.down_proj": {
|
| 1477 |
"group_size": 64,
|
| 1478 |
-
"bits":
|
| 1479 |
},
|
| 1480 |
"language_model.model.layers.53.self_attn.q_proj": {
|
| 1481 |
"group_size": 64,
|
| 1482 |
-
"bits":
|
| 1483 |
},
|
| 1484 |
"language_model.model.layers.53.self_attn.k_proj": {
|
| 1485 |
"group_size": 64,
|
| 1486 |
-
"bits":
|
| 1487 |
},
|
| 1488 |
"language_model.model.layers.53.self_attn.o_proj": {
|
| 1489 |
"group_size": 64,
|
| 1490 |
-
"bits":
|
| 1491 |
},
|
| 1492 |
"language_model.model.layers.53.mlp.gate_proj": {
|
| 1493 |
"group_size": 64,
|
|
@@ -1503,19 +1503,19 @@
|
|
| 1503 |
},
|
| 1504 |
"language_model.model.layers.54.self_attn.q_proj": {
|
| 1505 |
"group_size": 64,
|
| 1506 |
-
"bits":
|
| 1507 |
},
|
| 1508 |
"language_model.model.layers.54.self_attn.k_proj": {
|
| 1509 |
"group_size": 64,
|
| 1510 |
-
"bits":
|
| 1511 |
},
|
| 1512 |
"language_model.model.layers.54.self_attn.v_proj": {
|
| 1513 |
"group_size": 64,
|
| 1514 |
-
"bits":
|
| 1515 |
},
|
| 1516 |
"language_model.model.layers.54.self_attn.o_proj": {
|
| 1517 |
"group_size": 64,
|
| 1518 |
-
"bits":
|
| 1519 |
},
|
| 1520 |
"language_model.model.layers.54.mlp.gate_proj": {
|
| 1521 |
"group_size": 64,
|
|
@@ -1531,19 +1531,19 @@
|
|
| 1531 |
},
|
| 1532 |
"language_model.model.layers.55.self_attn.q_proj": {
|
| 1533 |
"group_size": 64,
|
| 1534 |
-
"bits":
|
| 1535 |
},
|
| 1536 |
"language_model.model.layers.55.self_attn.k_proj": {
|
| 1537 |
"group_size": 64,
|
| 1538 |
-
"bits":
|
| 1539 |
},
|
| 1540 |
"language_model.model.layers.55.self_attn.v_proj": {
|
| 1541 |
"group_size": 64,
|
| 1542 |
-
"bits":
|
| 1543 |
},
|
| 1544 |
"language_model.model.layers.55.self_attn.o_proj": {
|
| 1545 |
"group_size": 64,
|
| 1546 |
-
"bits":
|
| 1547 |
},
|
| 1548 |
"language_model.model.layers.55.mlp.gate_proj": {
|
| 1549 |
"group_size": 64,
|
|
@@ -1559,19 +1559,19 @@
|
|
| 1559 |
},
|
| 1560 |
"language_model.model.layers.56.self_attn.q_proj": {
|
| 1561 |
"group_size": 64,
|
| 1562 |
-
"bits":
|
| 1563 |
},
|
| 1564 |
"language_model.model.layers.56.self_attn.k_proj": {
|
| 1565 |
"group_size": 64,
|
| 1566 |
-
"bits":
|
| 1567 |
},
|
| 1568 |
"language_model.model.layers.56.self_attn.v_proj": {
|
| 1569 |
"group_size": 64,
|
| 1570 |
-
"bits":
|
| 1571 |
},
|
| 1572 |
"language_model.model.layers.56.self_attn.o_proj": {
|
| 1573 |
"group_size": 64,
|
| 1574 |
-
"bits":
|
| 1575 |
},
|
| 1576 |
"language_model.model.layers.56.mlp.gate_proj": {
|
| 1577 |
"group_size": 64,
|
|
@@ -1587,83 +1587,83 @@
|
|
| 1587 |
},
|
| 1588 |
"language_model.model.layers.57.self_attn.q_proj": {
|
| 1589 |
"group_size": 64,
|
| 1590 |
-
"bits":
|
| 1591 |
},
|
| 1592 |
"language_model.model.layers.57.self_attn.k_proj": {
|
| 1593 |
"group_size": 64,
|
| 1594 |
-
"bits":
|
| 1595 |
},
|
| 1596 |
"language_model.model.layers.57.self_attn.v_proj": {
|
| 1597 |
"group_size": 64,
|
| 1598 |
-
"bits":
|
| 1599 |
},
|
| 1600 |
"language_model.model.layers.57.self_attn.o_proj": {
|
| 1601 |
"group_size": 64,
|
| 1602 |
-
"bits":
|
| 1603 |
},
|
| 1604 |
"language_model.model.layers.57.mlp.gate_proj": {
|
| 1605 |
"group_size": 64,
|
| 1606 |
-
"bits":
|
| 1607 |
},
|
| 1608 |
"language_model.model.layers.57.mlp.up_proj": {
|
| 1609 |
"group_size": 64,
|
| 1610 |
-
"bits":
|
| 1611 |
},
|
| 1612 |
"language_model.model.layers.57.mlp.down_proj": {
|
| 1613 |
"group_size": 64,
|
| 1614 |
-
"bits":
|
| 1615 |
},
|
| 1616 |
"language_model.model.layers.58.self_attn.q_proj": {
|
| 1617 |
"group_size": 64,
|
| 1618 |
-
"bits":
|
| 1619 |
},
|
| 1620 |
"language_model.model.layers.58.self_attn.k_proj": {
|
| 1621 |
"group_size": 64,
|
| 1622 |
-
"bits":
|
| 1623 |
},
|
| 1624 |
"language_model.model.layers.58.self_attn.v_proj": {
|
| 1625 |
"group_size": 64,
|
| 1626 |
-
"bits":
|
| 1627 |
},
|
| 1628 |
"language_model.model.layers.58.self_attn.o_proj": {
|
| 1629 |
"group_size": 64,
|
| 1630 |
-
"bits":
|
| 1631 |
},
|
| 1632 |
"language_model.model.layers.58.mlp.gate_proj": {
|
| 1633 |
"group_size": 64,
|
| 1634 |
-
"bits":
|
| 1635 |
},
|
| 1636 |
"language_model.model.layers.58.mlp.up_proj": {
|
| 1637 |
"group_size": 64,
|
| 1638 |
-
"bits":
|
| 1639 |
},
|
| 1640 |
"language_model.model.layers.58.mlp.down_proj": {
|
| 1641 |
"group_size": 64,
|
| 1642 |
-
"bits":
|
| 1643 |
},
|
| 1644 |
"language_model.model.layers.59.self_attn.q_proj": {
|
| 1645 |
"group_size": 64,
|
| 1646 |
-
"bits":
|
| 1647 |
},
|
| 1648 |
"language_model.model.layers.59.self_attn.k_proj": {
|
| 1649 |
"group_size": 64,
|
| 1650 |
-
"bits":
|
| 1651 |
},
|
| 1652 |
"language_model.model.layers.59.self_attn.o_proj": {
|
| 1653 |
"group_size": 64,
|
| 1654 |
-
"bits":
|
| 1655 |
},
|
| 1656 |
"language_model.model.layers.59.mlp.gate_proj": {
|
| 1657 |
"group_size": 64,
|
| 1658 |
-
"bits":
|
| 1659 |
},
|
| 1660 |
"language_model.model.layers.59.mlp.up_proj": {
|
| 1661 |
"group_size": 64,
|
| 1662 |
-
"bits":
|
| 1663 |
},
|
| 1664 |
"language_model.model.layers.59.mlp.down_proj": {
|
| 1665 |
"group_size": 64,
|
| 1666 |
-
"bits":
|
| 1667 |
}
|
| 1668 |
},
|
| 1669 |
"quantization_config": {
|
|
@@ -1671,23 +1671,23 @@
|
|
| 1671 |
"bits": 4,
|
| 1672 |
"language_model.model.embed_tokens": {
|
| 1673 |
"group_size": 64,
|
| 1674 |
-
"bits":
|
| 1675 |
},
|
| 1676 |
"language_model.model.layers.0.self_attn.q_proj": {
|
| 1677 |
"group_size": 64,
|
| 1678 |
-
"bits":
|
| 1679 |
},
|
| 1680 |
"language_model.model.layers.0.self_attn.k_proj": {
|
| 1681 |
"group_size": 64,
|
| 1682 |
-
"bits":
|
| 1683 |
},
|
| 1684 |
"language_model.model.layers.0.self_attn.v_proj": {
|
| 1685 |
"group_size": 64,
|
| 1686 |
-
"bits":
|
| 1687 |
},
|
| 1688 |
"language_model.model.layers.0.self_attn.o_proj": {
|
| 1689 |
"group_size": 64,
|
| 1690 |
-
"bits":
|
| 1691 |
},
|
| 1692 |
"language_model.model.layers.0.mlp.gate_proj": {
|
| 1693 |
"group_size": 64,
|
|
@@ -1703,47 +1703,47 @@
|
|
| 1703 |
},
|
| 1704 |
"language_model.model.layers.1.self_attn.q_proj": {
|
| 1705 |
"group_size": 64,
|
| 1706 |
-
"bits":
|
| 1707 |
},
|
| 1708 |
"language_model.model.layers.1.self_attn.k_proj": {
|
| 1709 |
"group_size": 64,
|
| 1710 |
-
"bits":
|
| 1711 |
},
|
| 1712 |
"language_model.model.layers.1.self_attn.v_proj": {
|
| 1713 |
"group_size": 64,
|
| 1714 |
-
"bits":
|
| 1715 |
},
|
| 1716 |
"language_model.model.layers.1.self_attn.o_proj": {
|
| 1717 |
"group_size": 64,
|
| 1718 |
-
"bits":
|
| 1719 |
},
|
| 1720 |
"language_model.model.layers.1.mlp.gate_proj": {
|
| 1721 |
"group_size": 64,
|
| 1722 |
-
"bits":
|
| 1723 |
},
|
| 1724 |
"language_model.model.layers.1.mlp.up_proj": {
|
| 1725 |
"group_size": 64,
|
| 1726 |
-
"bits":
|
| 1727 |
},
|
| 1728 |
"language_model.model.layers.1.mlp.down_proj": {
|
| 1729 |
"group_size": 64,
|
| 1730 |
-
"bits":
|
| 1731 |
},
|
| 1732 |
"language_model.model.layers.2.self_attn.q_proj": {
|
| 1733 |
"group_size": 64,
|
| 1734 |
-
"bits":
|
| 1735 |
},
|
| 1736 |
"language_model.model.layers.2.self_attn.k_proj": {
|
| 1737 |
"group_size": 64,
|
| 1738 |
-
"bits":
|
| 1739 |
},
|
| 1740 |
"language_model.model.layers.2.self_attn.v_proj": {
|
| 1741 |
"group_size": 64,
|
| 1742 |
-
"bits":
|
| 1743 |
},
|
| 1744 |
"language_model.model.layers.2.self_attn.o_proj": {
|
| 1745 |
"group_size": 64,
|
| 1746 |
-
"bits":
|
| 1747 |
},
|
| 1748 |
"language_model.model.layers.2.mlp.gate_proj": {
|
| 1749 |
"group_size": 64,
|
|
@@ -1759,19 +1759,19 @@
|
|
| 1759 |
},
|
| 1760 |
"language_model.model.layers.3.self_attn.q_proj": {
|
| 1761 |
"group_size": 64,
|
| 1762 |
-
"bits":
|
| 1763 |
},
|
| 1764 |
"language_model.model.layers.3.self_attn.k_proj": {
|
| 1765 |
"group_size": 64,
|
| 1766 |
-
"bits":
|
| 1767 |
},
|
| 1768 |
"language_model.model.layers.3.self_attn.v_proj": {
|
| 1769 |
"group_size": 64,
|
| 1770 |
-
"bits":
|
| 1771 |
},
|
| 1772 |
"language_model.model.layers.3.self_attn.o_proj": {
|
| 1773 |
"group_size": 64,
|
| 1774 |
-
"bits":
|
| 1775 |
},
|
| 1776 |
"language_model.model.layers.3.mlp.gate_proj": {
|
| 1777 |
"group_size": 64,
|
|
@@ -1787,19 +1787,19 @@
|
|
| 1787 |
},
|
| 1788 |
"language_model.model.layers.4.self_attn.q_proj": {
|
| 1789 |
"group_size": 64,
|
| 1790 |
-
"bits":
|
| 1791 |
},
|
| 1792 |
"language_model.model.layers.4.self_attn.k_proj": {
|
| 1793 |
"group_size": 64,
|
| 1794 |
-
"bits":
|
| 1795 |
},
|
| 1796 |
"language_model.model.layers.4.self_attn.v_proj": {
|
| 1797 |
"group_size": 64,
|
| 1798 |
-
"bits":
|
| 1799 |
},
|
| 1800 |
"language_model.model.layers.4.self_attn.o_proj": {
|
| 1801 |
"group_size": 64,
|
| 1802 |
-
"bits":
|
| 1803 |
},
|
| 1804 |
"language_model.model.layers.4.mlp.gate_proj": {
|
| 1805 |
"group_size": 64,
|
|
@@ -1815,15 +1815,15 @@
|
|
| 1815 |
},
|
| 1816 |
"language_model.model.layers.5.self_attn.q_proj": {
|
| 1817 |
"group_size": 64,
|
| 1818 |
-
"bits":
|
| 1819 |
},
|
| 1820 |
"language_model.model.layers.5.self_attn.k_proj": {
|
| 1821 |
"group_size": 64,
|
| 1822 |
-
"bits":
|
| 1823 |
},
|
| 1824 |
"language_model.model.layers.5.self_attn.o_proj": {
|
| 1825 |
"group_size": 64,
|
| 1826 |
-
"bits":
|
| 1827 |
},
|
| 1828 |
"language_model.model.layers.5.mlp.gate_proj": {
|
| 1829 |
"group_size": 64,
|
|
@@ -1839,19 +1839,19 @@
|
|
| 1839 |
},
|
| 1840 |
"language_model.model.layers.6.self_attn.q_proj": {
|
| 1841 |
"group_size": 64,
|
| 1842 |
-
"bits":
|
| 1843 |
},
|
| 1844 |
"language_model.model.layers.6.self_attn.k_proj": {
|
| 1845 |
"group_size": 64,
|
| 1846 |
-
"bits":
|
| 1847 |
},
|
| 1848 |
"language_model.model.layers.6.self_attn.v_proj": {
|
| 1849 |
"group_size": 64,
|
| 1850 |
-
"bits":
|
| 1851 |
},
|
| 1852 |
"language_model.model.layers.6.self_attn.o_proj": {
|
| 1853 |
"group_size": 64,
|
| 1854 |
-
"bits":
|
| 1855 |
},
|
| 1856 |
"language_model.model.layers.6.mlp.gate_proj": {
|
| 1857 |
"group_size": 64,
|
|
@@ -1867,19 +1867,19 @@
|
|
| 1867 |
},
|
| 1868 |
"language_model.model.layers.7.self_attn.q_proj": {
|
| 1869 |
"group_size": 64,
|
| 1870 |
-
"bits":
|
| 1871 |
},
|
| 1872 |
"language_model.model.layers.7.self_attn.k_proj": {
|
| 1873 |
"group_size": 64,
|
| 1874 |
-
"bits":
|
| 1875 |
},
|
| 1876 |
"language_model.model.layers.7.self_attn.v_proj": {
|
| 1877 |
"group_size": 64,
|
| 1878 |
-
"bits":
|
| 1879 |
},
|
| 1880 |
"language_model.model.layers.7.self_attn.o_proj": {
|
| 1881 |
"group_size": 64,
|
| 1882 |
-
"bits":
|
| 1883 |
},
|
| 1884 |
"language_model.model.layers.7.mlp.gate_proj": {
|
| 1885 |
"group_size": 64,
|
|
@@ -1895,19 +1895,19 @@
|
|
| 1895 |
},
|
| 1896 |
"language_model.model.layers.8.self_attn.q_proj": {
|
| 1897 |
"group_size": 64,
|
| 1898 |
-
"bits":
|
| 1899 |
},
|
| 1900 |
"language_model.model.layers.8.self_attn.k_proj": {
|
| 1901 |
"group_size": 64,
|
| 1902 |
-
"bits":
|
| 1903 |
},
|
| 1904 |
"language_model.model.layers.8.self_attn.v_proj": {
|
| 1905 |
"group_size": 64,
|
| 1906 |
-
"bits":
|
| 1907 |
},
|
| 1908 |
"language_model.model.layers.8.self_attn.o_proj": {
|
| 1909 |
"group_size": 64,
|
| 1910 |
-
"bits":
|
| 1911 |
},
|
| 1912 |
"language_model.model.layers.8.mlp.gate_proj": {
|
| 1913 |
"group_size": 64,
|
|
@@ -1923,19 +1923,19 @@
|
|
| 1923 |
},
|
| 1924 |
"language_model.model.layers.9.self_attn.q_proj": {
|
| 1925 |
"group_size": 64,
|
| 1926 |
-
"bits":
|
| 1927 |
},
|
| 1928 |
"language_model.model.layers.9.self_attn.k_proj": {
|
| 1929 |
"group_size": 64,
|
| 1930 |
-
"bits":
|
| 1931 |
},
|
| 1932 |
"language_model.model.layers.9.self_attn.v_proj": {
|
| 1933 |
"group_size": 64,
|
| 1934 |
-
"bits":
|
| 1935 |
},
|
| 1936 |
"language_model.model.layers.9.self_attn.o_proj": {
|
| 1937 |
"group_size": 64,
|
| 1938 |
-
"bits":
|
| 1939 |
},
|
| 1940 |
"language_model.model.layers.9.mlp.gate_proj": {
|
| 1941 |
"group_size": 64,
|
|
@@ -1951,19 +1951,19 @@
|
|
| 1951 |
},
|
| 1952 |
"language_model.model.layers.10.self_attn.q_proj": {
|
| 1953 |
"group_size": 64,
|
| 1954 |
-
"bits":
|
| 1955 |
},
|
| 1956 |
"language_model.model.layers.10.self_attn.k_proj": {
|
| 1957 |
"group_size": 64,
|
| 1958 |
-
"bits":
|
| 1959 |
},
|
| 1960 |
"language_model.model.layers.10.self_attn.v_proj": {
|
| 1961 |
"group_size": 64,
|
| 1962 |
-
"bits":
|
| 1963 |
},
|
| 1964 |
"language_model.model.layers.10.self_attn.o_proj": {
|
| 1965 |
"group_size": 64,
|
| 1966 |
-
"bits":
|
| 1967 |
},
|
| 1968 |
"language_model.model.layers.10.mlp.gate_proj": {
|
| 1969 |
"group_size": 64,
|
|
@@ -1979,15 +1979,15 @@
|
|
| 1979 |
},
|
| 1980 |
"language_model.model.layers.11.self_attn.q_proj": {
|
| 1981 |
"group_size": 64,
|
| 1982 |
-
"bits":
|
| 1983 |
},
|
| 1984 |
"language_model.model.layers.11.self_attn.k_proj": {
|
| 1985 |
"group_size": 64,
|
| 1986 |
-
"bits":
|
| 1987 |
},
|
| 1988 |
"language_model.model.layers.11.self_attn.o_proj": {
|
| 1989 |
"group_size": 64,
|
| 1990 |
-
"bits":
|
| 1991 |
},
|
| 1992 |
"language_model.model.layers.11.mlp.gate_proj": {
|
| 1993 |
"group_size": 64,
|
|
@@ -2003,19 +2003,19 @@
|
|
| 2003 |
},
|
| 2004 |
"language_model.model.layers.12.self_attn.q_proj": {
|
| 2005 |
"group_size": 64,
|
| 2006 |
-
"bits":
|
| 2007 |
},
|
| 2008 |
"language_model.model.layers.12.self_attn.k_proj": {
|
| 2009 |
"group_size": 64,
|
| 2010 |
-
"bits":
|
| 2011 |
},
|
| 2012 |
"language_model.model.layers.12.self_attn.v_proj": {
|
| 2013 |
"group_size": 64,
|
| 2014 |
-
"bits":
|
| 2015 |
},
|
| 2016 |
"language_model.model.layers.12.self_attn.o_proj": {
|
| 2017 |
"group_size": 64,
|
| 2018 |
-
"bits":
|
| 2019 |
},
|
| 2020 |
"language_model.model.layers.12.mlp.gate_proj": {
|
| 2021 |
"group_size": 64,
|
|
@@ -2031,19 +2031,19 @@
|
|
| 2031 |
},
|
| 2032 |
"language_model.model.layers.13.self_attn.q_proj": {
|
| 2033 |
"group_size": 64,
|
| 2034 |
-
"bits":
|
| 2035 |
},
|
| 2036 |
"language_model.model.layers.13.self_attn.k_proj": {
|
| 2037 |
"group_size": 64,
|
| 2038 |
-
"bits":
|
| 2039 |
},
|
| 2040 |
"language_model.model.layers.13.self_attn.v_proj": {
|
| 2041 |
"group_size": 64,
|
| 2042 |
-
"bits":
|
| 2043 |
},
|
| 2044 |
"language_model.model.layers.13.self_attn.o_proj": {
|
| 2045 |
"group_size": 64,
|
| 2046 |
-
"bits":
|
| 2047 |
},
|
| 2048 |
"language_model.model.layers.13.mlp.gate_proj": {
|
| 2049 |
"group_size": 64,
|
|
@@ -2059,19 +2059,19 @@
|
|
| 2059 |
},
|
| 2060 |
"language_model.model.layers.14.self_attn.q_proj": {
|
| 2061 |
"group_size": 64,
|
| 2062 |
-
"bits":
|
| 2063 |
},
|
| 2064 |
"language_model.model.layers.14.self_attn.k_proj": {
|
| 2065 |
"group_size": 64,
|
| 2066 |
-
"bits":
|
| 2067 |
},
|
| 2068 |
"language_model.model.layers.14.self_attn.v_proj": {
|
| 2069 |
"group_size": 64,
|
| 2070 |
-
"bits":
|
| 2071 |
},
|
| 2072 |
"language_model.model.layers.14.self_attn.o_proj": {
|
| 2073 |
"group_size": 64,
|
| 2074 |
-
"bits":
|
| 2075 |
},
|
| 2076 |
"language_model.model.layers.14.mlp.gate_proj": {
|
| 2077 |
"group_size": 64,
|
|
@@ -2087,19 +2087,19 @@
|
|
| 2087 |
