rubenz-org/llama_3_2_1b_instruct_MLC
060
1{2 "metadata": {3 "ParamSize": 163,4 "ParamBytes": 695242752.0,5 "BitsPerParam": 4.5006289099722416 },7 "records": [8 {9 "dataPath": "params_shard_0.bin",10 "format": "raw-shard",11 "nbytes": 131334144,12 "records": [13 {14 "name": "model.embed_tokens.q_weight",15 "shape": [16 128256,17 25618 ],19 "dtype": "uint32",20 "format": "f32-to-bf16",21 "nbytes": 131334144,22 "byteOffset": 023 }24 ],25 "md5sum": "5b16cbf639061d595947e48a6029f6d1"26 },27 {28 "dataPath": "params_shard_1.bin",29 "format": "raw-shard",30 "nbytes": 16777216,31 "records": [32 {33 "name": "model.layers.0.mlp.gate_up_proj.q_weight",34 "shape": [35 16384,36 25637 ],38 "dtype": "uint32",39 "format": "f32-to-bf16",40 "nbytes": 16777216,41 "byteOffset": 042 }43 ],44 "md5sum": "399f19dfe98352fae7f5dd65d2134f53"45 },46 {47 "dataPath": "params_shard_2.bin",48 "format": "raw-shard",49 "nbytes": 31498240,50 "records": [51 {52 "name": "model.embed_tokens.q_scale",53 "shape": [54 128256,55 6456 ],57 "dtype": "float16",58 "format": "f32-to-bf16",59 "nbytes": 16416768,60 "byteOffset": 061 },62 {63 "name": "model.layers.0.input_layernorm.weight",64 "shape": [65 204866 ],67 "dtype": "float16",68 "format": "f32-to-bf16",69 "nbytes": 4096,70 "byteOffset": 1641676871 },72 {73 "name": "model.layers.0.mlp.down_proj.q_weight",74 "shape": [75 2048,76 102477 ],78 "dtype": "uint32",79 "format": "f32-to-bf16",80 "nbytes": 8388608,81 "byteOffset": 1642086482 },83 {84 "name": "model.layers.0.mlp.down_proj.q_scale",85 "shape": [86 2048,87 25688 ],89 "dtype": "float16",90 "format": "f32-to-bf16",91 "nbytes": 1048576,92 "byteOffset": 2480947293 },94 {95 "name": "model.layers.0.mlp.gate_up_proj.q_scale",96 "shape": [97 16384,98 6499 ],100 "dtype": "float16",101 "format": "f32-to-bf16",102 "nbytes": 2097152,103 "byteOffset": 25858048104 },105 {106 "name": "model.layers.0.post_attention_layernorm.weight",107 "shape": [108 2048109 ],110 "dtype": "float16",111 "format": "f32-to-bf16",112 "nbytes": 4096,113 "byteOffset": 27955200114 },115 {116 "name": "model.layers.0.self_attn.qkv_proj.q_weight",117 "shape": [118 3072,119 256120 ],121 "dtype": "uint32",122 "format": "f32-to-bf16",123 "nbytes": 3145728,124 "byteOffset": 27959296125 },126 {127 "name": "model.layers.0.self_attn.qkv_proj.q_scale",128 "shape": [129 3072,130 64131 ],132 "dtype": "float16",133 "format": "f32-to-bf16",134 "nbytes": 393216,135 "byteOffset": 31105024136 }137 ],138 "md5sum": "1df499000a92eb97bb6bfad5c9befc0b"139 },140 {141 "dataPath": "params_shard_3.bin",142 "format": "raw-shard",143 "nbytes": 30679040,144 "records": [145 {146 "name": "model.layers.0.self_attn.o_proj.q_weight",147 "shape": [148 2048,149 256150 ],151 "dtype": "uint32",152 "format": "f32-to-bf16",153 "nbytes": 2097152,154 "byteOffset": 0155 },156 {157 "name": "model.layers.0.self_attn.o_proj.q_scale",158 "shape": [159 2048,160 64161 ],162 "dtype": "float16",163 "format": "f32-to-bf16",164 "nbytes": 262144,165 "byteOffset": 2097152166 },167 {168 "name": "model.layers.1.input_layernorm.weight",169 "shape": [170 2048171 ],172 "dtype": "float16",173 "format": "f32-to-bf16",174 "nbytes": 4096,175 "byteOffset": 2359296176 },177 {178 "name": "model.layers.1.mlp.down_proj.q_weight",179 "shape": [180 2048,181 1024182 ],183 "dtype": "uint32",184 "format": "f32-to-bf16",185 "nbytes": 8388608,186 "byteOffset": 2363392187 },188 {189 "name": "model.layers.1.mlp.down_proj.q_scale",190 "shape": [191 2048,192 256193 ],194 "dtype": "float16",195 "format": "f32-to-bf16",196 "nbytes": 1048576,197 "byteOffset": 10752000198 },199 {200 "name": "model.layers.1.mlp.gate_up_proj.q_weight",201 "shape": [202 16384,203 256204 ],205 "dtype": "uint32",206 "format": "f32-to-bf16",207 "nbytes": 16777216,208 "byteOffset": 11800576209 },210 {211 "name": "model.layers.1.mlp.gate_up_proj.q_scale",212 "shape": [213 16384,214 64215 ],216 "dtype": "float16",217 "format": "f32-to-bf16",218 "nbytes": 2097152,219 "byteOffset": 28577792220 },221 {222 "name": "model.layers.1.post_attention_layernorm.weight",223 "shape": [224 2048225 ],226 "dtype": "float16",227 "format": "f32-to-bf16",228 "nbytes": 4096,229 "byteOffset": 30674944230 }231 ],232 "md5sum": "f92c70684df5f3b7c9377340b00388da"233 },234 {235 "dataPath": "params_shard_4.bin",236 "format": "raw-shard",237 "nbytes": 32116736,238 "records": [239 {240 "name": "model.layers.1.self_attn.qkv_proj.q_weight",241 "shape": [242 3072,243 256244 ],245 "dtype": "uint32",246 "format": "f32-to-bf16",247 "nbytes": 3145728,248 "byteOffset": 0249 },250 {251 "name": "model.layers.1.self_attn.qkv_proj.q_scale",252 "shape": [253 3072,254 64255 ],256 "dtype": "float16",257 "format": "f32-to-bf16",258 "nbytes": 393216,259 "byteOffset": 3145728260 },261 {262 "name": "model.layers.1.self_attn.o_proj.q_weight",263 "shape": [264 2048,265 256266 ],267 "dtype": "uint32",268 "format": "f32-to-bf16",269 "nbytes": 2097152,270 "byteOffset": 3538944271 },272 {273 "name": "model.layers.1.self_attn.o_proj.q_scale",274 "shape": [275 2048,276 64277 ],278 "dtype": "float16",279 "format": "f32-to-bf16",280 "nbytes": 262144,281 "byteOffset": 5636096282 },283 {284 "name": "model.layers.10.input_layernorm.weight",285 "shape": [286 2048287 ],288 "dtype": "float16",289 "format": "f32-to-bf16",290 "nbytes": 4096,291 "byteOffset": 5898240292 },293 {294 "name": "model.layers.10.mlp.down_proj.q_weight",295 "shape": [296 2048,297 1024298 ],299 "dtype": "uint32",300 "format": "f32-to-bf16",301 "nbytes": 8388608,302 "byteOffset": 5902336303 },304 {305 "name": "model.layers.10.mlp.down_proj.q_scale",306 "shape": [307 2048,308 256309 ],310 "dtype": "float16",311 "format": "f32-to-bf16",312 "nbytes": 1048576,313 "byteOffset": 14290944314 },315 {316 "name": "model.layers.10.mlp.gate_up_proj.q_weight",317 "shape": [318 16384,319 256320 ],321 "dtype": "uint32",322 "format": "f32-to-bf16",323 "nbytes": 16777216,324 "byteOffset": 15339520325 }326 ],327 "md5sum": "e563916b198cd302f4d06b2869e7575a"328 },329 {330 "dataPath": "params_shard_5.bin",331 "format": "raw-shard",332 "nbytes": 16777216,333 "records": [334 {335 "name": "model.layers.11.mlp.gate_up_proj.q_weight",336 "shape": [337 16384,338 256339 ],340 "dtype": "uint32",341 "format": "f32-to-bf16",342 "nbytes": 16777216,343 "byteOffset": 0344 }345 ],346 "md5sum": "cd24e35c5002b3f4844ade215d98aa3c"347 },348 {349 "dataPath": "params_shard_6.bin",350 "format": "raw-shard",351 "nbytes": 25444352,352 "records": [353 {354 "name": "model.layers.10.mlp.gate_up_proj.q_scale",355 "shape": [356 16384,357 64358 ],359 "dtype": "float16",360 "format": "f32-to-bf16",361 "nbytes": 2097152,362 "byteOffset": 0363 },364 {365 "name": "model.layers.10.post_attention_layernorm.weight",366 "shape": [367 2048368 ],369 "dtype": "float16",370 "format": "f32-to-bf16",371 "nbytes": 4096,372 "byteOffset": 2097152373 },374 {375 "name": "model.layers.10.self_attn.qkv_proj.q_weight",376 "shape": [377 3072,378 256379 ],380 "dtype": "uint32",381 "format": "f32-to-bf16",382 "nbytes": 3145728,383 "byteOffset": 2101248384 },385 {386 "name": "model.layers.10.self_attn.qkv_proj.q_scale",387 "shape": [388 3072,389 64390 ],391 "dtype": "float16",392 "format": "f32-to-bf16",393 "nbytes": 393216,394 "byteOffset": 5246976395 },396 {397 "name": "model.layers.10.self_attn.o_proj.q_weight",398 "shape": [399 2048,400 256401 ],402 "dtype": "uint32",403 "format": "f32-to-bf16",404 "nbytes": 2097152,405 "byteOffset": 5640192406 },407 {408 "name": "model.layers.10.self_attn.o_proj.q_scale",409 "shape": [410 2048,411 64412 ],413 "dtype": "float16",414 "format": "f32-to-bf16",415 "nbytes": 262144,416 "byteOffset": 7737344417 },418 {419 "name": "model.layers.11.input_layernorm.weight",420 "shape": [421 2048422 ],423 "dtype": "float16",424 "format": "f32-to-bf16",425 "nbytes": 4096,426 "byteOffset": 7999488427 },428 {429 "name": "model.layers.11.mlp.down_proj.q_weight",430 "shape": [431 2048,432 1024433 ],434 "dtype": "uint32",435 "format": "f32-to-bf16",436 "nbytes": 8388608,437 "byteOffset": 8003584438 },439 {440 "name": "model.layers.11.mlp.down_proj.q_scale",441 "shape": [442 2048,443 256444 ],445 "dtype": "float16",446 "format": "f32-to-bf16",447 "nbytes": 1048576,448 "byteOffset": 16392192449 },450 {451 "name": "model.layers.11.mlp.gate_up_proj.q_scale",452 "shape": [453 16384,454 64455 ],456 "dtype": "float16",457 "format": "f32-to-bf16",458 "nbytes": 2097152,459 "byteOffset": 17440768460 },461 {462 "name": "model.layers.11.post_attention_layernorm.weight",463 "shape": [464 2048465 ],466 "dtype": "float16",467 "format": "f32-to-bf16",468 "nbytes": 4096,469 "byteOffset": 19537920470 },471 {472 "name": "model.layers.11.self_attn.qkv_proj.q_weight",473 "shape": [474 3072,475 256476 ],477 "dtype": "uint32",478 "format": "f32-to-bf16",479 "nbytes": 3145728,480 "byteOffset": 19542016481 },482 {483 "name": "model.layers.11.self_attn.qkv_proj.q_scale",484 "shape": [485 3072,486 64487 ],488 "dtype": "float16",489 "format": "f32-to-bf16",490 "nbytes": 393216,491 "byteOffset": 22687744492 },493 {494 "name": "model.layers.11.self_attn.o_proj.q_weight",495 "shape": [496 2048,497 256498 ],499 "dtype": "uint32",500 "format": "f32-to-bf16",501 "nbytes": 2097152,502 "byteOffset": 23080960503 },504 {505 "name": "model.layers.11.self_attn.o_proj.q_scale",506 "shape": [507 2048,508 64509 ],510 "dtype": "float16",511 "format": "f32-to-bf16",512 "nbytes": 262144,513 "byteOffset": 25178112514 },515 {516 "name": "model.layers.12.input_layernorm.weight",517 "shape": [518 2048519 ],520 "dtype": "float16",521 "format": "f32-to-bf16",522 "nbytes": 4096,523 "byteOffset": 25440256524 }525 ],526 "md5sum": "790e5cd9179f76c89f50afb493e71005"527 },528 {529 "dataPath": "params_shard_7.bin",530 "format": "raw-shard",531 "nbytes": 31854592,532 "records": [533 {534 "name": "model.layers.12.mlp.down_proj.q_weight",535 "shape": [536 2048,537 1024538 ],539 "dtype": "uint32",540 "format": "f32-to-bf16",541 "nbytes": 8388608,542 "byteOffset": 0543 },544 {545 "name": "model.layers.12.mlp.down_proj.q_scale",546 "shape": [547 2048,548 256549 ],550 "dtype": "float16",551 "format": "f32-to-bf16",552 "nbytes": 1048576,553 "byteOffset": 8388608554 },555 {556 "name": "model.layers.12.mlp.gate_up_proj.q_weight",557 "shape": [558 16384,559 256560 ],561 "dtype": "uint32",562 "format": "f32-to-bf16",563 "nbytes": 16777216,564 "byteOffset": 9437184565 },566 {567 "name": "model.layers.12.mlp.gate_up_proj.q_scale",568 "shape": [569 16384,570 64571 ],572 "dtype": "float16",573 "format": "f32-to-bf16",574 "nbytes": 2097152,575 "byteOffset": 26214400576 },577 {578 "name": "model.layers.12.post_attention_layernorm.weight",579 "shape": [580 2048581 ],582 "dtype": "float16",583 "format": "f32-to-bf16",584 "nbytes": 4096,585 "byteOffset": 28311552586 },587 {588 "name": "model.layers.12.self_attn.qkv_proj.q_weight",589 "shape": [590 3072,591 256592 ],593 "dtype": "uint32",594 "format": "f32-to-bf16",595 "nbytes": 3145728,596 "byteOffset": 28315648597 },598 {599 "name": "model.layers.12.self_attn.qkv_proj.q_scale",600 "shape": [601 3072,602 64603 ],604 "dtype": "float16",605 "format": "f32-to-bf16",606 "nbytes": 393216,607 "byteOffset": 31461376608 }609 ],610 "md5sum": "10bc3282c23bc5b4c1cca579ea759645"611 },612 {613 "dataPath": "params_shard_8.bin",614 "format": "raw-shard",615 "nbytes": 30679040,616 "records": [617 {618 "name": "model.layers.12.self_attn.o_proj.q_weight",619 "shape": [620 2048,621 256622 ],623 "dtype": "uint32",624 "format": "f32-to-bf16",625 "nbytes": 2097152,626 "byteOffset": 0627 },628 {629 "name": "model.layers.12.self_attn.o_proj.q_scale",630 "shape": [631 2048,632 64633 ],634 "dtype": "float16",635 "format": "f32-to-bf16",636 "nbytes": 262144,637 "byteOffset": 2097152638 },639 {640 "name": "model.layers.13.input_layernorm.weight",641 "shape": [642 2048643 ],644 "dtype": "float16",645 "format": "f32-to-bf16",646 "nbytes": 4096,647 "byteOffset": 2359296648 },649 {650 "name": "model.layers.13.mlp.down_proj.q_weight",651 "shape": [652 2048,653 1024654 ],655 "dtype": "uint32",656 "format": "f32-to-bf16",657 "nbytes": 8388608,658 "byteOffset": 2363392659 },660 {661 "name": "model.layers.13.mlp.down_proj.q_scale",662 "shape": [663 2048,664 256665 ],666 "dtype": "float16",667 "format": "f32-to-bf16",668 "nbytes": 1048576,669 "byteOffset": 10752000670 },671 {672 "name": "model.layers.13.mlp.gate_up_proj.q_weight",673 "shape": [674 16384,675 256676 ],677 "dtype": "uint32",678 "format": "f32-to-bf16",679 "nbytes": 16777216,680 "byteOffset": 11800576681 },682 {683 "name": "model.layers.13.mlp.gate_up_proj.q_scale",684 "shape": [685 16384,686 64687 ],688 "dtype": "float16",689 "format": "f32-to-bf16",690 "nbytes": 2097152,691 "byteOffset": 28577792692 },693 {694 "name": "model.layers.13.post_attention_layernorm.weight",695 "shape": [696 2048697 ],698 "dtype": "float16",699 "format": "f32-to-bf16",700 "nbytes": 4096,701 "byteOffset": 30674944702 }703 ],704 "md5sum": "454f0626712f59c7a3df706f17e39932"705 },706 {707 "dataPath": "params_shard_9.bin",708 "format": "raw-shard",709 "nbytes": 32116736,710 "records": [711 {712 "name": "model.layers.13.self_attn.qkv_proj.q_weight",713 "shape": [714 3072,715 256716 ],717 "dtype": "uint32",718 "format": "f32-to-bf16",719 "nbytes": 3145728,720 "byteOffset": 0721 },722 {723 "name": "model.layers.13.self_attn.qkv_proj.q_scale",724 "shape": [725 3072,726 64727 ],728 "dtype": "float16",729 "format": "f32-to-bf16",730 "nbytes": 393216,731 "byteOffset": 3145728732 },733 {734 "name": "model.layers.13.self_attn.o_proj.q_weight",735 "shape": [736 2048,737 256738 ],739 "dtype": "uint32",740 "format": "f32-to-bf16",741 "nbytes": 2097152,742 "byteOffset": 3538944743 },744 {745 "name": "model.layers.13.self_attn.o_proj.q_scale",746 "shape": [747 2048,748 64749 ],750 "dtype": "float16",751 "format": "f32-to-bf16",752 "nbytes": 262144,753 "byteOffset": 5636096754 },755 {756 "name": "model.layers.14.input_layernorm.weight",757 "shape": [758 2048759 ],760 "dtype": "float16",761 "format": "f32-to-bf16",762 "nbytes": 4096,763 "byteOffset": 5898240764 },765 {766 "name": "model.layers.14.mlp.down_proj.q_weight",767 "shape": [768 2048,769 1024770 ],771 "dtype": "uint32",772 "format": "f32-to-bf16",773 "nbytes": 8388608,774 "byteOffset": 5902336775 },776 {777 "name": "model.layers.14.mlp.down_proj.q_scale",778 "shape": [779 2048,780 256781 ],782 "dtype": "float16",783 "format": "f32-to-bf16",784 "nbytes": 1048576,785 "byteOffset": 14290944786 },787 {788 "name": "model.layers.14.mlp.gate_up_proj.q_weight",789 "shape": [790 16384,791 256792 ],793 "dtype": "uint32",794 "format": "f32-to-bf16",795 "nbytes": 16777216,796 "byteOffset": 15339520797 }798 ],799 "md5sum": "44d2358aba02dc567f54998e6a75fb78"800 },801 {802 "dataPath": "params_shard_10.bin",803 "format": "raw-shard",804 "nbytes": 16777216,805 "records": [806 {807 "name": "model.layers.15.mlp.gate_up_proj.q_weight",808 "shape": [809 16384,810 256811 ],812 "dtype": "uint32",813 "format": "f32-to-bf16",814 "nbytes": 16777216,815 "byteOffset": 0816 }817 ],818 "md5sum": "d3fd5cd5438c74f5743d05c5324f3fd3"819 },820 {821 "dataPath": "params_shard_11.bin",822 "format": "raw-shard",823 "nbytes": 25444352,824 "records": [825 {826 "name": "model.layers.14.mlp.gate_up_proj.q_scale",827 "shape": [828 16384,829 64830 ],831 "dtype": "float16",832 "format": "f32-to-bf16",833 "nbytes": 2097152,834 "byteOffset": 0835 },836 {837 "name": "model.layers.14.post_attention_layernorm.weight",838 "shape": [839 2048840 ],841 "dtype": "float16",842 "format": "f32-to-bf16",843 "nbytes": 4096,844 "byteOffset": 2097152845 },846 {847 "name": "model.layers.14.self_attn.qkv_proj.q_weight",848 "shape": [849 3072,850 256851 ],852 "dtype": "uint32",853 "format": "f32-to-bf16",854 "nbytes": 3145728,855 "byteOffset": 2101248856 },857 {858 "name": "model.layers.14.self_attn.qkv_proj.q_scale",859 "shape": [860 3072,861 64862 ],863 "dtype": "float16",864 "format": "f32-to-bf16",865 "nbytes": 