},
|
| 2088 |
"language_model.model.layers.15.self_attn.q_proj": {
|
| 2089 |
"group_size": 64,
|
| 2090 |
-
"bits":
|
| 2091 |
},
|
| 2092 |
"language_model.model.layers.15.self_attn.k_proj": {
|
| 2093 |
"group_size": 64,
|
| 2094 |
-
"bits":
|
| 2095 |
},
|
| 2096 |
"language_model.model.layers.15.self_attn.v_proj": {
|
| 2097 |
"group_size": 64,
|
| 2098 |
-
"bits":
|
| 2099 |
},
|
| 2100 |
"language_model.model.layers.15.self_attn.o_proj": {
|
| 2101 |
"group_size": 64,
|
| 2102 |
-
"bits":
|
| 2103 |
},
|
| 2104 |
"language_model.model.layers.15.mlp.gate_proj": {
|
| 2105 |
"group_size": 64,
|
|
@@ -2115,19 +2115,19 @@
|
|
| 2115 |
},
|
| 2116 |
"language_model.model.layers.16.self_attn.q_proj": {
|
| 2117 |
"group_size": 64,
|
| 2118 |
-
"bits":
|
| 2119 |
},
|
| 2120 |
"language_model.model.layers.16.self_attn.k_proj": {
|
| 2121 |
"group_size": 64,
|
| 2122 |
-
"bits":
|
| 2123 |
},
|
| 2124 |
"language_model.model.layers.16.self_attn.v_proj": {
|
| 2125 |
"group_size": 64,
|
| 2126 |
-
"bits":
|
| 2127 |
},
|
| 2128 |
"language_model.model.layers.16.self_attn.o_proj": {
|
| 2129 |
"group_size": 64,
|
| 2130 |
-
"bits":
|
| 2131 |
},
|
| 2132 |
"language_model.model.layers.16.mlp.gate_proj": {
|
| 2133 |
"group_size": 64,
|
|
@@ -2143,15 +2143,15 @@
|
|
| 2143 |
},
|
| 2144 |
"language_model.model.layers.17.self_attn.q_proj": {
|
| 2145 |
"group_size": 64,
|
| 2146 |
-
"bits":
|
| 2147 |
},
|
| 2148 |
"language_model.model.layers.17.self_attn.k_proj": {
|
| 2149 |
"group_size": 64,
|
| 2150 |
-
"bits":
|
| 2151 |
},
|
| 2152 |
"language_model.model.layers.17.self_attn.o_proj": {
|
| 2153 |
"group_size": 64,
|
| 2154 |
-
"bits":
|
| 2155 |
},
|
| 2156 |
"language_model.model.layers.17.mlp.gate_proj": {
|
| 2157 |
"group_size": 64,
|
|
@@ -2167,19 +2167,19 @@
|
|
| 2167 |
},
|
| 2168 |
"language_model.model.layers.18.self_attn.q_proj": {
|
| 2169 |
"group_size": 64,
|
| 2170 |
-
"bits":
|
| 2171 |
},
|
| 2172 |
"language_model.model.layers.18.self_attn.k_proj": {
|
| 2173 |
"group_size": 64,
|
| 2174 |
-
"bits":
|
| 2175 |
},
|
| 2176 |
"language_model.model.layers.18.self_attn.v_proj": {
|
| 2177 |
"group_size": 64,
|
| 2178 |
-
"bits":
|
| 2179 |
},
|
| 2180 |
"language_model.model.layers.18.self_attn.o_proj": {
|
| 2181 |
"group_size": 64,
|
| 2182 |
-
"bits":
|
| 2183 |
},
|
| 2184 |
"language_model.model.layers.18.mlp.gate_proj": {
|
| 2185 |
"group_size": 64,
|
|
@@ -2195,19 +2195,19 @@
|
|
| 2195 |
},
|
| 2196 |
"language_model.model.layers.19.self_attn.q_proj": {
|
| 2197 |
"group_size": 64,
|
| 2198 |
-
"bits":
|
| 2199 |
},
|
| 2200 |
"language_model.model.layers.19.self_attn.k_proj": {
|
| 2201 |
"group_size": 64,
|
| 2202 |
-
"bits":
|
| 2203 |
},
|
| 2204 |
"language_model.model.layers.19.self_attn.v_proj": {
|
| 2205 |
"group_size": 64,
|
| 2206 |
-
"bits":
|
| 2207 |
},
|
| 2208 |
"language_model.model.layers.19.self_attn.o_proj": {
|
| 2209 |
"group_size": 64,
|
| 2210 |
-
"bits":
|
| 2211 |
},
|
| 2212 |
"language_model.model.layers.19.mlp.gate_proj": {
|
| 2213 |
"group_size": 64,
|
|
@@ -2223,19 +2223,19 @@
|
|
| 2223 |
},
|
| 2224 |
"language_model.model.layers.20.self_attn.q_proj": {
|
| 2225 |
"group_size": 64,
|
| 2226 |
-
"bits":
|
| 2227 |
},
|
| 2228 |
"language_model.model.layers.20.self_attn.k_proj": {
|
| 2229 |
"group_size": 64,
|
| 2230 |
-
"bits":
|
| 2231 |
},
|
| 2232 |
"language_model.model.layers.20.self_attn.v_proj": {
|
| 2233 |
"group_size": 64,
|
| 2234 |
-
"bits":
|
| 2235 |
},
|
| 2236 |
"language_model.model.layers.20.self_attn.o_proj": {
|
| 2237 |
"group_size": 64,
|
| 2238 |
-
"bits":
|
| 2239 |
},
|
| 2240 |
"language_model.model.layers.20.mlp.gate_proj": {
|
| 2241 |
"group_size": 64,
|
|
@@ -2251,19 +2251,19 @@
|
|
| 2251 |
},
|
| 2252 |
"language_model.model.layers.21.self_attn.q_proj": {
|
| 2253 |
"group_size": 64,
|
| 2254 |
-
"bits":
|
| 2255 |
},
|
| 2256 |
"language_model.model.layers.21.self_attn.k_proj": {
|
| 2257 |
"group_size": 64,
|
| 2258 |
-
"bits":
|
| 2259 |
},
|
| 2260 |
"language_model.model.layers.21.self_attn.v_proj": {
|
| 2261 |
"group_size": 64,
|
| 2262 |
-
"bits":
|
| 2263 |
},
|
| 2264 |
"language_model.model.layers.21.self_attn.o_proj": {
|
| 2265 |
"group_size": 64,
|
| 2266 |
-
"bits":
|
| 2267 |
},
|
| 2268 |
"language_model.model.layers.21.mlp.gate_proj": {
|
| 2269 |
"group_size": 64,
|
|
@@ -2279,19 +2279,19 @@
|
|
| 2279 |
},
|
| 2280 |
"language_model.model.layers.22.self_attn.q_proj": {
|
| 2281 |
"group_size": 64,
|
| 2282 |
-
"bits":
|
| 2283 |
},
|
| 2284 |
"language_model.model.layers.22.self_attn.k_proj": {
|
| 2285 |
"group_size": 64,
|
| 2286 |
-
"bits":
|
| 2287 |
},
|
| 2288 |
"language_model.model.layers.22.self_attn.v_proj": {
|
| 2289 |
"group_size": 64,
|
| 2290 |
-
"bits":
|
| 2291 |
},
|
| 2292 |
"language_model.model.layers.22.self_attn.o_proj": {
|
| 2293 |
"group_size": 64,
|
| 2294 |
-
"bits":
|
| 2295 |
},
|
| 2296 |
"language_model.model.layers.22.mlp.gate_proj": {
|
| 2297 |
"group_size": 64,
|
|
@@ -2307,15 +2307,15 @@
|
|
| 2307 |
},
|
| 2308 |
"language_model.model.layers.23.self_attn.q_proj": {
|
| 2309 |
"group_size": 64,
|
| 2310 |
-
"bits":
|
| 2311 |
},
|
| 2312 |
"language_model.model.layers.23.self_attn.k_proj": {
|
| 2313 |
"group_size": 64,
|
| 2314 |
-
"bits":
|
| 2315 |
},
|
| 2316 |
"language_model.model.layers.23.self_attn.o_proj": {
|
| 2317 |
"group_size": 64,
|
| 2318 |
-
"bits":
|
| 2319 |
},
|
| 2320 |
"language_model.model.layers.23.mlp.gate_proj": {
|
| 2321 |
"group_size": 64,
|
|
@@ -2331,19 +2331,19 @@
|
|
| 2331 |
},
|
| 2332 |
"language_model.model.layers.24.self_attn.q_proj": {
|
| 2333 |
"group_size": 64,
|
| 2334 |
-
"bits":
|
| 2335 |
},
|
| 2336 |
"language_model.model.layers.24.self_attn.k_proj": {
|
| 2337 |
"group_size": 64,
|
| 2338 |
-
"bits":
|
| 2339 |
},
|
| 2340 |
"language_model.model.layers.24.self_attn.v_proj": {
|
| 2341 |
"group_size": 64,
|
| 2342 |
-
"bits":
|
| 2343 |
},
|
| 2344 |
"language_model.model.layers.24.self_attn.o_proj": {
|
| 2345 |
"group_size": 64,
|
| 2346 |
-
"bits":
|
| 2347 |
},
|
| 2348 |
"language_model.model.layers.24.mlp.gate_proj": {
|
| 2349 |
"group_size": 64,
|
|
@@ -2359,19 +2359,19 @@
|
|
| 2359 |
},
|
| 2360 |
"language_model.model.layers.25.self_attn.q_proj": {
|
| 2361 |
"group_size": 64,
|
| 2362 |
-
"bits":
|
| 2363 |
},
|
| 2364 |
"language_model.model.layers.25.self_attn.k_proj": {
|
| 2365 |
"group_size": 64,
|
| 2366 |
-
"bits":
|
| 2367 |
},
|
| 2368 |
"language_model.model.layers.25.self_attn.v_proj": {
|
| 2369 |
"group_size": 64,
|
| 2370 |
-
"bits":
|
| 2371 |
},
|
| 2372 |
"language_model.model.layers.25.self_attn.o_proj": {
|
| 2373 |
"group_size": 64,
|
| 2374 |
-
"bits":
|
| 2375 |
},
|
| 2376 |
"language_model.model.layers.25.mlp.gate_proj": {
|
| 2377 |
"group_size": 64,
|
|
@@ -2387,19 +2387,19 @@
|
|
| 2387 |
},
|
| 2388 |
"language_model.model.layers.26.self_attn.q_proj": {
|
| 2389 |
"group_size": 64,
|
| 2390 |
-
"bits":
|
| 2391 |
},
|
| 2392 |
"language_model.model.layers.26.self_attn.k_proj": {
|
| 2393 |
"group_size": 64,
|
| 2394 |
-
"bits":
|
| 2395 |
},
|
| 2396 |
"language_model.model.layers.26.self_attn.v_proj": {
|
| 2397 |
"group_size": 64,
|
| 2398 |
-
"bits":
|
| 2399 |
},
|
| 2400 |
"language_model.model.layers.26.self_attn.o_proj": {
|
| 2401 |
"group_size": 64,
|
| 2402 |
-
"bits":
|
| 2403 |
},
|
| 2404 |
"language_model.model.layers.26.mlp.gate_proj": {
|
| 2405 |
"group_size": 64,
|
|
@@ -2415,19 +2415,19 @@
|
|
| 2415 |
},
|
| 2416 |
"language_model.model.layers.27.self_attn.q_proj": {
|
| 2417 |
"group_size": 64,
|
| 2418 |
-
"bits":
|
| 2419 |
},
|
| 2420 |
"language_model.model.layers.27.self_attn.k_proj": {
|
| 2421 |
"group_size": 64,
|
| 2422 |
-
"bits":
|
| 2423 |
},
|
| 2424 |
"language_model.model.layers.27.self_attn.v_proj": {
|
| 2425 |
"group_size": 64,
|
| 2426 |
-
"bits":
|
| 2427 |
},
|
| 2428 |
"language_model.model.layers.27.self_attn.o_proj": {
|
| 2429 |
"group_size": 64,
|
| 2430 |
-
"bits":
|
| 2431 |
},
|
| 2432 |
"language_model.model.layers.27.mlp.gate_proj": {
|
| 2433 |
"group_size": 64,
|
|
@@ -2443,19 +2443,19 @@
|
|
| 2443 |
},
|
| 2444 |
"language_model.model.layers.28.self_attn.q_proj": {
|
| 2445 |
"group_size": 64,
|
| 2446 |
-
"bits":
|
| 2447 |
},
|
| 2448 |
"language_model.model.layers.28.self_attn.k_proj": {
|
| 2449 |
"group_size": 64,
|
| 2450 |
-
"bits":
|
| 2451 |
},
|
| 2452 |
"language_model.model.layers.28.self_attn.v_proj": {
|
| 2453 |
"group_size": 64,
|
| 2454 |
-
"bits":
|
| 2455 |
},
|
| 2456 |
"language_model.model.layers.28.self_attn.o_proj": {
|
| 2457 |
"group_size": 64,
|
| 2458 |
-
"bits":
|
| 2459 |
},
|
| 2460 |
"language_model.model.layers.28.mlp.gate_proj": {
|
| 2461 |
"group_size": 64,
|
|
@@ -2471,15 +2471,15 @@
|
|
| 2471 |
},
|
| 2472 |
"language_model.model.layers.29.self_attn.q_proj": {
|
| 2473 |
"group_size": 64,
|
| 2474 |
-
"bits":
|
| 2475 |
},
|
| 2476 |
"language_model.model.layers.29.self_attn.k_proj": {
|
| 2477 |
"group_size": 64,
|
| 2478 |
-
"bits":
|
| 2479 |
},
|
| 2480 |
"language_model.model.layers.29.self_attn.o_proj": {
|
| 2481 |
"group_size": 64,
|
| 2482 |
-
"bits":
|
| 2483 |
},
|
| 2484 |
"language_model.model.layers.29.mlp.gate_proj": {
|
| 2485 |
"group_size": 64,
|
|
@@ -2495,19 +2495,19 @@
|
|
| 2495 |
},
|
| 2496 |
"language_model.model.layers.30.self_attn.q_proj": {
|
| 2497 |
"group_size": 64,
|
| 2498 |
-
"bits":
|
| 2499 |
},
|
| 2500 |
"language_model.model.layers.30.self_attn.k_proj": {
|
| 2501 |
"group_size": 64,
|
| 2502 |
-
"bits":
|
| 2503 |
},
|
| 2504 |
"language_model.model.layers.30.self_attn.v_proj": {
|
| 2505 |
"group_size": 64,
|
| 2506 |
-
"bits":
|
| 2507 |
},
|
| 2508 |
"language_model.model.layers.30.self_attn.o_proj": {
|
| 2509 |
"group_size": 64,
|
| 2510 |
-
"bits":
|
| 2511 |
},
|
| 2512 |
"language_model.model.layers.30.mlp.gate_proj": {
|
| 2513 |
"group_size": 64,
|
|
@@ -2523,19 +2523,19 @@
|
|
| 2523 |
},
|
| 2524 |
"language_model.model.layers.31.self_attn.q_proj": {
|
| 2525 |
"group_size": 64,
|
| 2526 |
-
"bits":
|
| 2527 |
},
|
| 2528 |
"language_model.model.layers.31.self_attn.k_proj": {
|
| 2529 |
"group_size": 64,
|
| 2530 |
-
"bits":
|
| 2531 |
},
|
| 2532 |
"language_model.model.layers.31.self_attn.v_proj": {
|
| 2533 |
"group_size": 64,
|
| 2534 |
-
"bits":
|
| 2535 |
},
|
| 2536 |
"language_model.model.layers.31.self_attn.o_proj": {
|
| 2537 |
"group_size": 64,
|
| 2538 |
-
"bits":
|
| 2539 |
},
|
| 2540 |
"language_model.model.layers.31.mlp.gate_proj": {
|
| 2541 |
"group_size": 64,
|
|
@@ -2551,19 +2551,19 @@
|
|
| 2551 |
},
|
| 2552 |
"language_model.model.layers.32.self_attn.q_proj": {
|
| 2553 |
"group_size": 64,
|
| 2554 |
-
"bits":
|
| 2555 |
},
|
| 2556 |
"language_model.model.layers.32.self_attn.k_proj": {
|
| 2557 |
"group_size": 64,
|
| 2558 |
-
"bits":
|
| 2559 |
},
|
| 2560 |
"language_model.model.layers.32.self_attn.v_proj": {
|
| 2561 |
"group_size": 64,
|
| 2562 |
-
"bits":
|
| 2563 |
},
|
| 2564 |
"language_model.model.layers.32.self_attn.o_proj": {
|
| 2565 |
"group_size": 64,
|
| 2566 |
-
"bits":
|
| 2567 |
},
|
| 2568 |
"language_model.model.layers.32.mlp.gate_proj": {
|
| 2569 |
"group_size": 64,
|
|
@@ -2579,19 +2579,19 @@
|
|
| 2579 |
},
|
| 2580 |
"language_model.model.layers.33.self_attn.q_proj": {
|
| 2581 |
"group_size": 64,
|
| 2582 |
-
"bits":
|
| 2583 |
},
|
| 2584 |
"language_model.model.layers.33.self_attn.k_proj": {
|
| 2585 |
"group_size": 64,
|
| 2586 |
-
"bits":
|
| 2587 |
},
|
| 2588 |
"language_model.model.layers.33.self_attn.v_proj": {
|
| 2589 |
"group_size": 64,
|
| 2590 |
-
"bits":
|
| 2591 |
},
|
| 2592 |
"language_model.model.layers.33.self_attn.o_proj": {
|
| 2593 |
"group_size": 64,
|
| 2594 |
-
"bits":
|
| 2595 |
},
|
| 2596 |
"language_model.model.layers.33.mlp.gate_proj": {
|
| 2597 |
"group_size": 64,
|
|
@@ -2607,19 +2607,19 @@
|
|
| 2607 |
},
|
| 2608 |
"language_model.model.layers.34.self_attn.q_proj": {
|
| 2609 |
"group_size": 64,
|
| 2610 |
-
"bits":
|
| 2611 |
},
|
| 2612 |
"language_model.model.layers.34.self_attn.k_proj": {
|
| 2613 |
"group_size": 64,
|
| 2614 |
-
"bits":
|
| 2615 |
},
|
| 2616 |
"language_model.model.layers.34.self_attn.v_proj": {
|
| 2617 |
"group_size": 64,
|
| 2618 |
-
"bits":
|
| 2619 |
},
|
| 2620 |
"language_model.model.layers.34.self_attn.o_proj": {
|
| 2621 |
"group_size": 64,
|
| 2622 |
-
"bits":
|
| 2623 |
},
|
| 2624 |
"language_model.model.layers.34.mlp.gate_proj": {
|
| 2625 |
"group_size": 64,
|
|
@@ -2635,15 +2635,15 @@
|
|
| 2635 |
},
|
| 2636 |
"language_model.model.layers.35.self_attn.q_proj": {
|
| 2637 |
"group_size": 64,
|
| 2638 |
-
"bits":
|
| 2639 |
},
|
| 2640 |
"language_model.model.layers.35.self_attn.k_proj": {
|
| 2641 |
"group_size": 64,
|
| 2642 |
-
"bits":
|
| 2643 |
},
|
| 2644 |
"language_model.model.layers.35.self_attn.o_proj": {
|
| 2645 |
"group_size": 64,
|
| 2646 |
-
"bits":
|
| 2647 |
},
|
| 2648 |
"language_model.model.layers.35.mlp.gate_proj": {
|
| 2649 |
"group_size": 64,
|
|
@@ -2659,19 +2659,19 @@
|
|
| 2659 |
},
|
| 2660 |
"language_model.model.layers.36.self_attn.q_proj": {
|
| 2661 |
"group_size": 64,
|
| 2662 |
-
"bits":
|
| 2663 |
},
|
| 2664 |
"language_model.model.layers.36.self_attn.k_proj": {
|
| 2665 |
"group_size": 64,
|
| 2666 |
-
"bits":
|
| 2667 |
},
|
| 2668 |
"language_model.model.layers.36.self_attn.v_proj": {
|
| 2669 |
"group_size": 64,
|
| 2670 |
-
"bits":
|
| 2671 |
},
|
| 2672 |
"language_model.model.layers.36.self_attn.o_proj": {
|
| 2673 |
"group_size": 64,
|
| 2674 |
-
"bits":
|
| 2675 |
},
|
| 2676 |
"language_model.model.layers.36.mlp.gate_proj": {
|
| 2677 |
"group_size": 64,
|
|
@@ -2687,19 +2687,19 @@
|
|
| 2687 |
},
|
| 2688 |
"language_model.model.layers.37.self_attn.q_proj": {
|
| 2689 |
"group_size": 64,
|
| 2690 |
-
"bits":
|
| 2691 |
},
|
| 2692 |
"language_model.model.layers.37.self_attn.k_proj": {
|
| 2693 |
"group_size": 64,
|
| 2694 |
-
"bits":
|
| 2695 |
},
|
| 2696 |
"language_model.model.layers.37.self_attn.v_proj": {
|
| 2697 |
"group_size": 64,
|
| 2698 |
-
"bits":
|
| 2699 |
},
|
| 2700 |
"language_model.model.layers.37.self_attn.o_proj": {
|
| 2701 |
"group_size": 64,
|
| 2702 |
-
"bits":
|
| 2703 |
},
|
| 2704 |
"language_model.model.layers.37.mlp.gate_proj": {
|
| 2705 |
"group_size": 64,
|
|
@@ -2715,19 +2715,19 @@
|
|
| 2715 |
},
|
| 2716 |
"language_model.model.layers.38.self_attn.q_proj": {
|
| 2717 |
"group_size": 64,
|
| 2718 |
-
"bits":
|
| 2719 |
},
|
| 2720 |
"language_model.model.layers.38.self_attn.k_proj": {
|
| 2721 |
"group_size": 64,
|
| 2722 |
-
"bits":
|
| 2723 |
},
|
| 2724 |
"language_model.model.layers.38.self_attn.v_proj": {
|
| 2725 |
"group_size": 64,
|
| 2726 |
-
"bits":
|
| 2727 |
},
|
| 2728 |
"language_model.model.layers.38.self_attn.o_proj": {
|
| 2729 |
"group_size": 64,
|
| 2730 |
-
"bits":
|
| 2731 |
},
|
| 2732 |
"language_model.model.layers.38.mlp.gate_proj": {
|
| 2733 |
"group_size": 64,
|
|
@@ -2743,19 +2743,19 @@
|
|
| 2743 |
},
|
| 2744 |
"language_model.model.layers.39.self_attn.q_proj": {
|
| 2745 |
"group_size": 64,
|
| 2746 |
-
"bits":
|
| 2747 |
},
|
| 2748 |
"language_model.model.layers.39.self_attn.k_proj": {
|
| 2749 |
"group_size": 64,
|
| 2750 |
-
"bits":
|
| 2751 |
},
|
| 2752 |
"language_model.model.layers.39.self_attn.v_proj": {
|
| 2753 |
"group_size": 64,
|
| 2754 |
-
"bits":
|
| 2755 |
},
|
| 2756 |
"language_model.model.layers.39.self_attn.o_proj": {
|
| 2757 |
"group_size": 64,
|
| 2758 |
-
"bits":
|
| 2759 |
},
|
| 2760 |
"language_model.model.layers.39.mlp.gate_proj": {
|
| 2761 |
"group_size": 64,
|
|
@@ -2771,19 +2771,19 @@
|
|
| 2771 |
},
|
| 2772 |
"language_model.model.layers.40.self_attn.q_proj": {
|
| 2773 |
"group_size": 64,
|
| 2774 |
-
"bits":
|
| 2775 |
},
|
| 2776 |
"language_model.model.layers.40.self_attn.k_proj": {
|
| 2777 |
"group_size": 64,
|
| 2778 |
-
"bits":
|
| 2779 |
},
|
| 2780 |
"language_model.model.layers.40.self_attn.v_proj": {
|
| 2781 |
"group_size": 64,
|
| 2782 |
-
"bits":
|
| 2783 |
},
|
| 2784 |
"language_model.model.layers.40.self_attn.o_proj": {
|
| 2785 |
"group_size": 64,
|
| 2786 |
-
"bits":
|
| 2787 |
},
|
| 2788 |
"language_model.model.layers.40.mlp.gate_proj": {
|
| 2789 |
"group_size": 64,
|
|
@@ -2799,15 +2799,15 @@
|
|
| 2799 |
},
|
| 2800 |
"language_model.model.layers.41.self_attn.q_proj": {
|
| 2801 |
"group_size": 64,
|
| 2802 |
-
"bits":
|
| 2803 |
},
|
| 2804 |
"language_model.model.layers.41.self_attn.k_proj": {
|
| 2805 |
"group_size": 64,
|
| 2806 |
-
"bits":
|
| 2807 |
},
|
| 2808 |
"language_model.model.layers.41.self_attn.o_proj": {
|
| 2809 |
"group_size": 64,
|
| 2810 |
-
"bits":
|
| 2811 |
},
|
| 2812 |
"language_model.model.layers.41.mlp.gate_proj": {
|
| 2813 |
"group_size": 64,
|
|
@@ -2823,19 +2823,19 @@
|
|
| 2823 |
},
|
| 2824 |
"language_model.model.layers.42.self_attn.q_proj": {
|
| 2825 |
"group_size": 64,
|
| 2826 |
-
"bits":
|
| 2827 |
},
|
| 2828 |
"language_model.model.layers.42.self_attn.k_proj": {
|
| 2829 |
"group_size": 64,
|
| 2830 |
-
"bits":
|
| 2831 |
},
|
| 2832 |
"language_model.model.layers.42.self_attn.v_proj": {
|
| 2833 |
"group_size": 64,
|
| 2834 |
-
"bits":
|
| 2835 |
},
|
| 2836 |
"language_model.model.layers.42.self_attn.o_proj": {
|
| 2837 |
"group_size": 64,
|
| 2838 |
-
"bits":
|
| 2839 |
},
|
| 2840 |
"language_model.model.layers.42.mlp.gate_proj": {
|
| 2841 |
"group_size": 64,
|
|
@@ -2851,19 +2851,19 @@
|
|
| 2851 |
},
|
| 2852 |
"language_model.model.layers.43.self_attn.q_proj": {
|
| 2853 |
"group_size": 64,
|
| 2854 |
-
"bits":
|
| 2855 |
},
|
| 2856 |
"language_model.model.layers.43.self_attn.k_proj": {
|
| 2857 |
"group_size": 64,
|
| 2858 |
-
"bits":
|
| 2859 |
},
|
| 2860 |
"language_model.model.layers.43.self_attn.v_proj": {
|
| 2861 |
"group_size": 64,
|
| 2862 |
-
"bits":
|
| 2863 |
},
|
| 2864 |
"language_model.model.layers.43.self_attn.o_proj": {
|
| 2865 |
"group_size": 64,
|
| 2866 |
-
"bits":
|
| 2867 |
},
|
| 2868 |
"language_model.model.layers.43.mlp.gate_proj": {
|
| 2869 |
"group_size": 64,
|
|
@@ -2879,19 +2879,19 @@
|
|
| 2879 |
},
|
| 2880 |
"language_model.model.layers.44.self_attn.q_proj": {
|
| 2881 |
"group_size": 64,
|
| 2882 |
-
"bits":
|
| 2883 |
},
|
| 2884 |
"language_model.model.layers.44.self_attn.k_proj": {
|
| 2885 |
"group_size": 64,
|
| 2886 |
-
"bits":
|
| 2887 |
},
|
| 2888 |
"language_model.model.layers.44.self_attn.v_proj": {
|
| 2889 |
"group_size": 64,
|
| 2890 |
-
"bits":
|
| 2891 |
},
|
| 2892 |
"language_model.model.layers.44.self_attn.o_proj": {
|
| 2893 |
"group_size": 64,
|
| 2894 |
-
"bits":
|
| 2895 |
},
|
| 2896 |
"language_model.model.layers.44.mlp.gate_proj": {
|
| 2897 |
"group_size": 64,
|
|
@@ -2907,19 +2907,19 @@
|
|
| 2907 |
},
|
| 2908 |
"language_model.model.layers.45.self_attn.q_proj": {
|
| 2909 |
"group_size": 64,
|
| 2910 |
-
"bits":
|
| 2911 |
},
|
| 2912 |
"language_model.model.layers.45.self_attn.k_proj": {
|
| 2913 |
"group_size": 64,
|
| 2914 |
-
"bits":
|
| 2915 |
},
|
| 2916 |
"language_model.model.layers.45.self_attn.v_proj": {
|
| 2917 |
"group_size": 64,
|
| 2918 |
-
"bits":
|
| 2919 |
},
|
| 2920 |
"language_model.model.layers.45.self_attn.o_proj": {
|
| 2921 |
"group_size": 64,
|
| 2922 |
-
"bits":
|
| 2923 |
},
|
| 2924 |
"language_model.model.layers.45.mlp.gate_proj": {
|
| 2925 |
"group_size": 64,
|
|
@@ -2935,19 +2935,19 @@
|
|
| 2935 |
},
|
| 2936 |
"language_model.model.layers.46.self_attn.q_proj": {
|
| 2937 |
"group_size": 64,
|
| 2938 |
-
"bits":
|
| 2939 |
},
|
| 2940 |
"language_model.model.layers.46.self_attn.k_proj": {
|
| 2941 |
"group_size": 64,
|
| 2942 |
-
"bits":
|
| 2943 |
},
|
| 2944 |
"language_model.model.layers.46.self_attn.v_proj": {
|
| 2945 |
"group_size": 64,
|
| 2946 |
-
"bits":
|
| 2947 |
},
|
| 2948 |
"language_model.model.layers.46.self_attn.o_proj": {
|
| 2949 |
"group_size": 64,
|
| 2950 |
-
"bits":
|
| 2951 |
},
|
| 2952 |
"language_model.model.layers.46.mlp.gate_proj": {
|
| 2953 |
"group_size": 64,
|
|
@@ -2963,15 +2963,15 @@
|
|
| 2963 |
},
|
| 2964 |
"language_model.model.layers.47.self_attn.q_proj": {
|
| 2965 |
"group_size": 64,
|
| 2966 |
-
"bits":
|
| 2967 |
},
|
| 2968 |
"language_model.model.layers.47.self_attn.k_proj": {
|
| 2969 |
"group_size": 64,
|
| 2970 |
-
"bits":
|
| 2971 |
},
|
| 2972 |
"language_model.model.layers.47.self_attn.o_proj": {
|
| 2973 |
"group_size": 64,
|
| 2974 |
-
"bits":
|
| 2975 |
},
|
| 2976 |
"language_model.model.layers.47.mlp.gate_proj": {
|
| 2977 |
"group_size": 64,
|
|
@@ -2987,19 +2987,19 @@
|
|
| 2987 |
},
|
| 2988 |
"language_model.model.layers.48.self_attn.q_proj": {
|
| 2989 |
"group_size": 64,
|
| 2990 |
-
"bits":
|
| 2991 |
},
|
| 2992 |
"language_model.model.layers.48.self_attn.k_proj": {
|
| 2993 |
"group_size": 64,
|
| 2994 |
-
"bits":
|
| 2995 |
},
|
| 2996 |
"language_model.model.layers.48.self_attn.v_proj": {
|
| 2997 |
"group_size": 64,
|
| 2998 |
-
"bits":
|
| 2999 |
},
|
| 3000 |
"language_model.model.layers.48.self_attn.o_proj": {
|
| 3001 |
"group_size": 64,
|
| 3002 |
-
"bits":
|
| 3003 |
},
|
| 3004 |
"language_model.model.layers.48.mlp.gate_proj": {
|
| 3005 |
"group_size": 64,
|
|
@@ -3015,19 +3015,19 @@
|
|
| 3015 |
},
|
| 3016 |
"language_model.model.layers.49.self_attn.q_proj": {
|
| 3017 |
"group_size": 64,
|
| 3018 |
-
"bits":
|
| 3019 |
},
|
| 3020 |
"language_model.model.layers.49.self_attn.k_proj": {
|
| 3021 |
"group_size": 64,
|
| 3022 |
-
"bits":
|
| 3023 |
},
|
| 3024 |
"language_model.model.layers.49.self_attn.v_proj": {
|
| 3025 |
"group_size": 64,
|
| 3026 |
-
"bits":
|
| 3027 |
},
|
| 3028 |
"language_model.model.layers.49.self_attn.o_proj": {
|
| 3029 |
"group_size": 64,
|
| 3030 |
-
"bits":
|
| 3031 |
},
|
| 3032 |
"language_model.model.layers.49.mlp.gate_proj": {
|
| 3033 |
"group_size": 64,
|
|
@@ -3043,19 +3043,19 @@
|
|
| 3043 |
},
|
| 3044 |
"language_model.model.layers.50.self_attn.q_proj": {
|
| 3045 |
"group_size": 64,
|
| 3046 |
-
"bits":
|
| 3047 |
},
|
| 3048 |
"language_model.model.layers.50.self_attn.k_proj": {
|
| 3049 |
"group_size": 64,
|
| 3050 |
-
"bits":
|
| 3051 |
},
|
| 3052 |
"language_model.model.layers.50.self_attn.v_proj": {
|
| 3053 |
"group_size": 64,
|
| 3054 |
-
"bits":
|
| 3055 |
},
|
| 3056 |
"language_model.model.layers.50.self_attn.o_proj": {
|
| 3057 |
"group_size": 64,
|
| 3058 |
-
"bits":
|
| 3059 |
},
|
| 3060 |
"language_model.model.layers.50.mlp.gate_proj": {
|
| 3061 |
"group_size": 64,
|
|
@@ -3071,19 +3071,19 @@
|
|
| 3071 |
},
|
| 3072 |
"language_model.model.layers.51.self_attn.q_proj": {
|
| 3073 |
"group_size": 64,
|
| 3074 |
-
"bits":
|
| 3075 |
},
|
| 3076 |
"language_model.model.layers.51.self_attn.k_proj": {
|
| 3077 |
"group_size": 64,
|
| 3078 |
-
"bits":
|
| 3079 |
},
|
| 3080 |
"language_model.model.layers.51.self_attn.v_proj": {
|
| 3081 |
"group_size": 64,
|
| 3082 |
-
"bits":
|
| 3083 |
},
|
| 3084 |
"language_model.model.layers.51.self_attn.o_proj": {
|
| 3085 |
"group_size": 64,
|
| 3086 |
-
"bits":
|
| 3087 |
},
|
| 3088 |
"language_model.model.layers.51.mlp.gate_proj": {
|
| 3089 |
"group_size": 64,
|
|
@@ -3099,43 +3099,43 @@
|
|
| 3099 |
},
|
| 3100 |
"language_model.model.layers.52.self_attn.q_proj": {
|
| 3101 |
"group_size": 64,
|
| 3102 |
-
"bits":
|
| 3103 |
},
|
| 3104 |
"language_model.model.layers.52.self_attn.k_proj": {
|
| 3105 |
"group_size": 64,
|
| 3106 |
-
"bits":
|
| 3107 |
},
|
| 3108 |
"language_model.model.layers.52.self_attn.v_proj": {
|
| 3109 |
"group_size": 64,
|
| 3110 |
-
"bits":
|
| 3111 |
},
|
| 3112 |
"language_model.model.layers.52.self_attn.o_proj": {
|
| 3113 |
"group_size": 64,
|
| 3114 |
-
"bits":
|
| 3115 |
},
|
| 3116 |
"language_model.model.layers.52.mlp.gate_proj": {
|
| 3117 |
"group_size": 64,
|
| 3118 |
-
"bits":
|
| 3119 |
},
|
| 3120 |
"language_model.model.layers.52.mlp.up_proj": {
|
| 3121 |
"group_size": 64,
|
| 3122 |
-
"bits":
|
| 3123 |
},
|
| 3124 |
"language_model.model.layers.52.mlp.down_proj": {
|
| 3125 |
"group_size": 64,
|
| 3126 |
-
"bits":
|
| 3127 |
},
|
| 3128 |
"language_model.model.layers.53.self_attn.q_proj": {
|
| 3129 |
"group_size": 64,
|
| 3130 |
-
"bits":
|
| 3131 |
},
|
| 3132 |
"language_model.model.layers.53.self_attn.k_proj": {
|
| 3133 |
"group_size": 64,
|
| 3134 |
-
"bits":
|
| 3135 |
},
|
| 3136 |
"language_model.model.layers.53.self_attn.o_proj": {
|
| 3137 |
"group_size": 64,
|
| 3138 |
-
"bits":
|
| 3139 |
},
|
| 3140 |
"language_model.model.layers.53.mlp.gate_proj": {
|
| 3141 |
"group_size": 64,
|
|
@@ -3151,19 +3151,19 @@
|
|
| 3151 |
},
|
| 3152 |
"language_model.model.layers.54.self_attn.q_proj": {
|
| 3153 |
"group_size": 64,
|
| 3154 |
-
"bits":
|
| 3155 |
},
|
| 3156 |
"language_model.model.layers.54.self_attn.k_proj": {
|
| 3157 |
"group_size": 64,
|
| 3158 |
-
"bits":
|
| 3159 |
},
|
| 3160 |
"language_model.model.layers.54.self_attn.v_proj": {
|
| 3161 |
"group_size": 64,
|
| 3162 |
-
"bits":
|
| 3163 |
},
|
| 3164 |
"language_model.model.layers.54.self_attn.o_proj": {
|
| 3165 |
"group_size": 64,
|
| 3166 |
-
"bits":
|
| 3167 |
},
|
| 3168 |
"language_model.model.layers.54.mlp.gate_proj": {
|
| 3169 |
"group_size": 64,
|
|
@@ -3179,19 +3179,19 @@
|
|
| 3179 |
},
|
| 3180 |
"language_model.model.layers.55.self_attn.q_proj": {
|
| 3181 |
"group_size": 64,
|
| 3182 |
-
"bits":
|
| 3183 |
},
|
| 3184 |
"language_model.model.layers.55.self_attn.k_proj": {
|
| 3185 |
"group_size": 64,
|
| 3186 |
-
"bits":
|
| 3187 |
},
|
| 3188 |
"language_model.model.layers.55.self_attn.v_proj": {
|
| 3189 |
"group_size": 64,
|
| 3190 |
-
"bits":
|
| 3191 |
},
|
| 3192 |
"language_model.model.layers.55.self_attn.o_proj": {
|
| 3193 |
"group_size": 64,
|
| 3194 |
-
"bits":
|
| 3195 |
},
|
| 3196 |
"language_model.model.layers.55.mlp.gate_proj": {
|
| 3197 |
"group_size": 64,
|
|
@@ -3207,19 +3207,19 @@
|
|
| 3207 |
},
|
| 3208 |
"language_model.model.layers.56.self_attn.q_proj": {
|
| 3209 |
"group_size": 64,
|
| 3210 |
-
"bits":
|
| 3211 |
},
|
| 3212 |
"language_model.model.layers.56.self_attn.k_proj": {
|
| 3213 |
"group_size": 64,
|
| 3214 |
-
"bits":
|
| 3215 |
},
|
| 3216 |
"language_model.model.layers.56.self_attn.v_proj": {
|
| 3217 |
"group_size": 64,
|
| 3218 |
-
"bits":
|
| 3219 |
},
|
| 3220 |
"language_model.model.layers.56.self_attn.o_proj": {
|
| 3221 |
"group_size": 64,
|
| 3222 |
-
"bits":
|
| 3223 |
},
|
| 3224 |
"language_model.model.layers.56.mlp.gate_proj": {
|
| 3225 |
"group_size": 64,
|
|
@@ -3235,83 +3235,83 @@
|
|
| 3235 |
},
|
| 3236 |
"language_model.model.layers.57.self_attn.q_proj": {
|
| 3237 |
"group_size": 64,
|
| 3238 |
-
"bits":
|
| 3239 |
},
|
| 3240 |
"language_model.model.layers.57.self_attn.k_proj": {
|
| 3241 |
"group_size": 64,
|
| 3242 |
-
"bits":
|
| 3243 |
},
|
| 3244 |
"language_model.model.layers.57.self_attn.v_proj": {
|
| 3245 |
"group_size": 64,
|
| 3246 |
-
"bits":
|
| 3247 |
},
|
| 3248 |
"language_model.model.layers.57.self_attn.o_proj": {
|
| 3249 |
"group_size": 64,
|
| 3250 |
-
"bits":
|
| 3251 |
},
|
| 3252 |
"language_model.model.layers.57.mlp.gate_proj": {
|
| 3253 |
"group_size": 64,
|
| 3254 |
-
"bits":
|
| 3255 |
},
|
| 3256 |
"language_model.model.layers.57.mlp.up_proj": {
|
| 3257 |
"group_size": 64,
|
| 3258 |
-
"bits":
|
| 3259 |
},
|
| 3260 |
"language_model.model.layers.57.mlp.down_proj": {
|
| 3261 |
"group_size": 64,
|
| 3262 |
-
"bits":
|
| 3263 |
},
|
| 3264 |
"language_model.model.layers.58.self_attn.q_proj": {
|
| 3265 |
"group_size": 64,
|
| 3266 |
-
"bits":
|
| 3267 |
},
|
| 3268 |
"language_model.model.layers.58.self_attn.k_proj": {
|
| 3269 |
"group_size": 64,
|
| 3270 |
-
"bits":
|
| 3271 |
},
|
| 3272 |
"language_model.model.layers.58.self_attn.v_proj": {
|
| 3273 |
"group_size": 64,
|
| 3274 |
-
"bits":
|
| 3275 |
},
|
| 3276 |
"language_model.model.layers.58.self_attn.o_proj": {
|
| 3277 |
"group_size": 64,
|
| 3278 |
-
"bits":
|
| 3279 |
},
|
| 3280 |
"language_model.model.layers.58.mlp.gate_proj": {
|
| 3281 |
"group_size": 64,
|
| 3282 |
-
"bits":
|
| 3283 |
},
|
| 3284 |
"language_model.model.layers.58.mlp.up_proj": {
|
| 3285 |
"group_size": 64,
|
| 3286 |
-
"bits":
|
| 3287 |
},
|
| 3288 |
"language_model.model.layers.58.mlp.down_proj": {
|
| 3289 |
"group_size": 64,
|
| 3290 |
-
"bits":
|
| 3291 |
},
|
| 3292 |
"language_model.model.layers.59.self_attn.q_proj": {
|
| 3293 |
"group_size": 64,
|
| 3294 |
-
"bits":
|
| 3295 |
},
|
| 3296 |
"language_model.model.layers.59.self_attn.k_proj": {
|
| 3297 |
"group_size": 64,
|
| 3298 |
-
"bits":
|
| 3299 |
},
|
| 3300 |
"language_model.model.layers.59.self_attn.o_proj": {
|
| 3301 |
"group_size": 64,
|
| 3302 |
-
"bits":
|
| 3303 |
},
|
| 3304 |
"language_model.model.layers.59.mlp.gate_proj": {
|
| 3305 |
"group_size": 64,
|
| 3306 |
-
"bits":
|
| 3307 |
},
|
| 3308 |
"language_model.model.layers.59.mlp.up_proj": {
|
| 3309 |
"group_size": 64,
|
| 3310 |
-
"bits":
|
| 3311 |
},
|
| 3312 |
"language_model.model.layers.59.mlp.down_proj": {
|
| 3313 |
"group_size": 64,
|
| 3314 |
-
"bits":
|
| 3315 |
}
|
| 3316 |
},
|
| 3317 |
"text_config": {
|
|
|
|
| 23 |
"bits": 4,
|
| 24 |
"language_model.model.embed_tokens": {
|
| 25 |
"group_size": 64,
|
| 26 |
+
"bits": 6
|
| 27 |
},
|
| 28 |
"language_model.model.layers.0.self_attn.q_proj": {
|
| 29 |
"group_size": 64,
|
| 30 |
+
"bits": 8
|
| 31 |
},
|
| 32 |
"language_model.model.layers.0.self_attn.k_proj": {
|
| 33 |
"group_size": 64,
|
| 34 |
+
"bits": 8
|
| 35 |
},
|
| 36 |
"language_model.model.layers.0.self_attn.v_proj": {
|
| 37 |
"group_size": 64,
|
| 38 |
+
"bits": 8
|
| 39 |
},
|
| 40 |
"language_model.model.layers.0.self_attn.o_proj": {
|
| 41 |
"group_size": 64,
|
| 42 |
+
"bits": 8
|
| 43 |
},
|
| 44 |
"language_model.model.layers.0.mlp.gate_proj": {
|
| 45 |
"group_size": 64,
|
|
|
|
| 55 |
},
|
| 56 |