393216,866 "byteOffset": 5246976867 },868 {869 "name": "model.layers.14.self_attn.o_proj.q_weight",870 "shape": [871 2048,872 256873 ],874 "dtype": "uint32",875 "format": "f32-to-bf16",876 "nbytes": 2097152,877 "byteOffset": 5640192878 },879 {880 "name": "model.layers.14.self_attn.o_proj.q_scale",881 "shape": [882 2048,883 64884 ],885 "dtype": "float16",886 "format": "f32-to-bf16",887 "nbytes": 262144,888 "byteOffset": 7737344889 },890 {891 "name": "model.layers.15.input_layernorm.weight",892 "shape": [893 2048894 ],895 "dtype": "float16",896 "format": "f32-to-bf16",897 "nbytes": 4096,898 "byteOffset": 7999488899 },900 {901 "name": "model.layers.15.mlp.down_proj.q_weight",902 "shape": [903 2048,904 1024905 ],906 "dtype": "uint32",907 "format": "f32-to-bf16",908 "nbytes": 8388608,909 "byteOffset": 8003584910 },911 {912 "name": "model.layers.15.mlp.down_proj.q_scale",913 "shape": [914 2048,915 256916 ],917 "dtype": "float16",918 "format": "f32-to-bf16",919 "nbytes": 1048576,920 "byteOffset": 16392192921 },922 {923 "name": "model.layers.15.mlp.gate_up_proj.q_scale",924 "shape": [925 16384,926 64927 ],928 "dtype": "float16",929 "format": "f32-to-bf16",930 "nbytes": 2097152,931 "byteOffset": 17440768932 },933 {934 "name": "model.layers.15.post_attention_layernorm.weight",935 "shape": [936 2048937 ],938 "dtype": "float16",939 "format": "f32-to-bf16",940 "nbytes": 4096,941 "byteOffset": 19537920942 },943 {944 "name": "model.layers.15.self_attn.qkv_proj.q_weight",945 "shape": [946 3072,947 256948 ],949 "dtype": "uint32",950 "format": "f32-to-bf16",951 "nbytes": 3145728,952 "byteOffset": 19542016953 },954 {955 "name": "model.layers.15.self_attn.qkv_proj.q_scale",956 "shape": [957 3072,958 64959 ],960 "dtype": "float16",961 "format": "f32-to-bf16",962 "nbytes": 393216,963 "byteOffset": 22687744964 },965 {966 "name": "model.layers.15.self_attn.o_proj.q_weight",967 "shape": [968 2048,969 256970 ],971 "dtype": "uint32",972 "format": "f32-to-bf16",973 "nbytes": 2097152,974 "byteOffset": 23080960975 },976 {977 "name": "model.layers.15.self_attn.o_proj.q_scale",978 "shape": [979 2048,980 64981 ],982 "dtype": "float16",983 "format": "f32-to-bf16",984 "nbytes": 262144,985 "byteOffset": 25178112986 },987 {988 "name": "model.layers.2.input_layernorm.weight",989 "shape": [990 2048991 ],992 "dtype": "float16",993 "format": "f32-to-bf16",994 "nbytes": 4096,995 "byteOffset": 25440256996 }997 ],998 "md5sum": "a4ec52fb4769bc90c99a46f9e57345ef"999 },1000 {1001 "dataPath": "params_shard_12.bin",1002 "format": "raw-shard",1003 "nbytes": 31854592,1004 "records": [1005 {1006 "name": "model.layers.2.mlp.down_proj.q_weight",1007 "shape": [1008 2048,1009 10241010 ],1011 "dtype": "uint32",1012 "format": "f32-to-bf16",1013 "nbytes": 8388608,1014 "byteOffset": 01015 },1016 {1017 "name": "model.layers.2.mlp.down_proj.q_scale",1018 "shape": [1019 2048,1020 2561021 ],1022 "dtype": "float16",1023 "format": "f32-to-bf16",1024 "nbytes": 1048576,1025 "byteOffset": 83886081026 },1027 {1028 "name": "model.layers.2.mlp.gate_up_proj.q_weight",1029 "shape": [1030 16384,1031 2561032 ],1033 "dtype": "uint32",1034 "format": "f32-to-bf16",1035 "nbytes": 16777216,1036 "byteOffset": 