"language_model.model.layers.1.self_attn.q_proj": {
|
| 57 |
"group_size": 64,
|
| 58 |
+
"bits": 8
|
| 59 |
},
|
| 60 |
"language_model.model.layers.1.self_attn.k_proj": {
|
| 61 |
"group_size": 64,
|
| 62 |
+
"bits": 8
|
| 63 |
},
|
| 64 |
"language_model.model.layers.1.self_attn.v_proj": {
|
| 65 |
"group_size": 64,
|
| 66 |
+
"bits": 8
|
| 67 |
},
|
| 68 |
"language_model.model.layers.1.self_attn.o_proj": {
|
| 69 |
"group_size": 64,
|
| 70 |
+
"bits": 8
|
| 71 |
},
|
| 72 |
"language_model.model.layers.1.mlp.gate_proj": {
|
| 73 |
"group_size": 64,
|
| 74 |
+
"bits": 5
|
| 75 |
},
|
| 76 |
"language_model.model.layers.1.mlp.up_proj": {
|
| 77 |
"group_size": 64,
|
| 78 |
+
"bits": 5
|
| 79 |
},
|
| 80 |
"language_model.model.layers.1.mlp.down_proj": {
|
| 81 |
"group_size": 64,
|
| 82 |
+
"bits": 5
|
| 83 |
},
|
| 84 |
"language_model.model.layers.2.self_attn.q_proj": {
|
| 85 |
"group_size": 64,
|
| 86 |
+
"bits": 8
|
| 87 |
},
|
| 88 |
"language_model.model.layers.2.self_attn.k_proj": {
|
| 89 |
"group_size": 64,
|
| 90 |
+
"bits": 8
|
| 91 |
},
|
| 92 |
"language_model.model.layers.2.self_attn.v_proj": {
|
| 93 |
"group_size": 64,
|
| 94 |
+
"bits": 8
|
| 95 |
},
|
| 96 |
"language_model.model.layers.2.self_attn.o_proj": {
|
| 97 |
"group_size": 64,
|
| 98 |
+
"bits": 8
|
| 99 |
},
|
| 100 |
"language_model.model.layers.2.mlp.gate_proj": {
|
| 101 |
"group_size": 64,
|
|
|
|
| 111 |
},
|
| 112 |
"language_model.model.layers.3.self_attn.q_proj": {
|
| 113 |
"group_size": 64,
|
| 114 |
+
"bits": 8
|
| 115 |
},
|
| 116 |
"language_model.model.layers.3.self_attn.k_proj": {
|
| 117 |
"group_size": 64,
|
| 118 |
+
"bits": 8
|
| 119 |
},
|
| 120 |
"language_model.model.layers.3.self_attn.v_proj": {
|
| 121 |
"group_size": 64,
|
| 122 |
+
"bits": 8
|
| 123 |
},
|
| 124 |
"language_model.model.layers.3.self_attn.o_proj": {
|
| 125 |
"group_size": 64,
|
| 126 |
+
"bits": 8
|
| 127 |
},
|
| 128 |
"language_model.model.layers.3.mlp.gate_proj": {
|
| 129 |
"group_size": 64,
|
|
|
|
| 139 |
},
|
| 140 |
"language_model.model.layers.4.self_attn.q_proj": {
|
| 141 |
"group_size": 64,
|
| 142 |
+
"bits": 8
|
| 143 |
},
|
| 144 |
"language_model.model.layers.4.self_attn.k_proj": {
|
| 145 |
"group_size": 64,
|
| 146 |
+
"bits": 8
|
| 147 |
},
|
| 148 |
"language_model.model.layers.4.self_attn.v_proj": {
|
| 149 |
"group_size": 64,
|
| 150 |
+
"bits": 8
|
| 151 |
},
|
| 152 |
"language_model.model.layers.4.self_attn.o_proj": {
|
| 153 |
"group_size": 64,
|
| 154 |
+
"bits": 8
|
| 155 |
},
|
| 156 |
"language_model.model.layers.4.mlp.gate_proj": {
|
| 157 |
"group_size": 64,
|
|
|
|
| 167 |
},
|
| 168 |
"language_model.model.layers.5.self_attn.q_proj": {
|
| 169 |
"group_size": 64,
|
| 170 |
+
"bits": 8
|
| 171 |
},
|
| 172 |
"language_model.model.layers.5.self_attn.k_proj": {
|
| 173 |
"group_size": 64,
|
| 174 |
+
"bits": 8
|
| 175 |
},
|
| 176 |
"language_model.model.layers.5.self_attn.o_proj": {
|
| 177 |
"group_size": 64,
|
| 178 |
+
"bits": 8
|
| 179 |
},
|
| 180 |
"language_model.model.layers.5.mlp.gate_proj": {
|
| 181 |
"group_size": 64,
|
|
|
|
| 191 |
},
|
| 192 |
"language_model.model.layers.6.self_attn.q_proj": {
|
| 193 |
"group_size": 64,
|
| 194 |
+
"bits": 8
|
| 195 |
},
|
| 196 |
"language_model.model.layers.6.self_attn.k_proj": {
|
| 197 |
"group_size": 64,
|
| 198 |
+
"bits": 8
|
| 199 |
},
|
| 200 |
"language_model.model.layers.6.self_attn.v_proj": {
|
| 201 |
"group_size": 64,
|
| 202 |
+
"bits": 8
|
| 203 |
},
|
| 204 |
"language_model.model.layers.6.self_attn.o_proj": {
|
| 205 |
"group_size": 64,
|
| 206 |
+
"bits": 8
|
| 207 |
},
|
| 208 |
"language_model.model.layers.6.mlp.gate_proj": {
|
| 209 |
"group_size": 64,
|
|
|
|
| 219 |
},
|
| 220 |
"language_model.model.layers.7.self_attn.q_proj": {
|
| 221 |
"group_size": 64,
|
| 222 |
+
"bits": 8
|
| 223 |
},
|
| 224 |
"language_model.model.layers.7.self_attn.k_proj": {
|
| 225 |
"group_size": 64,
|
| 226 |
+
"bits": 8
|
| 227 |
},
|
| 228 |
"language_model.model.layers.7.self_attn.v_proj": {
|
| 229 |
"group_size": 64,
|
| 230 |
+
"bits": 8
|
| 231 |
},
|
| 232 |
"language_model.model.layers.7.self_attn.o_proj": {
|
| 233 |
"group_size": 64,
|
| 234 |
+
"bits": 8
|
| 235 |
},
|
| 236 |
"language_model.model.layers.7.mlp.gate_proj": {
|
| 237 |
"group_size": 64,
|
|
|
|
| 247 |
},
|
| 248 |
"language_model.model.layers.8.self_attn.q_proj": {
|
| 249 |
"group_size": 64,
|
| 250 |
+
"bits": 8
|
| 251 |
},
|
| 252 |
"language_model.model.layers.8.self_attn.k_proj": {
|
| 253 |
"group_size": 64,
|
| 254 |
+
"bits": 8
|
| 255 |
},
|
| 256 |
"language_model.model.layers.8.self_attn.v_proj": {
|
| 257 |
"group_size": 64,
|
| 258 |
+
"bits": 8
|
| 259 |
},
|
| 260 |
"language_model.model.layers.8.self_attn.o_proj": {
|
| 261 |
"group_size": 64,
|
| 262 |
+
"bits": 8
|
| 263 |
},
|
| 264 |
"language_model.model.layers.8.mlp.gate_proj": {
|
| 265 |
"group_size": 64,
|
|
|
|
| 275 |
},
|
| 276 |
"language_model.model.layers.9.self_attn.q_proj": {
|
| 277 |
"group_size": 64,
|
| 278 |
+
"bits": 8
|
| 279 |
},
|
| 280 |
"language_model.model.layers.9.self_attn.k_proj": {
|
| 281 |
"group_size": 64,
|
| 282 |
+
"bits": 8
|
| 283 |
},
|
| 284 |
"language_model.model.layers.9.self_attn.v_proj": {
|
| 285 |
"group_size": 64,
|
| 286 |
+
"bits": 8
|
| 287 |
},
|
| 288 |
"language_model.model.layers.9.self_attn.o_proj": {
|
| 289 |
"group_size": 64,
|
| 290 |
+
"bits": 8
|
| 291 |
},
|
| 292 |
"language_model.model.layers.9.mlp.gate_proj": {
|
| 293 |
"group_size": 64,
|
|
|
|
| 303 |
},
|
| 304 |
"language_model.model.layers.10.self_attn.q_proj": {
|
| 305 |
"group_size": 64,
|
| 306 |
+
"bits": 8
|
| 307 |
},
|
| 308 |
"language_model.model.layers.10.self_attn.k_proj": {
|
| 309 |
"group_size": 64,
|
| 310 |
+
"bits": 8
|
| 311 |
},
|
| 312 |
"language_model.model.layers.10.self_attn.v_proj": {
|
| 313 |
"group_size": 64,
|
| 314 |
+
"bits": 8
|
| 315 |
},
|
| 316 |
"language_model.model.layers.10.self_attn.o_proj": {
|
| 317 |
"group_size": 64,
|
| 318 |
+
"bits": 8
|
| 319 |
},
|
| 320 |
"language_model.model.layers.10.mlp.gate_proj": {
|
| 321 |
"group_size": 64,
|
|
|
|
| 331 |
},
|
| 332 |
"language_model.model.layers.11.self_attn.q_proj": {
|
| 333 |
"group_size": 64,
|
| 334 |
+
"bits": 8
|
| 335 |
},
|
| 336 |
"language_model.model.layers.11.self_attn.k_proj": {
|
| 337 |
"group_size": 64,
|
| 338 |
+
"bits": 8
|
| 339 |
},
|
| 340 |
"language_model.model.layers.11.self_attn.o_proj": {
|
| 341 |
"group_size": 64,
|
| 342 |
+
"bits": 8
|
| 343 |
},
|
| 344 |
"language_model.model.layers.11.mlp.gate_proj": {
|
| 345 |
"group_size": 64,
|
|
|
|
| 355 |
},
|
| 356 |
"language_model.model.layers.12.self_attn.q_proj": {
|
| 357 |
"group_size": 64,
|
| 358 |
+
"bits": 8
|
| 359 |
},
|
| 360 |
"language_model.model.layers.12.self_attn.k_proj": {
|
| 361 |
"group_size": 64,
|
| 362 |
+
"bits": 8
|
| 363 |
},
|
| 364 |
"language_model.model.layers.12.self_attn.v_proj": {
|
| 365 |
"group_size": 64,
|
| 366 |
+
"bits": 8
|
| 367 |
},
|
| 368 |
"language_model.model.layers.12.self_attn.o_proj": {
|
| 369 |
"group_size": 64,
|
| 370 |
+
"bits": 8
|
| 371 |
},
|
| 372 |
"language_model.model.layers.12.mlp.gate_proj": {
|
| 373 |
"group_size": 64,
|
|
|
|
| 383 |
},
|
| 384 |
"language_model.model.layers.13.self_attn.q_proj": {
|
| 385 |
"group_size": 64,
|
| 386 |
+
"bits": 8
|
| 387 |
},
|
| 388 |
"language_model.model.layers.13.self_attn.k_proj": {
|
| 389 |
"group_size": 64,
|
| 390 |
+
"bits": 8
|
| 391 |
},
|
| 392 |
"language_model.model.layers.13.self_attn.v_proj": {
|
| 393 |
"group_size": 64,
|
| 394 |
+
"bits": 8
|
| 395 |
},
|
| 396 |
"language_model.model.layers.13.self_attn.o_proj": {
|
| 397 |
"group_size": 64,
|
| 398 |
+
"bits": 8
|
| 399 |
},
|
| 400 |
"language_model.model.layers.13.mlp.gate_proj": {
|
| 401 |
"group_size": 64,
|
|
|
|
| 411 |
},
|
| 412 |
"language_model.model.layers.14.self_attn.q_proj": {
|
| 413 |
"group_size": 64,
|
| 414 |
+
"bits": 8
|
| 415 |
},
|
| 416 |
"language_model.model.layers.14.self_attn.k_proj": {
|
| 417 |
"group_size": 64,
|
| 418 |
+
"bits": 8
|
| 419 |
},
|
| 420 |
"language_model.model.layers.14.self_attn.v_proj": {
|
| 421 |
"group_size": 64,
|
| 422 |
+
"bits": 8
|
| 423 |
},
|
| 424 |
"language_model.model.layers.14.self_attn.o_proj": {
|
| 425 |
"group_size": 64,
|
| 426 |
+
"bits": 8
|
| 427 |
},
|
| 428 |
"language_model.model.layers.14.mlp.gate_proj": {
|
| 429 |
"group_size": 64,
|
|
|
|
| 439 |
},
|
| 440 |
"language_model.model.layers.15.self_attn.q_proj": {
|
| 441 |
"group_size": 64,
|
| 442 |
+
"bits": 8
|
| 443 |
},
|
| 444 |
"language_model.model.layers.15.self_attn.k_proj": {
|
| 445 |
"group_size": 64,
|
| 446 |
+
"bits": 8
|
| 447 |
},
|
| 448 |
"language_model.model.layers.15.self_attn.v_proj": {
|
| 449 |
"group_size": 64,
|
| 450 |
+
"bits": 8
|
| 451 |
},
|
| 452 |
"language_model.model.layers.15.self_attn.o_proj": {
|
| 453 |
"group_size": 64,
|
| 454 |
+
"bits": 8
|
| 455 |
},
|
| 456 |
"language_model.model.layers.15.mlp.gate_proj": {
|
| 457 |
"group_size": 64,
|
|
|
|
| 467 |
},
|
| 468 |
"language_model.model.layers.16.self_attn.q_proj": {
|
| 469 |
"group_size": 64,
|
| 470 |
+
"bits": 8
|
| 471 |
},
|
| 472 |
"language_model.model.layers.16.self_attn.k_proj": {
|
| 473 |
"group_size": 64,
|
| 474 |
+
"bits": 8
|
| 475 |
},
|
| 476 |
"language_model.model.layers.16.self_attn.v_proj": {
|
| 477 |
"group_size": 64,
|
| 478 |
+
"bits": 8
|
| 479 |
},
|
| 480 |
"language_model.model.layers.16.self_attn.o_proj": {
|
| 481 |
"group_size": 64,
|
| 482 |
+
"bits": 8
|
| 483 |
},
|
| 484 |
"language_model.model.layers.16.mlp.gate_proj": {
|
| 485 |
"group_size": 64,
|
|
|
|
| 495 |
},
|
| 496 |
"language_model.model.layers.17.self_attn.q_proj": {
|
| 497 |
"group_size": 64,
|
| 498 |
+
"bits": 8
|
| 499 |
},
|
| 500 |
"language_model.model.layers.17.self_attn.k_proj": {
|
| 501 |
"group_size": 64,
|
| 502 |
+
"bits": 8
|
| 503 |
},
|
| 504 |
"language_model.model.layers.17.self_attn.o_proj": {
|
| 505 |
"group_size": 64,
|
| 506 |
+
"bits": 8
|
| 507 |
},
|
| 508 |
"language_model.model.layers.17.mlp.gate_proj": {
|
| 509 |
"group_size": 64,
|
|
|
|
| 519 |
},
|
| 520 |
"language_model.model.layers.18.self_attn.q_proj": {
|
| 521 |
"group_size": 64,
|
| 522 |
+
"bits": 8
|
| 523 |
},
|
| 524 |
"language_model.model.layers.18.self_attn.k_proj": {
|
| 525 |
"group_size": 64,
|
| 526 |
+
"bits": 8
|
| 527 |
},
|
| 528 |
"language_model.model.layers.18.self_attn.v_proj": {
|
| 529 |
"group_size": 64,
|
| 530 |
+
"bits": 8
|
| 531 |
},
|
| 532 |
"language_model.model.layers.18.self_attn.o_proj": {
|
| 533 |
"group_size": 64,
|
| 534 |
+
"bits": 8
|
| 535 |
},
|
| 536 |
"language_model.model.layers.18.mlp.gate_proj": {
|
| 537 |
"group_size": 64,
|
|
|
|
| 547 |
},
|
| 548 |
"language_model.model.layers.19.self_attn.q_proj": {
|
| 549 |
"group_size": 64,
|
| 550 |
+
"bits": 8
|
| 551 |
},
|
| 552 |
"language_model.model.layers.19.self_attn.k_proj": {
|
| 553 |
"group_size": 64,
|
| 554 |
+
"bits": 8
|
| 555 |
},
|
| 556 |
"language_model.model.layers.19.self_attn.v_proj": {
|
| 557 |
"group_size": 64,
|
| 558 |
+
"bits": 8
|
| 559 |
},
|
| 560 |
"language_model.model.layers.19.self_attn.o_proj": {
|
| 561 |
"group_size": 64,
|
| 562 |
+
"bits": 8
|
| 563 |
},
|
| 564 |
"language_model.model.layers.19.mlp.gate_proj": {
|
| 565 |
"group_size": 64,
|
|
|
|
| 575 |
},
|
| 576 |
"language_model.model.layers.20.self_attn.q_proj": {
|
| 577 |
"group_size": 64,
|
| 578 |
+
"bits": 8
|
| 579 |
},
|
| 580 |
"language_model.model.layers.20.self_attn.k_proj": {
|
| 581 |
"group_size": 64,
|
| 582 |
+
"bits": 8
|
| 583 |
},
|
| 584 |
"language_model.model.layers.20.self_attn.v_proj": {
|
| 585 |
"group_size": 64,
|
| 586 |
+
"bits": 8
|
| 587 |
},
|
| 588 |
"language_model.model.layers.20.self_attn.o_proj": {
|
| 589 |
"group_size": 64,
|
| 590 |
+
"bits": 8
|
| 591 |
},
|
| 592 |
"language_model.model.layers.20.mlp.gate_proj": {
|
| 593 |
"group_size": 64,
|
|
|
|
| 603 |
},
|
| 604 |
"language_model.model.layers.21.self_attn.q_proj": {
|
| 605 |
"group_size": 64,
|
| 606 |
+
"bits": 8
|
| 607 |
},
|
| 608 |
"language_model.model.layers.21.self_attn.k_proj": {
|
| 609 |
"group_size": 64,
|
| 610 |
+
"bits": 8
|
| 611 |
},
|
| 612 |
"language_model.model.layers.21.self_attn.v_proj": {
|
| 613 |
"group_size": 64,
|
| 614 |
+
"bits": 8
|
| 615 |
},
|
| 616 |
"language_model.model.layers.21.self_attn.o_proj": {
|
| 617 |
"group_size": 64,
|
| 618 |
+
"bits": 8
|
| 619 |
},
|
| 620 |
"language_model.model.layers.21.mlp.gate_proj": {
|
| 621 |
"group_size": 64,
|
|
|
|
| 631 |
},
|
| 632 |
"language_model.model.layers.22.self_attn.q_proj": {
|
| 633 |
"group_size": 64,
|
| 634 |
+
"bits": 8
|
| 635 |
},
|
| 636 |
"language_model.model.layers.22.self_attn.k_proj": {
|
| 637 |
"group_size": 64,
|
| 638 |
+
"bits": 8
|
| 639 |
},
|
| 640 |
"language_model.model.layers.22.self_attn.v_proj": {
|
| 641 |
"group_size": 64,
|
| 642 |
+
"bits": 8
|
| 643 |
},
|
| 644 |
"language_model.model.layers.22.self_attn.o_proj": {
|
| 645 |
"group_size": 64,
|
| 646 |
+
"bits": 8
|
| 647 |
},
|
| 648 |
"language_model.model.layers.22.mlp.gate_proj": {
|
| 649 |
"group_size": 64,
|
|
|
|
| 659 |
},
|
| 660 |
"language_model.model.layers.23.self_attn.q_proj": {
|
| 661 |
"group_size": 64,
|
| 662 |
+
"bits": 8
|
| 663 |
},
|
| 664 |
"language_model.model.layers.23.self_attn.k_proj": {
|
| 665 |
"group_size": 64,
|
| 666 |
+
"bits": 8
|
| 667 |
},
|
| 668 |
"language_model.model.layers.23.self_attn.o_proj": {
|
| 669 |
"group_size": 64,
|
| 670 |
+
"bits": 8
|
| 671 |
},
|
| 672 |
"language_model.model.layers.23.mlp.gate_proj": {
|
| 673 |
"group_size": 64,
|
|
|
|
| 683 |
},
|
| 684 |
"language_model.model.layers.24.self_attn.q_proj": {
|
| 685 |
"group_size": 64,
|
| 686 |
+
"bits": 8
|
| 687 |
},
|
| 688 |
"language_model.model.layers.24.self_attn.k_proj": {
|
| 689 |
"group_size": 64,
|
| 690 |
+
"bits": 8
|
| 691 |
},
|
| 692 |
"language_model.model.layers.24.self_attn.v_proj": {
|
| 693 |
"group_size": 64,
|
| 694 |
+
"bits": 8
|
| 695 |
},
|
| 696 |
"language_model.model.layers.24.self_attn.o_proj": {
|
| 697 |
"group_size": 64,
|
| 698 |
+
"bits": 8
|
| 699 |
},
|
| 700 |
"language_model.model.layers.24.mlp.gate_proj": {
|
| 701 |
"group_size": 64,
|
|
|
|
| 711 |
},
|
| 712 |
"language_model.model.layers.25.self_attn.q_proj": {
|
| 713 |
"group_size": 64,
|
| 714 |
+
"bits": 8
|
| 715 |
},
|
| 716 |
"language_model.model.layers.25.self_attn.k_proj": {
|
| 717 |
"group_size": 64,
|
| 718 |
+
"bits": 8
|
| 719 |
},
|
| 720 |
"language_model.model.layers.25.self_attn.v_proj": {
|
| 721 |
"group_size": 64,
|
| 722 |
+
"bits": 8
|
| 723 |
},
|
| 724 |
"language_model.model.layers.25.self_attn.o_proj": {
|
| 725 |
"group_size": 64,
|
| 726 |
+
"bits": 8
|
| 727 |
},
|
| 728 |
"language_model.model.layers.25.mlp.gate_proj": {
|
| 729 |
"group_size": 64,
|
|
|
|
| 739 |
},
|
| 740 |
"language_model.model.layers.26.self_attn.q_proj": {
|
| 741 |
"group_size": 64,
|
| 742 |
+
"bits": 8
|
| 743 |
},
|
| 744 |
"language_model.model.layers.26.self_attn.k_proj": {
|
| 745 |
"group_size": 64,
|
| 746 |
+
"bits": 8
|
| 747 |
},
|
| 748 |
"language_model.model.layers.26.self_attn.v_proj": {
|
| 749 |
"group_size": 64,
|
| 750 |
+
"bits": 8
|
| 751 |
},
|
| 752 |
"language_model.model.layers.26.self_attn.o_proj": {
|
| 753 |
"group_size": 64,
|
| 754 |
+
"bits": 8
|
| 755 |
},
|
| 756 |
"language_model.model.layers.26.mlp.gate_proj": {
|
| 757 |
"group_size": 64,
|
|
|
|
| 767 |
},
|
| 768 |
"language_model.model.layers.27.self_attn.q_proj": {
|
| 769 |
"group_size": 64,
|
| 770 |
+
"bits": 8
|
| 771 |
},
|
| 772 |
"language_model.model.layers.27.self_attn.k_proj": {
|
| 773 |
"group_size": 64,
|
| 774 |
+
"bits": 8
|
| 775 |
},
|
| 776 |
"language_model.model.layers.27.self_attn.v_proj": {
|
| 777 |
"group_size": 64,
|
| 778 |
+
"bits": 8
|
| 779 |
},
|
| 780 |
"language_model.model.layers.27.self_attn.o_proj": {
|
| 781 |
"group_size": 64,
|
| 782 |
+
"bits": 8
|
| 783 |
},
|
| 784 |
"language_model.model.layers.27.mlp.gate_proj": {
|
| 785 |
"group_size": 64,
|
|
|
|
| 795 |
},
|
| 796 |
"language_model.model.layers.28.self_attn.q_proj": {
|
| 797 |
"group_size": 64,
|
| 798 |
+
"bits": 8
|
| 799 |
},
|
| 800 |
"language_model.model.layers.28.self_attn.k_proj": {
|
| 801 |
"group_size": 64,
|
| 802 |
+
"bits": 8
|
| 803 |
},
|
| 804 |
"language_model.model.layers.28.self_attn.v_proj": {
|
| 805 |
"group_size": 64,
|
| 806 |
+
"bits": 8
|
| 807 |
},
|
| 808 |
"language_model.model.layers.28.self_attn.o_proj": {
|
| 809 |
"group_size": 64,
|
| 810 |
+
"bits": 8
|
| 811 |
},
|
| 812 |
"language_model.model.layers.28.mlp.gate_proj": {
|
| 813 |
"group_size": 64,
|
|
|
|
| 823 |
},
|
| 824 |
"language_model.model.layers.29.self_attn.q_proj": {
|
| 825 |
"group_size": 64,
|
| 826 |
+
"bits": 8
|
| 827 |
},
|
| 828 |
"language_model.model.layers.29.self_attn.k_proj": {
|
| 829 |
"group_size": 64,
|
| 830 |
+
"bits": 8
|
| 831 |
},
|
| 832 |
"language_model.model.layers.29.self_attn.o_proj": {
|
| 833 |
"group_size": 64,
|
| 834 |
+
"bits": 8
|
| 835 |
},
|
| 836 |
"language_model.model.layers.29.mlp.gate_proj": {
|
| 837 |
"group_size": 64,
|
|
|
|
| 847 |
},
|
| 848 |
"language_model.model.layers.30.self_attn.q_proj": {
|
| 849 |
"group_size": 64,
|
| 850 |
+
"bits": 8
|
| 851 |
},
|
| 852 |
"language_model.model.layers.30.self_attn.k_proj": {
|
| 853 |
"group_size": 64,
|
| 854 |
+
"bits": 8
|
| 855 |
},
|
| 856 |
"language_model.model.layers.30.self_attn.v_proj": {
|
| 857 |
"group_size": 64,
|
| 858 |
+
"bits": 8
|
| 859 |
},
|
| 860 |
"language_model.model.layers.30.self_attn.o_proj": {
|
| 861 |
"group_size": 64,
|
| 862 |
+
"bits": 8
|
| 863 |
},
|
| 864 |
"language_model.model.layers.30.mlp.gate_proj": {
|
| 865 |
"group_size": 64,
|
|
|
|
| 875 |
},
|
| 876 |
"language_model.model.layers.31.self_attn.q_proj": {
|
| 877 |
"group_size": 64,
|
| 878 |
+
"bits": 8
|
| 879 |
},
|
| 880 |
"language_model.model.layers.31.self_attn.k_proj": {
|
| 881 |
"group_size": 64,
|
| 882 |
+
"bits": 8
|
| 883 |
},
|
| 884 |
"language_model.model.layers.31.self_attn.v_proj": {
|
| 885 |
"group_size": 64,
|
| 886 |
+
"bits": 8
|
| 887 |
},
|
| 888 |
"language_model.model.layers.31.self_attn.o_proj": {
|
| 889 |
"group_size": 64,
|
| 890 |
+
"bits": 8
|
| 891 |
},
|
| 892 |
"language_model.model.layers.31.mlp.gate_proj": {
|
| 893 |
"group_size": 64,
|
|
|
|
| 903 |
},
|
| 904 |
"language_model.model.layers.32.self_attn.q_proj": {
|
| 905 |
"group_size": 64,
|
| 906 |
+
"bits": 8
|
| 907 |
},
|
| 908 |
"language_model.model.layers.32.self_attn.k_proj": {
|
| 909 |
"group_size": 64,
|
| 910 |
+
"bits": 8
|
| 911 |
},
|
| 912 |
"language_model.model.layers.32.self_attn.v_proj": {
|
| 913 |
"group_size": 64,
|
| 914 |
+
"bits": 8
|
| 915 |
},
|
| 916 |
"language_model.model.layers.32.self_attn.o_proj": {
|
| 917 |
"group_size": 64,
|
| 918 |
+
"bits": 8
|
| 919 |
},
|
| 920 |
"language_model.model.layers.32.mlp.gate_proj": {
|
| 921 |
"group_size": 64,
|
|
|
|
| 931 |
},
|
| 932 |
"language_model.model.layers.33.self_attn.q_proj": {
|
| 933 |
"group_size": 64,
|
| 934 |
+
"bits": 8
|
| 935 |
},
|
| 936 |
"language_model.model.layers.33.self_attn.k_proj": {
|
| 937 |
"group_size": 64,
|
| 938 |
+
"bits": 8
|
| 939 |
},
|
| 940 |
"language_model.model.layers.33.self_attn.v_proj": {
|
| 941 |
"group_size": 64,
|
| 942 |
+
"bits": 8
|
| 943 |
},
|
| 944 |
"language_model.model.layers.33.self_attn.o_proj": {
|
| 945 |
"group_size": 64,
|
| 946 |
+
"bits": 8
|
| 947 |
},
|
| 948 |
"language_model.model.layers.33.mlp.gate_proj": {
|
| 949 |
"group_size": 64,
|
|
|
|
| 959 |
},
|
| 960 |
"language_model.model.layers.34.self_attn.q_proj": {
|
| 961 |
"group_size": 64,
|
| 962 |
+
"bits": 8
|
| 963 |
},
|
| 964 |
"language_model.model.layers.34.self_attn.k_proj": {
|
| 965 |
"group_size": 64,
|
| 966 |
+
"bits": 8
|
| 967 |
},
|
| 968 |
"language_model.model.layers.34.self_attn.v_proj": {
|
| 969 |
"group_size": 64,
|
| 970 |
+
"bits": 8
|
| 971 |
},
|
| 972 |
"language_model.model.layers.34.self_attn.o_proj": {
|
| 973 |
"group_size": 64,
|
| 974 |
+
"bits": 8
|
| 975 |
},
|
| 976 |
"language_model.model.layers.34.mlp.gate_proj": {
|
| 977 |
"group_size": 64,
|
|
|
|
| 987 |
},
|
| 988 |
"language_model.model.layers.35.self_attn.q_proj": {
|
| 989 |
"group_size": 64,
|
| 990 |
+
"bits": 8
|
| 991 |
},
|
| 992 |
"language_model.model.layers.35.self_attn.k_proj": {
|
| 993 |
"group_size": 64,
|
| 994 |
+
"bits": 8
|
| 995 |
},
|
| 996 |
"language_model.model.layers.35.self_attn.o_proj": {
|
| 997 |
"group_size": 64,
|
| 998 |
+
"bits": 8
|
| 999 |
},
|
| 1000 |
"language_model.model.layers.35.mlp.gate_proj": {
|
| 1001 |
"group_size": 64,
|
|
|
|
| 1011 |
},
|
| 1012 |
"language_model.model.layers.36.self_attn.q_proj": {
|
| 1013 |
"group_size": 64,
|
| 1014 |
+
"bits": 8
|
| 1015 |
},
|
| 1016 |
"language_model.model.layers.36.self_attn.k_proj": {
|
| 1017 |
"group_size": 64,
|
| 1018 |
+
"bits": 8
|
| 1019 |
},
|
| 1020 |
"language_model.model.layers.36.self_attn.v_proj": {
|
| 1021 |
"group_size": 64,
|
| 1022 |
+
"bits": 8
|
| 1023 |
},
|
| 1024 |
"language_model.model.layers.36.self_attn.o_proj": {
|
| 1025 |
"group_size": 64,
|
| 1026 |
+
"bits": 8
|
| 1027 |
},
|
| 1028 |
"language_model.model.layers.36.mlp.gate_proj": {
|
| 1029 |
"group_size": 64,
|
|
|
|
| 1039 |
},
|
| 1040 |
"language_model.model.layers.37.self_attn.q_proj": {
|
| 1041 |
"group_size": 64,
|
| 1042 |
+
"bits": 8
|
| 1043 |
},
|
| 1044 |
"language_model.model.layers.37.self_attn.k_proj": {
|
| 1045 |
"group_size": 64,
|
| 1046 |
+
"bits": 8
|
| 1047 |
},
|
| 1048 |
"language_model.model.layers.37.self_attn.v_proj": {
|
| 1049 |
"group_size": 64,
|
| 1050 |
+
"bits": 8
|
| 1051 |
},
|
| 1052 |
"language_model.model.layers.37.self_attn.o_proj": {
|
| 1053 |
"group_size": 64,
|
| 1054 |
+
"bits": 8
|
| 1055 |
},
|
| 1056 |
"language_model.model.layers.37.mlp.gate_proj": {
|
| 1057 |
"group_size": 64,
|
|
|
|
| 1067 |
},
|
| 1068 |
"language_model.model.layers.38.self_attn.q_proj": {
|
| 1069 |
"group_size": 64,
|
| 1070 |
+
"bits": 8
|
| 1071 |
},
|
| 1072 |
"language_model.model.layers.38.self_attn.k_proj": {
|
| 1073 |
"group_size": 64,
|
| 1074 |
+
"bits": 8
|
| 1075 |
},
|
| 1076 |
"language_model.model.layers.38.self_attn.v_proj": {
|
| 1077 |
"group_size": 64,
|
| 1078 |
+
"bits": 8
|
| 1079 |
},
|
| 1080 |
"language_model.model.layers.38.self_attn.o_proj": {
|
| 1081 |
"group_size": 64,
|
| 1082 |
+
"bits": 8
|
| 1083 |
},
|
| 1084 |
"language_model.model.layers.38.mlp.gate_proj": {
|
| 1085 |
"group_size": 64,
|
|
|
|
| 1095 |
},
|
| 1096 |
"language_model.model.layers.39.self_attn.q_proj": {
|
| 1097 |
"group_size": 64,
|
| 1098 |
+
"bits": 8
|
| 1099 |
},
|
| 1100 |
"language_model.model.layers.39.self_attn.k_proj": {
|
| 1101 |
"group_size": 64,
|
| 1102 |
+
"bits": 8
|
| 1103 |
},
|
| 1104 |
"language_model.model.layers.39.self_attn.v_proj": {
|
| 1105 |
"group_size": 64,
|
| 1106 |
+
"bits": 8
|
| 1107 |
},
|
| 1108 |
"language_model.model.layers.39.self_attn.o_proj": {
|
| 1109 |
"group_size": 64,
|
| 1110 |
+
"bits": 8
|
| 1111 |
},
|
| 1112 |
"language_model.model.layers.39.mlp.gate_proj": {
|
| 1113 |
"group_size": 64,
|
|
|
|
| 1123 |
},
|
| 1124 |
"language_model.model.layers.40.self_attn.q_proj": {
|
| 1125 |
"group_size": 64,
|
| 1126 |
+
"bits": 8
|
| 1127 |
},
|
| 1128 |
"language_model.model.layers.40.self_attn.k_proj": {
|
| 1129 |
"group_size": 64,
|
| 1130 |
+
"bits": 8
|
| 1131 |
},
|
| 1132 |
"language_model.model.layers.40.self_attn.v_proj": {
|
| 1133 |
"group_size": 64,
|
| 1134 |
+
"bits": 8
|
| 1135 |
},
|
| 1136 |
"language_model.model.layers.40.self_attn.o_proj": {
|
| 1137 |
"group_size": 64,
|
| 1138 |
+
"bits": 8
|
| 1139 |
},
|
| 1140 |
"language_model.model.layers.40.mlp.gate_proj": {
|
| 1141 |
"group_size": 64,
|
|
|
|
| 1151 |
},
|
| 1152 |
"language_model.model.layers.41.self_attn.q_proj": {
|
| 1153 |
"group_size": 64,
|
| 1154 |
+
"bits": 8
|
| 1155 |
},
|
| 1156 |
"language_model.model.layers.41.self_attn.k_proj": {
|
| 1157 |
"group_size": 64,
|
| 1158 |
+
"bits": 8
|
| 1159 |
},
|
| 1160 |
"language_model.model.layers.41.self_attn.o_proj": {
|
| 1161 |
"group_size": 64,
|
| 1162 |
+
"bits": 8
|
| 1163 |
},
|
| 1164 |
"language_model.model.layers.41.mlp.gate_proj": {
|
| 1165 |
"group_size": 64,
|
|
|
|
| 1175 |
},
|
| 1176 |
"language_model.model.layers.42.self_attn.q_proj": {
|
| 1177 |
"group_size": 64,
|
| 1178 |
+
"bits": 8
|
| 1179 |
},
|
| 1180 |
"language_model.model.layers.42.self_attn.k_proj": {
|
| 1181 |
"group_size": 64,
|
| 1182 |
+
"bits": 8
|
| 1183 |
},
|
| 1184 |
"language_model.model.layers.42.self_attn.v_proj": {
|
| 1185 |
"group_size": 64,
|
| 1186 |
+
"bits": 8
|
| 1187 |
},
|
| 1188 |
"language_model.model.layers.42.self_attn.o_proj": {
|
| 1189 |
"group_size": 64,
|
| 1190 |
+
"bits": 8
|
| 1191 |
},
|
| 1192 |
"language_model.model.layers.42.mlp.gate_proj": {
|
| 1193 |
"group_size": 64,
|
|
|
|
| 1203 |
},
|
| 1204 |
"language_model.model.layers.43.self_attn.q_proj": {
|
| 1205 |
"group_size": 64,
|
| 1206 |
+
"bits": 8
|
| 1207 |
},
|
| 1208 |
"language_model.model.layers.43.self_attn.k_proj": {
|
| 1209 |
"group_size": 64,
|
| 1210 |
+
"bits": 8
|
| 1211 |
},
|
| 1212 |
"language_model.model.layers.43.self_attn.v_proj": {
|
| 1213 |
"group_size": 64,
|
| 1214 |
+
"bits": 8
|
| 1215 |
},
|
| 1216 |
"language_model.model.layers.43.self_attn.o_proj": {
|
| 1217 |
"group_size": 64,
|
| 1218 |
+
"bits": 8
|
| 1219 |
},
|
| 1220 |
"language_model.model.layers.43.mlp.gate_proj": {
|
| 1221 |
"group_size": 64,
|
|
|
|
| 1231 |
},
|
| 1232 |
"language_model.model.layers.44.self_attn.q_proj": {
|
| 1233 |
"group_size": 64,
|
| 1234 |
+
"bits": 8
|
| 1235 |
},
|
| 1236 |
"language_model.model.layers.44.self_attn.k_proj": {
|
| 1237 |
"group_size": 64,
|
| 1238 |
+
"bits": 8
|
| 1239 |
},
|
| 1240 |
"language_model.model.layers.44.self_attn.v_proj": {
|
| 1241 |
"group_size": 64,
|
| 1242 |
+
"bits": 8
|
| 1243 |
},
|
| 1244 |
"language_model.model.layers.44.self_attn.o_proj": {
|
| 1245 |
"group_size": 64,
|
| 1246 |
+
"bits": 8
|
| 1247 |
},
|
| 1248 |
"language_model.model.layers.44.mlp.gate_proj": {
|
| 1249 |
"group_size": 64,
|
|
|
|
| 1259 |
},
|
| 1260 |
"language_model.model.layers.45.self_attn.q_proj": {
|
| 1261 |
"group_size": 64,
|
| 1262 |
+
"bits": 8
|
| 1263 |
},
|
| 1264 |
"language_model.model.layers.45.self_attn.k_proj": {
|
| 1265 |
"group_size": 64,
|
| 1266 |
+
"bits": 8
|
| 1267 |
},
|
| 1268 |
"language_model.model.layers.45.self_attn.v_proj": {
|
| 1269 |
"group_size": 64,
|
| 1270 |
+
"bits": 8
|
| 1271 |
},
|
| 1272 |
"language_model.model.layers.45.self_attn.o_proj": {
|
| 1273 |
"group_size": 64,
|
| 1274 |
+
"bits": 8
|
| 1275 |
},
|
| 1276 |
"language_model.model.layers.45.mlp.gate_proj": {
|
| 1277 |
"group_size": 64,
|
|
|
|
| 1287 |
},
|
| 1288 |
"language_model.model.layers.46.self_attn.q_proj": {
|
| 1289 |
"group_size": 64,
|
| 1290 |
+
"bits": 8
|
| 1291 |
},
|
| 1292 |
"language_model.model.layers.46.self_attn.k_proj": {
|
| 1293 |
"group_size": 64,
|
| 1294 |
+
"bits": 8
|
| 1295 |
},
|
| 1296 |
"language_model.model.layers.46.self_attn.v_proj": {
|
| 1297 |
"group_size": 64,
|
| 1298 |
+
"bits": 8
|
| 1299 |
},
|
| 1300 |
"language_model.model.layers.46.self_attn.o_proj": {
|
| 1301 |
"group_size": 64,
|
| 1302 |
+
"bits": 8
|
| 1303 |
},
|
| 1304 |
"language_model.model.layers.46.mlp.gate_proj": {
|
| 1305 |
"group_size": 64,
|
|
|
|
| 1315 |
},
|
| 1316 |
"language_model.model.layers.47.self_attn.q_proj": {
|
| 1317 |
"group_size": 64,
|
| 1318 |
+
"bits": 8
|
| 1319 |
},
|
| 1320 |
"language_model.model.layers.47.self_attn.k_proj": {
|
| 1321 |
"group_size": 64,
|
| 1322 |
+
"bits": 8
|
| 1323 |
},
|
| 1324 |
"language_model.model.layers.47.self_attn.o_proj": {
|
| 1325 |
"group_size": 64,
|
| 1326 |
+
"bits": 8
|
| 1327 |
},
|
| 1328 |
"language_model.model.layers.47.mlp.gate_proj": {
|
| 1329 |
"group_size": 64,
|
|
|
|
| 1339 |
},
|
| 1340 |
"language_model.model.layers.48.self_attn.q_proj": {
|
| 1341 |
"group_size": 64,
|
| 1342 |
+
"bits": 8
|
| 1343 |
},
|
| 1344 |
"language_model.model.layers.48.self_attn.k_proj": {
|
| 1345 |
"group_size": 64,
|
| 1346 |
+
"bits": 8
|
| 1347 |
},
|
| 1348 |
"language_model.model.layers.48.self_attn.v_proj": {
|
| 1349 |
"group_size": 64,
|
| 1350 |
+
"bits": 8
|
| 1351 |
},
|
| 1352 |
"language_model.model.layers.48.self_attn.o_proj": {
|
| 1353 |
"group_size": 64,
|
| 1354 |
+
"bits": 8
|
| 1355 |
},
|
| 1356 |
"language_model.model.layers.48.mlp.gate_proj": {
|
| 1357 |
"group_size": 64,
|
|
|
|
| 1367 |
},
|
| 1368 |
"language_model.model.layers.49.self_attn.q_proj": {
|
| 1369 |
"group_size": 64,
|
| 1370 |
+
"bits": 8
|
| 1371 |
},
|
| 1372 |
"language_model.model.layers.49.self_attn.k_proj": {
|
| 1373 |
"group_size": 64,
|
| 1374 |
+
"bits": 8
|
| 1375 |
},
|
| 1376 |
"language_model.model.layers.49.self_attn.v_proj": {
|
| 1377 |
"group_size": 64,
|
| 1378 |
+
"bits": 8
|
| 1379 |
},
|
| 1380 |
"language_model.model.layers.49.self_attn.o_proj": {
|
| 1381 |
"group_size": 64,
|
| 1382 |
+
"bits": 8
|
| 1383 |
},
|
| 1384 |
"language_model.model.layers.49.mlp.gate_proj": {
|
| 1385 |
"group_size": 64,
|
|
|
|
| 1395 |
},
|
| 1396 |
"language_model.model.layers.50.self_attn.q_proj": {
|
| 1397 |
"group_size": 64,
|
| 1398 |
+
"bits": 8
|
| 1399 |
},
|
| 1400 |
"language_model.model.layers.50.self_attn.k_proj": {
|
| 1401 |
"group_size": 64,
|
| 1402 |
+
"bits": 8
|
| 1403 |
},
|
| 1404 |
"language_model.model.layers.50.self_attn.v_proj": {
|
| 1405 |
"group_size": 64,
|
| 1406 |
+
"bits": 8
|
| 1407 |
},
|
| 1408 |
"language_model.model.layers.50.self_attn.o_proj": {
|