94371841037 },1038 {1039 "name": "model.layers.2.mlp.gate_up_proj.q_scale",1040 "shape": [1041 16384,1042 641043 ],1044 "dtype": "float16",1045 "format": "f32-to-bf16",1046 "nbytes": 2097152,1047 "byteOffset": 262144001048 },1049 {1050 "name": "model.layers.2.post_attention_layernorm.weight",1051 "shape": [1052 20481053 ],1054 "dtype": "float16",1055 "format": "f32-to-bf16",1056 "nbytes": 4096,1057 "byteOffset": 283115521058 },1059 {1060 "name": "model.layers.2.self_attn.qkv_proj.q_weight",1061 "shape": [1062 3072,1063 2561064 ],1065 "dtype": "uint32",1066 "format": "f32-to-bf16",1067 "nbytes": 3145728,1068 "byteOffset": 283156481069 },1070 {1071 "name": "model.layers.2.self_attn.qkv_proj.q_scale",1072 "shape": [1073 3072,1074 641075 ],1076 "dtype": "float16",1077 "format": "f32-to-bf16",1078 "nbytes": 393216,1079 "byteOffset": 314613761080 }1081 ],1082 "md5sum": "8405b7ebfd1e3cf7aafbb65bf48246c9"1083 },1084 {1085 "dataPath": "params_shard_13.bin",1086 "format": "raw-shard",1087 "nbytes": 30679040,1088 "records": [1089 {1090 "name": "model.layers.2.self_attn.o_proj.q_weight",1091 "shape": [1092 2048,1093 2561094 ],1095 "dtype": "uint32",1096 "format": "f32-to-bf16",1097 "nbytes": 2097152,1098 "byteOffset": 01099 },1100 {1101 "name": "model.layers.2.self_attn.o_proj.q_scale",1102 "shape": [1103 2048,1104 641105 ],1106 "dtype": "float16",1107 "format": "f32-to-bf16",1108 "nbytes": 262144,1109 "byteOffset": 20971521110 },1111 {1112 "name": "model.layers.3.input_layernorm.weight",1113 "shape": [1114 20481115 ],1116 "dtype": "float16",1117 "format": "f32-to-bf16",1118 "nbytes": 4096,1119 "byteOffset": 23592961120 },1121 {1122 "name": "model.layers.3.mlp.down_proj.q_weight",1123 "shape": [1124 2048,1125 10241126 ],1127 "dtype": "uint32",1128 "format": "f32-to-bf16",1129 "nbytes": 8388608,1130 "byteOffset": 23633921131 },1132 {1133 "name": "model.layers.3.mlp.down_proj.q_scale",1134 "shape": [1135 2048,1136 2561137 ],1138 "dtype": "float16",1139 "format": "f32-to-bf16",1140 "nbytes": 1048576,1141 "byteOffset": 107520001142 },1143 {1144 "name": "model.layers.3.mlp.gate_up_proj.q_weight",1145 "shape": [1146 16384,1147 2561148 ],1149 "dtype": "uint32",1150 "format": "f32-to-bf16",1151 "nbytes": 16777216,1152 "byteOffset": 118005761153 },1154 {1155 "name": "model.layers.3.mlp.gate_up_proj.q_scale",1156 "shape": [1157 16384,1158 641159 ],1160 "dtype": "float16",1161 "format": "f32-to-bf16",1162 "nbytes": 2097152,1163 "byteOffset": 285777921164 },1165 {1166 "name": "model.layers.3.post_attention_layernorm.weight",1167 "shape": [1168 20481169 ],1170 "dtype": "float16",1171 "format": "f32-to-bf16",1172 "nbytes": 4096,1173 "byteOffset": 306749441174 }1175 ],1176 "md5sum": "66223f02c1a4358cf574a8cc34322376"1177 },1178 {1179 "dataPath": "params_shard_14.bin",1180 "format": "raw-shard",1181 "nbytes": 32116736,1182 "records": [1183 {1184 "name": "model.layers.3.self_attn.qkv_proj.q_weight",1185 "shape": [1186 3072,1187 2561188 ],1189 "dtype": "uint32",1190 "format": "f32-to-bf16",1191 "nbytes": 3145728,1192 "byteOffset": 01193 },1194 {1195 "name": "model.layers.3.self_attn.qkv_proj.q_scale",1196 "shape": [1197 3072,1198 641199 ],1200 "dtype": "float16",