| 1409 |
"group_size": 64,
|
| 1410 |
+
"bits": 8
|
| 1411 |
},
|
| 1412 |
"language_model.model.layers.50.mlp.gate_proj": {
|
| 1413 |
"group_size": 64,
|
|
|
|
| 1423 |
},
|
| 1424 |
"language_model.model.layers.51.self_attn.q_proj": {
|
| 1425 |
"group_size": 64,
|
| 1426 |
+
"bits": 8
|
| 1427 |
},
|
| 1428 |
"language_model.model.layers.51.self_attn.k_proj": {
|
| 1429 |
"group_size": 64,
|
| 1430 |
+
"bits": 8
|
| 1431 |
},
|
| 1432 |
"language_model.model.layers.51.self_attn.v_proj": {
|
| 1433 |
"group_size": 64,
|
| 1434 |
+
"bits": 8
|
| 1435 |
},
|
| 1436 |
"language_model.model.layers.51.self_attn.o_proj": {
|
| 1437 |
"group_size": 64,
|
| 1438 |
+
"bits": 8
|
| 1439 |
},
|
| 1440 |
"language_model.model.layers.51.mlp.gate_proj": {
|
| 1441 |
"group_size": 64,
|
|
|
|
| 1451 |
},
|
| 1452 |
"language_model.model.layers.52.self_attn.q_proj": {
|
| 1453 |
"group_size": 64,
|
| 1454 |
+
"bits": 8
|
| 1455 |
},
|
| 1456 |
"language_model.model.layers.52.self_attn.k_proj": {
|
| 1457 |
"group_size": 64,
|
| 1458 |
+
"bits": 8
|
| 1459 |
},
|
| 1460 |
"language_model.model.layers.52.self_attn.v_proj": {
|
| 1461 |
"group_size": 64,
|
| 1462 |
+
"bits": 8
|
| 1463 |
},
|
| 1464 |
"language_model.model.layers.52.self_attn.o_proj": {
|
| 1465 |
"group_size": 64,
|
| 1466 |
+
"bits": 8
|
| 1467 |
},
|
| 1468 |
"language_model.model.layers.52.mlp.gate_proj": {
|
| 1469 |
"group_size": 64,
|
| 1470 |
+
"bits": 5
|
| 1471 |
},
|
| 1472 |
"language_model.model.layers.52.mlp.up_proj": {
|
| 1473 |
"group_size": 64,
|
| 1474 |
+
"bits": 5
|
| 1475 |
},
|
| 1476 |
"language_model.model.layers.52.mlp.down_proj": {
|
| 1477 |
"group_size": 64,
|
| 1478 |
+
"bits": 5
|
| 1479 |
},
|
| 1480 |
"language_model.model.layers.53.self_attn.q_proj": {
|
| 1481 |
"group_size": 64,
|
| 1482 |
+
"bits": 8
|
| 1483 |
},
|
| 1484 |
"language_model.model.layers.53.self_attn.k_proj": {
|
| 1485 |
"group_size": 64,
|
| 1486 |
+
"bits": 8
|
| 1487 |
},
|
| 1488 |
"language_model.model.layers.53.self_attn.o_proj": {
|
| 1489 |
"group_size": 64,
|
| 1490 |
+
"bits": 8
|
| 1491 |
},
|
| 1492 |
"language_model.model.layers.53.mlp.gate_proj": {
|
| 1493 |
"group_size": 64,
|
|
|
|
| 1503 |
},
|
| 1504 |
"language_model.model.layers.54.self_attn.q_proj": {
|
| 1505 |
"group_size": 64,
|
| 1506 |
+
"bits": 8
|
| 1507 |
},
|
| 1508 |
"language_model.model.layers.54.self_attn.k_proj": {
|
| 1509 |
"group_size": 64,
|
| 1510 |
+
"bits": 8
|
| 1511 |
},
|
| 1512 |
"language_model.model.layers.54.self_attn.v_proj": {
|
| 1513 |
"group_size": 64,
|
| 1514 |
+
"bits": 8
|
| 1515 |
},
|
| 1516 |
"language_model.model.layers.54.self_attn.o_proj": {
|
| 1517 |
"group_size": 64,
|
| 1518 |
+
"bits": 8
|
| 1519 |
},
|
| 1520 |
"language_model.model.layers.54.mlp.gate_proj": {
|
| 1521 |
"group_size": 64,
|
|
|
|
| 1531 |
},
|
| 1532 |
"language_model.model.layers.55.self_attn.q_proj": {
|
| 1533 |
"group_size": 64,
|
| 1534 |
+
"bits": 8
|
| 1535 |
},
|
| 1536 |
"language_model.model.layers.55.self_attn.k_proj": {
|
| 1537 |
"group_size": 64,
|
| 1538 |
+
"bits": 8
|
| 1539 |
},
|
| 1540 |
"language_model.model.layers.55.self_attn.v_proj": {
|
| 1541 |
"group_size": 64,
|
| 1542 |
+
"bits": 8
|
| 1543 |
},
|
| 1544 |
"language_model.model.layers.55.self_attn.o_proj": {
|
| 1545 |
"group_size": 64,
|
| 1546 |
+
"bits": 8
|
| 1547 |
},
|
| 1548 |
"language_model.model.layers.55.mlp.gate_proj": {
|
| 1549 |
"group_size": 64,
|
|
|
|
| 1559 |
},
|
| 1560 |
"language_model.model.layers.56.self_attn.q_proj": {
|
| 1561 |
"group_size": 64,
|
| 1562 |
+
"bits": 8
|
| 1563 |
},
|
| 1564 |
"language_model.model.layers.56.self_attn.k_proj": {
|
| 1565 |
"group_size": 64,
|
| 1566 |
+
"bits": 8
|
| 1567 |
},
|
| 1568 |
"language_model.model.layers.56.self_attn.v_proj": {
|
| 1569 |
"group_size": 64,
|
| 1570 |
+
"bits": 8
|
| 1571 |
},
|
| 1572 |
"language_model.model.layers.56.self_attn.o_proj": {
|
| 1573 |
"group_size": 64,
|
| 1574 |
+
"bits": 8
|
| 1575 |
},
|
| 1576 |
"language_model.model.layers.56.mlp.gate_proj": {
|
| 1577 |
"group_size": 64,
|
|
|
|
| 1587 |
},
|
| 1588 |
"language_model.model.layers.57.self_attn.q_proj": {
|
| 1589 |
"group_size": 64,
|
| 1590 |
+
"bits": 8
|
| 1591 |
},
|
| 1592 |
"language_model.model.layers.57.self_attn.k_proj": {
|
| 1593 |
"group_size": 64,
|
| 1594 |
+
"bits": 8
|
| 1595 |
},
|
| 1596 |
"language_model.model.layers.57.self_attn.v_proj": {
|
| 1597 |
"group_size": 64,
|
| 1598 |
+
"bits": 8
|
| 1599 |
},
|
| 1600 |
"language_model.model.layers.57.self_attn.o_proj": {
|
| 1601 |
"group_size": 64,
|
| 1602 |
+
"bits": 8
|
| 1603 |
},
|
| 1604 |
"language_model.model.layers.57.mlp.gate_proj": {
|
| 1605 |
"group_size": 64,
|
| 1606 |
+
"bits": 5
|
| 1607 |
},
|
| 1608 |
"language_model.model.layers.57.mlp.up_proj": {
|
| 1609 |
"group_size": 64,
|
| 1610 |
+
"bits": 5
|
| 1611 |
},
|
| 1612 |
"language_model.model.layers.57.mlp.down_proj": {
|
| 1613 |
"group_size": 64,
|
| 1614 |
+
"bits": 5
|
| 1615 |
},
|
| 1616 |
"language_model.model.layers.58.self_attn.q_proj": {
|
| 1617 |
"group_size": 64,
|
| 1618 |
+
"bits": 8
|
| 1619 |
},
|
| 1620 |
"language_model.model.layers.58.self_attn.k_proj": {
|
| 1621 |
"group_size": 64,
|
| 1622 |
+
"bits": 8
|
| 1623 |
},
|
| 1624 |
"language_model.model.layers.58.self_attn.v_proj": {
|
| 1625 |
"group_size": 64,
|
| 1626 |
+
"bits": 8
|
| 1627 |
},
|
| 1628 |
"language_model.model.layers.58.self_attn.o_proj": {
|
| 1629 |
"group_size": 64,
|
| 1630 |
+
"bits": 8
|
| 1631 |
},
|
| 1632 |
"language_model.model.layers.58.mlp.gate_proj": {
|
| 1633 |
"group_size": 64,
|
| 1634 |
+
"bits": 5
|
| 1635 |
},
|
| 1636 |
"language_model.model.layers.58.mlp.up_proj": {
|
| 1637 |
"group_size": 64,
|
| 1638 |
+
"bits": 5
|
| 1639 |
},
|
| 1640 |
"language_model.model.layers.58.mlp.down_proj": {
|
| 1641 |
"group_size": 64,
|
| 1642 |
+
"bits": 5
|
| 1643 |
},
|
| 1644 |
"language_model.model.layers.59.self_attn.q_proj": {
|
| 1645 |
"group_size": 64,
|
| 1646 |
+
"bits": 8
|
| 1647 |
},
|
| 1648 |
"language_model.model.layers.59.self_attn.k_proj": {
|
| 1649 |
"group_size": 64,
|
| 1650 |
+
"bits": 8
|
| 1651 |
},
|
| 1652 |
"language_model.model.layers.59.self_attn.o_proj": {
|
| 1653 |
"group_size": 64,
|
| 1654 |
+
"bits": 8
|
| 1655 |
},
|
| 1656 |
"language_model.model.layers.59.mlp.gate_proj": {
|
| 1657 |
"group_size": 64,
|
| 1658 |
+
"bits": 6
|
| 1659 |
},
|
| 1660 |
"language_model.model.layers.59.mlp.up_proj": {
|
| 1661 |
"group_size": 64,
|
| 1662 |
+
"bits": 6
|
| 1663 |
},
|
| 1664 |
"language_model.model.layers.59.mlp.down_proj": {
|
| 1665 |
"group_size": 64,
|
| 1666 |
+
"bits": 6
|
| 1667 |
}
|
| 1668 |
},
|
| 1669 |
"quantization_config": {
|
|
|
|
| 1671 |
"bits": 4,
|
| 1672 |
"language_model.model.embed_tokens": {
|
| 1673 |
"group_size": 64,
|
| 1674 |
+
"bits": 6
|
| 1675 |
},
|
| 1676 |
"language_model.model.layers.0.self_attn.q_proj": {
|
| 1677 |
"group_size": 64,
|
| 1678 |
+
"bits": 8
|
| 1679 |
},
|
| 1680 |
"language_model.model.layers.0.self_attn.k_proj": {
|
| 1681 |
"group_size": 64,
|
| 1682 |
+
"bits": 8
|
| 1683 |
},
|
| 1684 |
"language_model.model.layers.0.self_attn.v_proj": {
|
| 1685 |
"group_size": 64,
|
| 1686 |
+
"bits": 8
|
| 1687 |
},
|
| 1688 |
"language_model.model.layers.0.self_attn.o_proj": {
|
| 1689 |
"group_size": 64,
|
| 1690 |
+
"bits": 8
|
| 1691 |
},
|
| 1692 |
"language_model.model.layers.0.mlp.gate_proj": {
|
| 1693 |
"group_size": 64,
|
|
|
|
| 1703 |
},
|
| 1704 |
"language_model.model.layers.1.self_attn.q_proj": {
|
| 1705 |
"group_size": 64,
|
| 1706 |
+
"bits": 8
|
| 1707 |
},
|
| 1708 |
"language_model.model.layers.1.self_attn.k_proj": {
|
| 1709 |
"group_size": 64,
|
| 1710 |
+
"bits": 8
|
| 1711 |
},
|
| 1712 |
"language_model.model.layers.1.self_attn.v_proj": {
|
| 1713 |
"group_size": 64,
|
| 1714 |
+
"bits": 8
|
| 1715 |
},
|
| 1716 |
"language_model.model.layers.1.self_attn.o_proj": {
|
| 1717 |
"group_size": 64,
|
| 1718 |
+
"bits": 8
|
| 1719 |
},
|
| 1720 |
"language_model.model.layers.1.mlp.gate_proj": {
|
| 1721 |
"group_size": 64,
|
| 1722 |
+
"bits": 5
|
| 1723 |
},
|
| 1724 |
"language_model.model.layers.1.mlp.up_proj": {
|
| 1725 |
"group_size": 64,
|
| 1726 |
+
"bits": 5
|
| 1727 |
},
|
| 1728 |
"language_model.model.layers.1.mlp.down_proj": {
|
| 1729 |
"group_size": 64,
|
| 1730 |
+
"bits": 5
|
| 1731 |
},
|
| 1732 |
"language_model.model.layers.2.self_attn.q_proj": {
|
| 1733 |
"group_size": 64,
|
| 1734 |
+
"bits": 8
|
| 1735 |
},
|
| 1736 |
"language_model.model.layers.2.self_attn.k_proj": {
|
| 1737 |
"group_size": 64,
|
| 1738 |
+
"bits": 8
|
| 1739 |
},
|
| 1740 |
"language_model.model.layers.2.self_attn.v_proj": {
|
| 1741 |
"group_size": 64,
|
| 1742 |
+
"bits": 8
|
| 1743 |
},
|
| 1744 |
"language_model.model.layers.2.self_attn.o_proj": {
|
| 1745 |
"group_size": 64,
|
| 1746 |
+
"bits": 8
|
| 1747 |
},
|
| 1748 |
"language_model.model.layers.2.mlp.gate_proj": {
|
| 1749 |
"group_size": 64,
|
|
|
|
| 1759 |
},
|
| 1760 |
"language_model.model.layers.3.self_attn.q_proj": {
|
| 1761 |
"group_size": 64,
|
| 1762 |
+
"bits": 8
|
| 1763 |
},
|
| 1764 |
"language_model.model.layers.3.self_attn.k_proj": {
|
| 1765 |
"group_size": 64,
|
| 1766 |
+
"bits": 8
|
| 1767 |
},
|
| 1768 |
"language_model.model.layers.3.self_attn.v_proj": {
|
| 1769 |
"group_size": 64,
|
| 1770 |
+
"bits": 8
|
| 1771 |
},
|
| 1772 |
"language_model.model.layers.3.self_attn.o_proj": {
|
| 1773 |
"group_size": 64,
|
| 1774 |
+
"bits": 8
|
| 1775 |
},
|
| 1776 |
"language_model.model.layers.3.mlp.gate_proj": {
|
| 1777 |
"group_size": 64,
|
|
|
|
| 1787 |
},
|
| 1788 |
"language_model.model.layers.4.self_attn.q_proj": {
|
| 1789 |
"group_size": 64,
|
| 1790 |
+
"bits": 8
|
| 1791 |
},
|
| 1792 |
"language_model.model.layers.4.self_attn.k_proj": {
|
| 1793 |
"group_size": 64,
|
| 1794 |
+
"bits": 8
|
| 1795 |
},
|
| 1796 |
"language_model.model.layers.4.self_attn.v_proj": {
|
| 1797 |
"group_size": 64,
|
| 1798 |
+
"bits": 8
|
| 1799 |
},
|
| 1800 |
"language_model.model.layers.4.self_attn.o_proj": {
|
| 1801 |
"group_size": 64,
|
| 1802 |
+
"bits": 8
|
| 1803 |
},
|
| 1804 |
"language_model.model.layers.4.mlp.gate_proj": {
|
| 1805 |
"group_size": 64,
|
|
|
|
| 1815 |
},
|
| 1816 |
"language_model.model.layers.5.self_attn.q_proj": {
|
| 1817 |
"group_size": 64,
|
| 1818 |
+
"bits": 8
|
| 1819 |
},
|
| 1820 |
"language_model.model.layers.5.self_attn.k_proj": {
|
| 1821 |
"group_size": 64,
|
| 1822 |
+
"bits": 8
|
| 1823 |
},
|
| 1824 |
"language_model.model.layers.5.self_attn.o_proj": {
|
| 1825 |
"group_size": 64,
|
| 1826 |
+
"bits": 8
|
| 1827 |
},
|
| 1828 |
"language_model.model.layers.5.mlp.gate_proj": {
|
| 1829 |
"group_size": 64,
|
|
|
|
| 1839 |
},
|
| 1840 |
"language_model.model.layers.6.self_attn.q_proj": {
|
| 1841 |
"group_size": 64,
|
| 1842 |
+
"bits": 8
|
| 1843 |
},
|
| 1844 |
"language_model.model.layers.6.self_attn.k_proj": {
|
| 1845 |
"group_size": 64,
|
| 1846 |
+
"bits": 8
|
| 1847 |
},
|
| 1848 |
"language_model.model.layers.6.self_attn.v_proj": {
|
| 1849 |
"group_size": 64,
|
| 1850 |
+
"bits": 8
|
| 1851 |
},
|
| 1852 |
"language_model.model.layers.6.self_attn.o_proj": {
|
| 1853 |
"group_size": 64,
|
| 1854 |
+
"bits": 8
|
| 1855 |
},
|
| 1856 |
"language_model.model.layers.6.mlp.gate_proj": {
|
| 1857 |
"group_size": 64,
|
|
|
|
| 1867 |
},
|
| 1868 |
"language_model.model.layers.7.self_attn.q_proj": {
|
| 1869 |
"group_size": 64,
|
| 1870 |
+
"bits": 8
|
| 1871 |
},
|
| 1872 |
"language_model.model.layers.7.self_attn.k_proj": {
|
| 1873 |
"group_size": 64,
|
| 1874 |
+
"bits": 8
|
| 1875 |
},
|
| 1876 |
"language_model.model.layers.7.self_attn.v_proj": {
|
| 1877 |
"group_size": 64,
|
| 1878 |
+
"bits": 8
|
| 1879 |
},
|
| 1880 |
"language_model.model.layers.7.self_attn.o_proj": {
|
| 1881 |
"group_size": 64,
|
| 1882 |
+
"bits": 8
|
| 1883 |
},
|
| 1884 |
"language_model.model.layers.7.mlp.gate_proj": {
|
| 1885 |
"group_size": 64,
|
|
|
|
| 1895 |
},
|
| 1896 |
"language_model.model.layers.8.self_attn.q_proj": {
|
| 1897 |
"group_size": 64,
|
| 1898 |
+
"bits": 8
|
| 1899 |
},
|
| 1900 |
"language_model.model.layers.8.self_attn.k_proj": {
|
| 1901 |
"group_size": 64,
|
| 1902 |
+
"bits": 8
|
| 1903 |
},
|
| 1904 |
"language_model.model.layers.8.self_attn.v_proj": {
|
| 1905 |
"group_size": 64,
|
| 1906 |
+
"bits": 8
|
| 1907 |
},
|
| 1908 |
"language_model.model.layers.8.self_attn.o_proj": {
|
| 1909 |
"group_size": 64,
|
| 1910 |
+
"bits": 8
|
| 1911 |
},
|
| 1912 |
"language_model.model.layers.8.mlp.gate_proj": {
|
| 1913 |
"group_size": 64,
|
|
|
|
| 1923 |
},
|
| 1924 |
"language_model.model.layers.9.self_attn.q_proj": {
|
| 1925 |
"group_size": 64,
|
| 1926 |
+
"bits": 8
|
| 1927 |
},
|
| 1928 |
"language_model.model.layers.9.self_attn.k_proj": {
|
| 1929 |
"group_size": 64,
|
| 1930 |
+
"bits": 8
|
| 1931 |
},
|
| 1932 |
"language_model.model.layers.9.self_attn.v_proj": {
|
| 1933 |
"group_size": 64,
|
| 1934 |
+
"bits": 8
|
| 1935 |
},
|
| 1936 |
"language_model.model.layers.9.self_attn.o_proj": {
|
| 1937 |
"group_size": 64,
|
| 1938 |
+
"bits": 8
|
| 1939 |
},
|
| 1940 |
"language_model.model.layers.9.mlp.gate_proj": {
|
| 1941 |
"group_size": 64,
|
|
|
|
| 1951 |
},
|
| 1952 |
"language_model.model.layers.10.self_attn.q_proj": {
|
| 1953 |
"group_size": 64,
|
| 1954 |
+
"bits": 8
|
| 1955 |
},
|
| 1956 |
"language_model.model.layers.10.self_attn.k_proj": {
|
| 1957 |
"group_size": 64,
|
| 1958 |
+
"bits": 8
|
| 1959 |
},
|
| 1960 |
"language_model.model.layers.10.self_attn.v_proj": {
|
| 1961 |
"group_size": 64,
|
| 1962 |
+
"bits": 8
|
| 1963 |
},
|
| 1964 |
"language_model.model.layers.10.self_attn.o_proj": {
|
| 1965 |
"group_size": 64,
|
| 1966 |
+
"bits": 8
|
| 1967 |
},
|
| 1968 |
"language_model.model.layers.10.mlp.gate_proj": {
|
| 1969 |
"group_size": 64,
|
|
|
|
| 1979 |
},
|
| 1980 |
"language_model.model.layers.11.self_attn.q_proj": {
|
| 1981 |
"group_size": 64,
|
| 1982 |
+
"bits": 8
|
| 1983 |
},
|
| 1984 |
"language_model.model.layers.11.self_attn.k_proj": {
|
| 1985 |
"group_size": 64,
|
| 1986 |
+
"bits": 8
|
| 1987 |
},
|
| 1988 |
"language_model.model.layers.11.self_attn.o_proj": {
|
| 1989 |
"group_size": 64,
|
| 1990 |
+
"bits": 8
|
| 1991 |
},
|
| 1992 |
"language_model.model.layers.11.mlp.gate_proj": {
|
| 1993 |
"group_size": 64,
|
|
|
|
| 2003 |
},
|
| 2004 |
"language_model.model.layers.12.self_attn.q_proj": {
|
| 2005 |
"group_size": 64,
|
| 2006 |
+
"bits": 8
|
| 2007 |
},
|
| 2008 |
"language_model.model.layers.12.self_attn.k_proj": {
|
| 2009 |
"group_size": 64,
|
| 2010 |
+
"bits": 8
|
| 2011 |
},
|
| 2012 |
"language_model.model.layers.12.self_attn.v_proj": {
|
| 2013 |
"group_size": 64,
|
| 2014 |
+
"bits": 8
|
| 2015 |
},
|
| 2016 |
"language_model.model.layers.12.self_attn.o_proj": {
|
| 2017 |
"group_size": 64,
|
| 2018 |
+
"bits": 8
|
| 2019 |
},
|
| 2020 |
"language_model.model.layers.12.mlp.gate_proj": {
|
| 2021 |
"group_size": 64,
|
|
|
|
| 2031 |
},
|
| 2032 |
"language_model.model.layers.13.self_attn.q_proj": {
|
| 2033 |
"group_size": 64,
|
| 2034 |
+
"bits": 8
|
| 2035 |
},
|
| 2036 |
"language_model.model.layers.13.self_attn.k_proj": {
|
| 2037 |
"group_size": 64,
|
| 2038 |
+
"bits": 8
|
| 2039 |
},
|
| 2040 |
"language_model.model.layers.13.self_attn.v_proj": {
|
| 2041 |
"group_size": 64,
|
| 2042 |
+
"bits": 8
|
| 2043 |
},
|
| 2044 |
"language_model.model.layers.13.self_attn.o_proj": {
|
| 2045 |
"group_size": 64,
|
| 2046 |
+
"bits": 8
|
| 2047 |
},
|
| 2048 |
"language_model.model.layers.13.mlp.gate_proj": {
|
| 2049 |
"group_size": 64,
|
|
|
|
| 2059 |
},
|
| 2060 |
"language_model.model.layers.14.self_attn.q_proj": {
|
| 2061 |
"group_size": 64,
|
| 2062 |
+
"bits": 8
|
| 2063 |
},
|
| 2064 |
"language_model.model.layers.14.self_attn.k_proj": {
|
| 2065 |
"group_size": 64,
|
| 2066 |
+
"bits": 8
|
| 2067 |
},
|
| 2068 |
"language_model.model.layers.14.self_attn.v_proj": {
|
| 2069 |
"group_size": 64,
|
| 2070 |
+
"bits": 8
|
| 2071 |
},
|
| 2072 |
"language_model.model.layers.14.self_attn.o_proj": {
|
| 2073 |
"group_size": 64,
|
| 2074 |
+
"bits": 8
|
| 2075 |
},
|
| 2076 |
"language_model.model.layers.14.mlp.gate_proj": {
|
| 2077 |
"group_size": 64,
|
|
|
|
| 2087 |
},
|
| 2088 |
"language_model.model.layers.15.self_attn.q_proj": {
|
| 2089 |
"group_size": 64,
|
| 2090 |
+
"bits": 8
|
| 2091 |
},
|
| 2092 |
"language_model.model.layers.15.self_attn.k_proj": {
|
| 2093 |
"group_size": 64,
|
| 2094 |
+
"bits": 8
|
| 2095 |
},
|
| 2096 |
"language_model.model.layers.15.self_attn.v_proj": {
|
| 2097 |
"group_size": 64,
|
| 2098 |
+
"bits": 8
|
| 2099 |
},
|
| 2100 |
"language_model.model.layers.15.self_attn.o_proj": {
|
| 2101 |
"group_size": 64,
|
| 2102 |
+
"bits": 8
|
| 2103 |
},
|
| 2104 |
"language_model.model.layers.15.mlp.gate_proj": {
|
| 2105 |
"group_size": 64,
|
|
|
|
| 2115 |
},
|
| 2116 |
"language_model.model.layers.16.self_attn.q_proj": {
|
| 2117 |
"group_size": 64,
|
| 2118 |
+
"bits": 8
|
| 2119 |
},
|
| 2120 |
"language_model.model.layers.16.self_attn.k_proj": {
|
| 2121 |
"group_size": 64,
|
| 2122 |
+
"bits": 8
|
| 2123 |
},
|
| 2124 |
"language_model.model.layers.16.self_attn.v_proj": {
|
| 2125 |
"group_size": 64,
|
| 2126 |
+
"bits": 8
|
| 2127 |
},
|
| 2128 |
"language_model.model.layers.16.self_attn.o_proj": {
|
| 2129 |
"group_size": 64,
|
| 2130 |
+
"bits": 8
|
| 2131 |
},
|
| 2132 |
"language_model.model.layers.16.mlp.gate_proj": {
|
| 2133 |
"group_size": 64,
|
|
|
|
| 2143 |
},
|
| 2144 |
"language_model.model.layers.17.self_attn.q_proj": {
|
| 2145 |
"group_size": 64,
|
| 2146 |
+
"bits": 8
|
| 2147 |
},
|
| 2148 |
"language_model.model.layers.17.self_attn.k_proj": {
|
| 2149 |
"group_size": 64,
|
| 2150 |
+
"bits": 8
|
| 2151 |
},
|
| 2152 |
"language_model.model.layers.17.self_attn.o_proj": {
|
| 2153 |
"group_size": 64,
|
| 2154 |
+
"bits": 8
|
| 2155 |
},
|
| 2156 |
"language_model.model.layers.17.mlp.gate_proj": {
|
| 2157 |
"group_size": 64,
|
|
|
|
| 2167 |
},
|
| 2168 |
"language_model.model.layers.18.self_attn.q_proj": {
|
| 2169 |
"group_size": 64,
|
| 2170 |
+
"bits": 8
|
| 2171 |
},
|
| 2172 |
"language_model.model.layers.18.self_attn.k_proj": {
|
| 2173 |
"group_size": 64,
|
| 2174 |
+
"bits": 8
|
| 2175 |
},
|
| 2176 |
"language_model.model.layers.18.self_attn.v_proj": {
|
| 2177 |
"group_size": 64,
|
| 2178 |
+
"bits": 8
|
| 2179 |
},
|
| 2180 |
"language_model.model.layers.18.self_attn.o_proj": {
|
| 2181 |
"group_size": 64,
|
| 2182 |
+
"bits": 8
|
| 2183 |
},
|
| 2184 |
"language_model.model.layers.18.mlp.gate_proj": {
|
| 2185 |
"group_size": 64,
|
|
|
|
| 2195 |
},
|
| 2196 |
"language_model.model.layers.19.self_attn.q_proj": {
|
| 2197 |
"group_size": 64,
|
| 2198 |
+
"bits": 8
|
| 2199 |
},
|
| 2200 |
"language_model.model.layers.19.self_attn.k_proj": {
|
| 2201 |
"group_size": 64,
|
| 2202 |
+
"bits": 8
|
| 2203 |
},
|
| 2204 |
"language_model.model.layers.19.self_attn.v_proj": {
|
| 2205 |
"group_size": 64,
|
| 2206 |
+
"bits": 8
|
| 2207 |
},
|
| 2208 |
"language_model.model.layers.19.self_attn.o_proj": {
|
| 2209 |
"group_size": 64,
|
| 2210 |
+
"bits": 8
|
| 2211 |
},
|
| 2212 |
"language_model.model.layers.19.mlp.gate_proj": {
|
| 2213 |
"group_size": 64,
|
|
|
|
| 2223 |
},
|
| 2224 |
"language_model.model.layers.20.self_attn.q_proj": {
|
| 2225 |
"group_size": 64,
|
| 2226 |
+
"bits": 8
|
| 2227 |
},
|
| 2228 |
"language_model.model.layers.20.self_attn.k_proj": {
|
| 2229 |
"group_size": 64,
|
| 2230 |
+
"bits": 8
|
| 2231 |
},
|
| 2232 |
"language_model.model.layers.20.self_attn.v_proj": {
|
| 2233 |
"group_size": 64,
|
| 2234 |
+
"bits": 8
|
| 2235 |
},
|
| 2236 |
"language_model.model.layers.20.self_attn.o_proj": {
|
| 2237 |
"group_size": 64,
|
| 2238 |
+
"bits": 8
|
| 2239 |
},
|
| 2240 |
"language_model.model.layers.20.mlp.gate_proj": {
|
| 2241 |
"group_size": 64,
|
|
|
|
| 2251 |
},
|
| 2252 |
"language_model.model.layers.21.self_attn.q_proj": {
|
| 2253 |
"group_size": 64,
|
| 2254 |
+
"bits": 8
|
| 2255 |
},
|
| 2256 |
"language_model.model.layers.21.self_attn.k_proj": {
|
| 2257 |
"group_size": 64,
|
| 2258 |
+
"bits": 8
|
| 2259 |
},
|
| 2260 |
"language_model.model.layers.21.self_attn.v_proj": {
|
| 2261 |
"group_size": 64,
|
| 2262 |
+
"bits": 8
|
| 2263 |
},
|
| 2264 |
"language_model.model.layers.21.self_attn.o_proj": {
|
| 2265 |
"group_size": 64,
|
| 2266 |
+
"bits": 8
|
| 2267 |
},
|
| 2268 |
"language_model.model.layers.21.mlp.gate_proj": {
|
| 2269 |
"group_size": 64,
|
|
|
|
| 2279 |
},
|
| 2280 |
"language_model.model.layers.22.self_attn.q_proj": {
|
| 2281 |
"group_size": 64,
|
| 2282 |
+
"bits": 8
|
| 2283 |
},
|
| 2284 |
"language_model.model.layers.22.self_attn.k_proj": {
|
| 2285 |
"group_size": 64,
|
| 2286 |
+
"bits": 8
|
| 2287 |
},
|
| 2288 |
"language_model.model.layers.22.self_attn.v_proj": {
|
| 2289 |
"group_size": 64,
|
| 2290 |
+
"bits": 8
|
| 2291 |
},
|
| 2292 |
"language_model.model.layers.22.self_attn.o_proj": {
|
| 2293 |
"group_size": 64,
|
| 2294 |
+
"bits": 8
|
| 2295 |
},
|
| 2296 |
"language_model.model.layers.22.mlp.gate_proj": {
|
| 2297 |
"group_size": 64,
|
|
|
|
| 2307 |
},
|
| 2308 |
"language_model.model.layers.23.self_attn.q_proj": {
|
| 2309 |
"group_size": 64,
|
| 2310 |
+
"bits": 8
|
| 2311 |
},
|
| 2312 |
"language_model.model.layers.23.self_attn.k_proj": {
|
| 2313 |
"group_size": 64,
|
| 2314 |
+
"bits": 8
|
| 2315 |
},
|
| 2316 |
"language_model.model.layers.23.self_attn.o_proj": {
|
| 2317 |
"group_size": 64,
|
| 2318 |
+
"bits": 8
|
| 2319 |
},
|
| 2320 |
"language_model.model.layers.23.mlp.gate_proj": {
|
| 2321 |
"group_size": 64,
|
|
|
|
| 2331 |
},
|
| 2332 |
"language_model.model.layers.24.self_attn.q_proj": {
|
| 2333 |
"group_size": 64,
|
| 2334 |
+
"bits": 8
|
| 2335 |
},
|
| 2336 |
"language_model.model.layers.24.self_attn.k_proj": {
|
| 2337 |
"group_size": 64,
|
| 2338 |
+
"bits": 8
|
| 2339 |
},
|
| 2340 |
"language_model.model.layers.24.self_attn.v_proj": {
|
| 2341 |
"group_size": 64,
|
| 2342 |
+
"bits": 8
|
| 2343 |
},
|
| 2344 |
"language_model.model.layers.24.self_attn.o_proj": {
|
| 2345 |
"group_size": 64,
|
| 2346 |
+
"bits": 8
|
| 2347 |
},
|
| 2348 |
"language_model.model.layers.24.mlp.gate_proj": {
|
| 2349 |
"group_size": 64,
|
|
|
|
| 2359 |
},
|
| 2360 |
"language_model.model.layers.25.self_attn.q_proj": {
|
| 2361 |
"group_size": 64,
|
| 2362 |
+
"bits": 8
|
| 2363 |
},
|
| 2364 |
"language_model.model.layers.25.self_attn.k_proj": {
|
| 2365 |
"group_size": 64,
|
| 2366 |
+
"bits": 8
|
| 2367 |
},
|
| 2368 |
"language_model.model.layers.25.self_attn.v_proj": {
|
| 2369 |
"group_size": 64,
|
| 2370 |
+
"bits": 8
|
| 2371 |
},
|
| 2372 |
"language_model.model.layers.25.self_attn.o_proj": {
|
| 2373 |
"group_size": 64,
|
| 2374 |
+
"bits": 8
|
| 2375 |
},
|
| 2376 |
"language_model.model.layers.25.mlp.gate_proj": {
|
| 2377 |
"group_size": 64,
|
|
|
|
| 2387 |
},
|
| 2388 |
"language_model.model.layers.26.self_attn.q_proj": {
|
| 2389 |
"group_size": 64,
|
| 2390 |
+
"bits": 8
|
| 2391 |
},
|
| 2392 |
"language_model.model.layers.26.self_attn.k_proj": {
|
| 2393 |
"group_size": 64,
|
| 2394 |
+
"bits": 8
|
| 2395 |
},
|
| 2396 |
"language_model.model.layers.26.self_attn.v_proj": {
|
| 2397 |
"group_size": 64,
|
| 2398 |
+
"bits": 8
|
| 2399 |
},
|
| 2400 |
"language_model.model.layers.26.self_attn.o_proj": {
|
| 2401 |
"group_size": 64,
|
| 2402 |
+
"bits": 8
|
| 2403 |
},
|
| 2404 |
"language_model.model.layers.26.mlp.gate_proj": {
|
| 2405 |
"group_size": 64,
|
|
|
|
| 2415 |
},
|
| 2416 |
"language_model.model.layers.27.self_attn.q_proj": {
|
| 2417 |
"group_size": 64,
|
| 2418 |
+
"bits": 8
|
| 2419 |
},
|
| 2420 |
"language_model.model.layers.27.self_attn.k_proj": {
|
| 2421 |
"group_size": 64,
|
| 2422 |
+
"bits": 8
|
| 2423 |
},
|
| 2424 |
"language_model.model.layers.27.self_attn.v_proj": {
|
| 2425 |
"group_size": 64,
|
| 2426 |
+
"bits": 8
|
| 2427 |
},
|
| 2428 |
"language_model.model.layers.27.self_attn.o_proj": {
|
| 2429 |
"group_size": 64,
|
| 2430 |
+
"bits": 8
|
| 2431 |
},
|
| 2432 |
"language_model.model.layers.27.mlp.gate_proj": {
|
| 2433 |
"group_size": 64,
|
|
|
|
| 2443 |
},
|
| 2444 |
"language_model.model.layers.28.self_attn.q_proj": {
|
| 2445 |
"group_size": 64,
|
| 2446 |
+
"bits": 8
|
| 2447 |
},
|
| 2448 |
"language_model.model.layers.28.self_attn.k_proj": {
|
| 2449 |
"group_size": 64,
|
| 2450 |
+
"bits": 8
|
| 2451 |
},
|
| 2452 |
"language_model.model.layers.28.self_attn.v_proj": {
|
| 2453 |
"group_size": 64,
|
| 2454 |
+
"bits": 8
|
| 2455 |
},
|
| 2456 |
"language_model.model.layers.28.self_attn.o_proj": {
|
| 2457 |
"group_size": 64,
|
| 2458 |
+
"bits": 8
|
| 2459 |
},
|
| 2460 |
"language_model.model.layers.28.mlp.gate_proj": {
|
| 2461 |
"group_size": 64,
|
|
|
|
| 2471 |
},
|
| 2472 |
"language_model.model.layers.29.self_attn.q_proj": {
|
| 2473 |
"group_size": 64,
|
| 2474 |
+
"bits": 8
|
| 2475 |
},
|
| 2476 |
"language_model.model.layers.29.self_attn.k_proj": {
|
| 2477 |
"group_size": 64,
|
| 2478 |
+
"bits": 8
|
| 2479 |
},
|
| 2480 |
"language_model.model.layers.29.self_attn.o_proj": {
|
| 2481 |
"group_size": 64,
|
| 2482 |
+
"bits": 8
|
| 2483 |
},
|
| 2484 |
"language_model.model.layers.29.mlp.gate_proj": {
|
| 2485 |
"group_size": 64,
|
|
|
|
| 2495 |
},
|
| 2496 |
"language_model.model.layers.30.self_attn.q_proj": {
|
| 2497 |
"group_size": 64,
|
| 2498 |
+
"bits": 8
|
| 2499 |
},
|
| 2500 |
"language_model.model.layers.30.self_attn.k_proj": {
|
| 2501 |
"group_size": 64,
|
| 2502 |
+
"bits": 8
|
| 2503 |
},
|
| 2504 |
"language_model.model.layers.30.self_attn.v_proj": {
|
| 2505 |
"group_size": 64,
|
| 2506 |
+
"bits": 8
|
| 2507 |
},
|
| 2508 |
"language_model.model.layers.30.self_attn.o_proj": {
|
| 2509 |
"group_size": 64,
|
| 2510 |
+
"bits": 8
|
| 2511 |
},
|
| 2512 |
"language_model.model.layers.30.mlp.gate_proj": {
|
| 2513 |
"group_size": 64,
|
|
|
|
| 2523 |
},
|
| 2524 |
"language_model.model.layers.31.self_attn.q_proj": {
|
| 2525 |
"group_size": 64,
|
| 2526 |
+
"bits": 8
|
| 2527 |
},
|
| 2528 |
"language_model.model.layers.31.self_attn.k_proj": {
|
| 2529 |
"group_size": 64,
|
| 2530 |
+
"bits": 8
|
| 2531 |
},
|
| 2532 |
"language_model.model.layers.31.self_attn.v_proj": {
|
| 2533 |
"group_size": 64,
|
| 2534 |
+
"bits": 8
|
| 2535 |
},
|
| 2536 |
"language_model.model.layers.31.self_attn.o_proj": {
|
| 2537 |
"group_size": 64,
|
| 2538 |
+
"bits": 8
|
| 2539 |
},
|
| 2540 |
"language_model.model.layers.31.mlp.gate_proj": {
|
| 2541 |
"group_size": 64,
|
|
|
|
| 2551 |
},
|
| 2552 |
"language_model.model.layers.32.self_attn.q_proj": {
|
| 2553 |
"group_size": 64,
|
| 2554 |
+
"bits": 8
|
| 2555 |
},
|
| 2556 |
"language_model.model.layers.32.self_attn.k_proj": {
|
| 2557 |
"group_size": 64,
|
| 2558 |
+
"bits": 8
|
| 2559 |
},
|
| 2560 |
"language_model.model.layers.32.self_attn.v_proj": {
|
| 2561 |
"group_size": 64,
|
| 2562 |
+
"bits": 8
|
| 2563 |
},
|
| 2564 |
"language_model.model.layers.32.self_attn.o_proj": {
|
| 2565 |
"group_size": 64,
|
| 2566 |
+
"bits": 8
|
| 2567 |
},
|
| 2568 |
"language_model.model.layers.32.mlp.gate_proj": {
|
| 2569 |
"group_size": 64,
|
|
|
|
| 2579 |
},
|
| 2580 |
"language_model.model.layers.33.self_attn.q_proj": {
|
| 2581 |
"group_size": 64,
|
| 2582 |
+
"bits": 8
|
| 2583 |
},
|
| 2584 |
"language_model.model.layers.33.self_attn.k_proj": {
|
| 2585 |
"group_size": 64,
|
| 2586 |
+
"bits": 8
|
| 2587 |
},
|
| 2588 |
"language_model.model.layers.33.self_attn.v_proj": {
|
| 2589 |
"group_size": 64,
|
| 2590 |
+
"bits": 8
|
| 2591 |
},
|
| 2592 |
"language_model.model.layers.33.self_attn.o_proj": {
|
| 2593 |
"group_size": 64,
|
| 2594 |
+
"bits": 8
|
| 2595 |
},
|
| 2596 |
"language_model.model.layers.33.mlp.gate_proj": {
|
| 2597 |
"group_size": 64,
|
|
|
|
| 2607 |
},
|
| 2608 |
"language_model.model.layers.34.self_attn.q_proj": {
|
| 2609 |
"group_size": 64,
|
| 2610 |
+
"bits": 8
|
| 2611 |
},
|
| 2612 |
"language_model.model.layers.34.self_attn.k_proj": {
|
| 2613 |
"group_size": 64,
|
| 2614 |
+
"bits": 8
|
| 2615 |
},
|
| 2616 |
"language_model.model.layers.34.self_attn.v_proj": {
|
| 2617 |
"group_size": 64,
|
| 2618 |
+
"bits": 8
|
| 2619 |
},
|
| 2620 |
"language_model.model.layers.34.self_attn.o_proj": {
|
| 2621 |
"group_size": 64,
|
| 2622 |
+
"bits": 8
|
| 2623 |
},
|
| 2624 |
"language_model.model.layers.34.mlp.gate_proj": {
|
| 2625 |
"group_size": 64,
|
|
|
|
| 2635 |
},
|
| 2636 |
"language_model.model.layers.35.self_attn.q_proj": {
|
| 2637 |
"group_size": 64,
|
| 2638 |
+
"bits": 8
|
| 2639 |
},
|
| 2640 |
"language_model.model.layers.35.self_attn.k_proj": {
|
| 2641 |
"group_size": 64,
|
| 2642 |
+
"bits": 8
|
| 2643 |
},
|
| 2644 |
"language_model.model.layers.35.self_attn.o_proj": {
|
| 2645 |
"group_size": 64,
|
| 2646 |
+
"bits": 8
|
| 2647 |
},
|
| 2648 |
"language_model.model.layers.35.mlp.gate_proj": {
|
| 2649 |
"group_size": 64,
|
|
|
|
| 2659 |
},
|
| 2660 |
"language_model.model.layers.36.self_attn.q_proj": {
|
| 2661 |
"group_size": 64,
|
| 2662 |
+
"bits": 8
|
| 2663 |
},
|
| 2664 |
"language_model.model.layers.36.self_attn.k_proj": {
|
| 2665 |
"group_size": 64,
|
| 2666 |
+
"bits": 8
|
| 2667 |
},
|
| 2668 |
"language_model.model.layers.36.self_attn.v_proj": {
|
| 2669 |
"group_size": 64,
|
| 2670 |
+
"bits": 8
|
| 2671 |
},
|
| 2672 |
"language_model.model.layers.36.self_attn.o_proj": {
|
| 2673 |
"group_size": 64,
|
| 2674 |
+
"bits": 8
|
| 2675 |
},
|
| 2676 |
"language_model.model.layers.36.mlp.gate_proj": {
|
| 2677 |
"group_size": 64,
|
|
|
|
| 2687 |
},
|
| 2688 |
"language_model.model.layers.37.self_attn.q_proj": {
|
| 2689 |
"group_size": 64,
|
| 2690 |
+
"bits": 8
|
| 2691 |
},
|
| 2692 |
"language_model.model.layers.37.self_attn.k_proj": {
|
| 2693 |
"group_size": 64,
|
| 2694 |
+
"bits": 8
|
| 2695 |
},
|
| 2696 |
"language_model.model.layers.37.self_attn.v_proj": {
|
| 2697 |
"group_size": 64,
|
| 2698 |
+
"bits": 8
|
| 2699 |
},
|
| 2700 |
"language_model.model.layers.37.self_attn.o_proj": {
|
| 2701 |
"group_size": 64,
|
| 2702 |
+
"bits": 8
|
| 2703 |
},
|
| 2704 |
"language_model.model.layers.37.mlp.gate_proj": {
|
| 2705 |
"group_size": 64,
|
|
|
|
| 2715 |
},
|
| 2716 |
"language_model.model.layers.38.self_attn.q_proj": {
|
| 2717 |
"group_size": 64,
|
| 2718 |
+
"bits": 8
|
| 2719 |
},
|
| 2720 |
"language_model.model.layers.38.self_attn.k_proj": {
|
| 2721 |
"group_size": 64,
|
| 2722 |
+
"bits": 8
|
| 2723 |
},
|
| 2724 |
"language_model.model.layers.38.self_attn.v_proj": {
|
| 2725 |
"group_size": 64,
|
| 2726 |
+
"bits": 8
|
| 2727 |
},
|
| 2728 |
"language_model.model.layers.38.self_attn.o_proj": {
|
| 2729 |
"group_size": 64,
|
| 2730 |
+
"bits": 8
|
| 2731 |
},
|
| 2732 |
"language_model.model.layers.38.mlp.gate_proj": {
|
| 2733 |
"group_size": 64,
|
|
|
|
| 2743 |
},
|
| 2744 |
"language_model.model.layers.39.self_attn.q_proj": {
|
| 2745 |
"group_size": 64,
|
| 2746 |
+
"bits": 8
|
| 2747 |
},
|
| 2748 |
"language_model.model.layers.39.self_attn.k_proj": {
|
| 2749 |
"group_size": 64,
|
| 2750 |
+
"bits": 8
|
| 2751 |
},
|
| 2752 |
"language_model.model.layers.39.self_attn.v_proj": {
|
| 2753 |
"group_size": 64,
|
| 2754 |
+
"bits": 8
|
| 2755 |
},
|
| 2756 |
"language_model.model.layers.39.self_attn.o_proj": {
|
| 2757 |
"group_size": 64,
|
| 2758 |
+
"bits": 8
|
| 2759 |
},
|
| 2760 |
"language_model.model.layers.39.mlp.gate_proj": {
|
| 2761 |
"group_size": 64,
|
|
|
|
| 2771 |
},
|
| 2772 |
"language_model.model.layers.40.self_attn.q_proj": {
|
| 2773 |
"group_size": 64,
|
| 2774 |
+
"bits": 8
|
| 2775 |
},
|
| 2776 |
"language_model.model.layers.40.self_attn.k_proj": {
|
| 2777 |
"group_size": 64,
|
| 2778 |
+
"bits": 8
|
| 2779 |
},
|
| 2780 |
"language_model.model.layers.40.self_attn.v_proj": {
|
| 2781 |
"group_size": 64,
|
| 2782 |
+
"bits": 8
|
| 2783 |
},
|
| 2784 |
"language_model.model.layers.40.self_attn.o_proj": {
|
| 2785 |
"group_size": 64,
|
| 2786 |
+
"bits": 8
|
| 2787 |
},
|
| 2788 |
"language_model.model.layers.40.mlp.gate_proj": {
|
| 2789 |
"group_size": 64,
|
|
|
|
| 2799 |
},
|
| 2800 |
"language_model.model.layers.41.self_attn.q_proj": {
|
| 2801 |
"group_size": 64,
|
| 2802 |
+
"bits": 8
|
| 2803 |
},
|
| 2804 |
"language_model.model.layers.41.self_attn.k_proj": {
|
| 2805 |
"group_size": 64,
|
| 2806 |
+
"bits": 8
|
| 2807 |
},
|
| 2808 |
"language_model.model.layers.41.self_attn.o_proj": {
|
| 2809 |
"group_size": 64,
|
| 2810 |
+
"bits": 8
|
| 2811 |
},
|
| 2812 |
"language_model.model.layers.41.mlp.gate_proj": {
|
| 2813 |
"group_size": 64,
|
|
|
|
| 2823 |
},
|
| 2824 |
"language_model.model.layers.42.self_attn.q_proj": {
|
| 2825 |
"group_size": 64,
|
| 2826 |
+
"bits": 8
|
| 2827 |
},
|
| 2828 |
"language_model.model.layers.42.self_attn.k_proj": {
|
| 2829 |
"group_size": 64,
|
| 2830 |
+
"bits": 8
|
| 2831 |
},
|
| 2832 |
"language_model.model.layers.42.self_attn.v_proj": {
|
| 2833 |
"group_size": 64,
|
| 2834 |
+
"bits": 8
|
| 2835 |
},
|
| 2836 |
"language_model.model.layers.42.self_attn.o_proj": {
|
| 2837 |
"group_size": 64,
|
| 2838 |
+
"bits": 8
|
| 2839 |
},
|
| 2840 |
"language_model.model.layers.42.mlp.gate_proj": {
|
| 2841 |
"group_size": 64,
|
|
|
|
| 2851 |
},
|
| 2852 |
"language_model.model.layers.43.self_attn.q_proj": {
|
| 2853 |
"group_size": 64,
|
| 2854 |
+
"bits": 8
|
| 2855 |
},
|
| 2856 |
"language_model.model.layers.43.self_attn.k_proj": {
|
| 2857 |
"group_size": 64,
|
| 2858 |
+
"bits": 8
|
| 2859 |
},
|
| 2860 |
"language_model.model.layers.43.self_attn.v_proj": {
|
| 2861 |
"group_size": 64,
|
| 2862 |
+
"bits": 8
|
| 2863 |
},
|
| 2864 |
"language_model.model.layers.43.self_attn.o_proj": {
|
| 2865 |
"group_size": 64,
|
| 2866 |
+
"bits": 8
|
| 2867 |
},
|
| 2868 |
"language_model.model.layers.43.mlp.gate_proj": {
|
| 2869 |
"group_size": 64,
|
|
|
|
| 2879 |
},
|
| 2880 |
"language_model.model.layers.44.self_attn.q_proj": {
|
| 2881 |
"group_size": 64,
|
| 2882 |
+
"bits": 8
|
| 2883 |
},
|
| 2884 |
"language_model.model.layers.44.self_attn.k_proj": {
|
| 2885 |
"group_size": 64,
|
| 2886 |
+
"bits": 8
|
| 2887 |
},
|
| 2888 |
"language_model.model.layers.44.self_attn.v_proj": {
|
| 2889 |
"group_size": 64,
|
| 2890 |
+
"bits": 8
|
| 2891 |
},
|
| 2892 |
"language_model.model.layers.44.self_attn.o_proj": {
|
| 2893 |
"group_size": 64,
|
| 2894 |
+
"bits": 8
|
| 2895 |
},
|
| 2896 |
"language_model.model.layers.44.mlp.gate_proj": {
|
| 2897 |
"group_size": 64,
|
|
|
|
| 2907 |
},
|
| 2908 |
"language_model.model.layers.45.self_attn.q_proj": {
|
| 2909 |
"group_size": 64,
|
| 2910 |
+
"bits": 8
|
| 2911 |
},
|
| 2912 |
"language_model.model.layers.45.self_attn.k_proj": {
|
| 2913 |
"group_size": 64,
|
| 2914 |
+
"bits": 8
|
| 2915 |
},
|
| 2916 |
"language_model.model.layers.45.self_attn.v_proj": {
|
| 2917 |
"group_size": 64,
|
| 2918 |
+
"bits": 8
|
| 2919 |
},
|
| 2920 |
"language_model.model.layers.45.self_attn.o_proj": {
|
| 2921 |
"group_size": 64,
|
| 2922 |
+
"bits": 8
|
| 2923 |
},
|
| 2924 |
"language_model.model.layers.45.mlp.gate_proj": {
|
| 2925 |
"group_size": 64,
|
|
|
|
| 2935 |
},
|
| 2936 |
"language_model.model.layers.46.self_attn.q_proj": {
|
| 2937 |
"group_size": 64,
|
| 2938 |
+
"bits": 8
|
| 2939 |
},
|
| 2940 |
"language_model.model.layers.46.self_attn.k_proj": {
|
| 2941 |
"group_size": 64,
|
| 2942 |
+
"bits": 8
|
| 2943 |
},
|
| 2944 |
"language_model.model.layers.46.self_attn.v_proj": {
|
| 2945 |
"group_size": 64,
|
| 2946 |
+
"bits": 8
|
| 2947 |
},
|
| 2948 |
"language_model.model.layers.46.self_attn.o_proj": {
|
| 2949 |
"group_size": 64,
|
| 2950 |
+
"bits": 8
|
| 2951 |
},
|
| 2952 |
"language_model.model.layers.46.mlp.gate_proj": {
|
| 2953 |
"group_size": 64,
|
|
|
|
| 2963 |
},
|
| 2964 |
"language_model.model.layers.47.self_attn.q_proj": {
|
| 2965 |
"group_size": 64,
|
| 2966 |
+
"bits": 8
|
| 2967 |
},
|
| 2968 |
"language_model.model.layers.47.self_attn.k_proj": {
|
| 2969 |
"group_size": 64,
|
| 2970 |
+
"bits": 8
|
| 2971 |
},
|
| 2972 |
"language_model.model.layers.47.self_attn.o_proj": {
|
| 2973 |
"group_size": 64,
|
| 2974 |
+
"bits": 8
|
| 2975 |
},
|
| 2976 |
"language_model.model.layers.47.mlp.gate_proj": {
|
| 2977 |
"group_size": 64,
|
|
|
|
| 2987 |
},
|
| 2988 |
"language_model.model.layers.48.self_attn.q_proj": {
|
| 2989 |
"group_size": 64,
|
| 2990 |
+
"bits": 8
|
| 2991 |
},
|
| 2992 |
"language_model.model.layers.48.self_attn.k_proj": {
|
| 2993 |
"group_size": 64,
|
| 2994 |
+
"bits": 8
|
| 2995 |
},
|
| 2996 |
"language_model.model.layers.48.self_attn.v_proj": {
|
| 2997 |
"group_size": 64,
|
| 2998 |
+
"bits": 8
|
| 2999 |
},
|
| 3000 |
"language_model.model.layers.48.self_attn.o_proj": {
|
| 3001 |
"group_size": 64,
|
| 3002 |
+
"bits": 8
|
| 3003 |
},
|
| 3004 |
"language_model.model.layers.48.mlp.gate_proj": {
|
| 3005 |
"group_size": 64,
|
|
|
|
| 3015 |
},
|
| 3016 |
"language_model.model.layers.49.self_attn.q_proj": {
|
| 3017 |
"group_size": 64,
|
| 3018 |
+
"bits": 8
|
| 3019 |
},
|
| 3020 |
"language_model.model.layers.49.self_attn.k_proj": {
|
| 3021 |
"group_size": 64,
|
| 3022 |
+
"bits": 8
|
| 3023 |
},
|
| 3024 |
"language_model.model.layers.49.self_attn.v_proj": {
|
| 3025 |
"group_size": 64,
|
| 3026 |
+
"bits": 8
|
| 3027 |
},
|
| 3028 |
"language_model.model.layers.49.self_attn.o_proj": {
|
| 3029 |
"group_size": 64,
|
| 3030 |
+
"bits": 8
|
| 3031 |
},
|
| 3032 |
"language_model.model.layers.49.mlp.gate_proj": {
|
| 3033 |
"group_size": 64,
|
|
|
|
| 3043 |
},
|
| 3044 |
"language_model.model.layers.50.self_attn.q_proj": {
|
| 3045 |
"group_size": 64,
|
| 3046 |
+
"bits": 8
|
| 3047 |
},
|
| 3048 |
"language_model.model.layers.50.self_attn.k_proj": {
|
| 3049 |
"group_size": 64,
|
| 3050 |
+
"bits": 8
|
| 3051 |
},
|
| 3052 |
"language_model.model.layers.50.self_attn.v_proj": {
|
| 3053 |
"group_size": 64,
|
| 3054 |
+
"bits": 8
|
| 3055 |
},
|
| 3056 |
"language_model.model.layers.50.self_attn.o_proj": {
|
| 3057 |
"group_size": 64,
|
| 3058 |
+
"bits": 8
|
| 3059 |
},
|
| 3060 |
"language_model.model.layers.50.mlp.gate_proj": {
|
| 3061 |
"group_size": 64,
|
|
|
|
| 3071 |
},
|
| 3072 |
"language_model.model.layers.51.self_attn.q_proj": {
|
| 3073 |
"group_size": 64,
|
| 3074 |
+
"bits": 8
|
| 3075 |
},
|
| 3076 |
"language_model.model.layers.51.self_attn.k_proj": {
|
| 3077 |
"group_size": 64,
|
| 3078 |
+
"bits": 8
|
| 3079 |
},
|
| 3080 |
"language_model.model.layers.51.self_attn.v_proj": {
|
| 3081 |
"group_size": 64,
|
| 3082 |
+
"bits": 8
|
| 3083 |
},
|
| 3084 |
"language_model.model.layers.51.self_attn.o_proj": {
|
| 3085 |
"group_size": 64,
|
| 3086 |
+
"bits": 8
|
| 3087 |
},
|
| 3088 |
"language_model.model.layers.51.mlp.gate_proj": {
|
| 3089 |
"group_size": 64,
|
|
|
|
| 3099 |
},
|
| 3100 |
"language_model.model.layers.52.self_attn.q_proj": {
|
| 3101 |
"group_size": 64,
|
| 3102 |
+
"bits": 8
|
| 3103 |
},
|
| 3104 |
"language_model.model.layers.52.self_attn.k_proj": {
|
| 3105 |
"group_size": 64,
|
| 3106 |
+
"bits": 8
|
| 3107 |
},
|
| 3108 |
"language_model.model.layers.52.self_attn.v_proj": {
|
| 3109 |
"group_size": 64,
|
| 3110 |
+
"bits": 8
|
| 3111 |
},
|
| 3112 |
"language_model.model.layers.52.self_attn.o_proj": {
|
| 3113 |
"group_size": 64,
|
| 3114 |
+
"bits": 8
|
| 3115 |
},
|
| 3116 |
"language_model.model.layers.52.mlp.gate_proj": {
|
| 3117 |
"group_size": 64,
|
| 3118 |
+
"bits": 5
|
| 3119 |
},
|
| 3120 |
"language_model.model.layers.52.mlp.up_proj": {
|
| 3121 |
"group_size": 64,
|
| 3122 |
+
"bits": 5
|
| 3123 |
},
|
| 3124 |
"language_model.model.layers.52.mlp.down_proj": {
|
| 3125 |
"group_size": 64,
|
| 3126 |
+
"bits": 5
|
| 3127 |
},
|
| 3128 |
"language_model.model.layers.53.self_attn.q_proj": {
|
| 3129 |
"group_size": 64,
|
| 3130 |
+
"bits": 8
|
| 3131 |
},
|
| 3132 |
"language_model.model.layers.53.self_attn.k_proj": {
|
| 3133 |
"group_size": 64,
|
| 3134 |
+
"bits": 8
|
| 3135 |
},
|
| 3136 |
"language_model.model.layers.53.self_attn.o_proj": {
|
| 3137 |
"group_size": 64,
|
| 3138 |
+
"bits": 8
|
| 3139 |
},
|
| 3140 |
"language_model.model.layers.53.mlp.gate_proj": {
|
| 3141 |
"group_size": 64,
|
|
|
|
| 3151 |
},
|
| 3152 |
"language_model.model.layers.54.self_attn.q_proj": {
|
| 3153 |
"group_size": 64,
|
| 3154 |
+
"bits": 8
|
| 3155 |
},
|
| 3156 |
"language_model.model.layers.54.self_attn.k_proj": {
|
| 3157 |
"group_size": 64,
|
| 3158 |
+
"bits": 8
|
| 3159 |
},
|
| 3160 |
"language_model.model.layers.54.self_attn.v_proj": {
|
| 3161 |
"group_size": 64,
|
| 3162 |
+
"bits": 8
|
| 3163 |
},
|
| 3164 |
"language_model.model.layers.54.self_attn.o_proj": {
|
| 3165 |
"group_size": 64,
|
| 3166 |
+
"bits": 8
|
| 3167 |
},
|
| 3168 |
"language_model.model.layers.54.mlp.gate_proj": {
|
| 3169 |
"group_size": 64,
|
|
|
|
| 3179 |
},
|
| 3180 |
"language_model.model.layers.55.self_attn.q_proj": {
|
| 3181 |
"group_size": 64,
|
| 3182 |
+
"bits": 8
|
| 3183 |
},
|
| 3184 |
"language_model.model.layers.55.self_attn.k_proj": {
|
| 3185 |
"group_size": 64,
|
| 3186 |
+
"bits": 8
|
| 3187 |
},
|
| 3188 |
"language_model.model.layers.55.self_attn.v_proj": {
|
| 3189 |
"group_size": 64,
|
| 3190 |
+
"bits": 8
|
| 3191 |
},
|
| 3192 |
"language_model.model.layers.55.self_attn.o_proj": {
|
| 3193 |
"group_size": 64,
|
| 3194 |
+
"bits": 8
|
| 3195 |
},
|
| 3196 |
"language_model.model.layers.55.mlp.gate_proj": {
|
| 3197 |
"group_size": 64,
|
|
|
|
| 3207 |
},
|
| 3208 |
"language_model.model.layers.56.self_attn.q_proj": {
|
| 3209 |
"group_size": 64,
|
| 3210 |
+
"bits": 8
|
| 3211 |
},
|
| 3212 |
"language_model.model.layers.56.self_attn.k_proj": {
|
| 3213 |
"group_size": 64,
|
| 3214 |
+
"bits": 8
|
| 3215 |
},
|
| 3216 |
"language_model.model.layers.56.self_attn.v_proj": {
|
| 3217 |
"group_size": 64,
|
| 3218 |
+
"bits": 8
|
| 3219 |
},
|
| 3220 |
"language_model.model.layers.56.self_attn.o_proj": {
|
| 3221 |
"group_size": 64,
|
| 3222 |
+
"bits": 8
|
| 3223 |
},
|
| 3224 |
"language_model.model.layers.56.mlp.gate_proj": {
|
| 3225 |
"group_size": 64,
|
|
|
|
| 3235 |
},
|
| 3236 |
"language_model.model.layers.57.self_attn.q_proj": {
|
| 3237 |
"group_size": 64,
|
| 3238 |
+
"bits": 8
|
| 3239 |
},
|
| 3240 |
"language_model.model.layers.57.self_attn.k_proj": {
|
| 3241 |
"group_size": 64,
|
| 3242 |
+
"bits": 8
|
| 3243 |
},
|
| 3244 |
"language_model.model.layers.57.self_attn.v_proj": {
|
| 3245 |
"group_size": 64,
|
| 3246 |
+
"bits": 8
|
| 3247 |
},
|
| 3248 |
"language_model.model.layers.57.self_attn.o_proj": {
|
| 3249 |
"group_size": 64,
|
| 3250 |
+
"bits": 8
|
| 3251 |
},
|
| 3252 |
"language_model.model.layers.57.mlp.gate_proj": {
|
| 3253 |
"group_size": 64,
|
| 3254 |
+
"bits": 5
|
| 3255 |
},
|
| 3256 |
"language_model.model.layers.57.mlp.up_proj": {
|
| 3257 |
"group_size": 64,
|
| 3258 |
+
"bits": 5
|
| 3259 |
},
|
| 3260 |
"language_model.model.layers.57.mlp.down_proj": {
|
| 3261 |
"group_size": 64,
|
| 3262 |
+
"bits": 5
|
| 3263 |
},
|
| 3264 |
"language_model.model.layers.58.self_attn.q_proj": {
|
| 3265 |
"group_size": 64,
|
| 3266 |
+
"bits": 8
|
| 3267 |
},
|
| 3268 |
"language_model.model.layers.58.self_attn.k_proj": {
|
| 3269 |
"group_size": 64,
|
| 3270 |
+
"bits": 8
|
| 3271 |
},
|
| 3272 |
"language_model.model.layers.58.self_attn.v_proj": {
|
| 3273 |
"group_size": 64,
|
| 3274 |
+
"bits": 8
|
| 3275 |
},
|
| 3276 |
"language_model.model.layers.58.self_attn.o_proj": {
|
| 3277 |
"group_size": 64,
|
| 3278 |
+
"bits": 8
|
| 3279 |
},
|
| 3280 |
"language_model.model.layers.58.mlp.gate_proj": {
|
| 3281 |
"group_size": 64,
|
| 3282 |
+
"bits": 5
|
| 3283 |
},
|
| 3284 |
"language_model.model.layers.58.mlp.up_proj": {
|
| 3285 |
"group_size": 64,
|
| 3286 |
+
"bits": 5
|
| 3287 |
},
|
| 3288 |
"language_model.model.layers.58.mlp.down_proj": {
|
| 3289 |
"group_size": 64,
|
| 3290 |
+
"bits": 5
|
| 3291 |
},
|
| 3292 |
"language_model.model.layers.59.self_attn.q_proj": {
|
| 3293 |
"group_size": 64,
|
| 3294 |
+
"bits": 8
|
| 3295 |
},
|
| 3296 |
"language_model.model.layers.59.self_attn.k_proj": {
|
| 3297 |
"group_size": 64,
|
| 3298 |
+
"bits": 8
|
| 3299 |
},
|
| 3300 |
"language_model.model.layers.59.self_attn.o_proj": {
|
| 3301 |
"group_size": 64,
|
| 3302 |
+
"bits": 8
|
| 3303 |
},
|
| 3304 |
"language_model.model.layers.59.mlp.gate_proj": {
|
| 3305 |
"group_size": 64,
|
| 3306 |
+
"bits": 6
|
| 3307 |
},
|
| 3308 |
"language_model.model.layers.59.mlp.up_proj": {
|
| 3309 |
"group_size": 64,
|
| 3310 |
+
"bits": 6
|
| 3311 |
},
|
| 3312 |
"language_model.model.layers.59.mlp.down_proj": {
|
| 3313 |
"group_size": 64,
|
| 3314 |
+
"bits": 6
|
| 3315 |
}
|
| 3316 |
},
|
| 3317 |
"text_config": {
|