Jeethu/Meta-Llama-3.1-8B-Instruct-PLLM
06
1{2 "metadata": {3 "ParamSize": 325,4 "ParamBytes": 4140998656.0,5 "BitsPerParam": 4.0701209831028266 },7 "records": [8 {9 "dataPath": "params_shard_0.bin",10 "format": "raw-shard",11 "nbytes": 262668288,12 "records": [13 {14 "name": "lm_head.q_weight",15 "shape": [16 512,17 12825618 ],19 "dtype": "uint32",20 "format": "f32-to-bf16",21 "nbytes": 262668288,22 "byteOffset": 023 }24 ],25 "md5sum": "58bb3a884ba5f20edb72182486a45c0c"26 },27 {28 "dataPath": "params_shard_1.bin",29 "format": "raw-shard",30 "nbytes": 29360128,31 "records": [32 {33 "name": "model.layers.31.mlp.down_proj.q_weight",34 "shape": [35 1792,36 409637 ],38 "dtype": "uint32",39 "format": "f32-to-bf16",40 "nbytes": 29360128,41 "byteOffset": 042 }43 ],44 "md5sum": "90243819b8fe4c6fef274410da882b85"45 },46 {47 "dataPath": "params_shard_2.bin",48 "format": "raw-shard",49 "nbytes": 58720256,50 "records": [51 {52 "name": "model.layers.31.mlp.gate_up_proj.q_weight",53 "shape": [54 512,55 2867256 ],57 "dtype": "uint32",58 "format": "f32-to-bf16",59 "nbytes": 58720256,60 "byteOffset": 061 }62 ],63 "md5sum": "0e20f69c3aa3a38db5d646ffbb7daca4"64 },65 {66 "dataPath": "params_shard_3.bin",67 "format": "raw-shard",68 "nbytes": 262668288,69 "records": [70 {71 "name": "model.embed_tokens.q_weight",72 "shape": [73 128256,74 51275 ],76 "dtype": "uint32",77 "format": "f32-to-bf16",78 "nbytes": 262668288,79 "byteOffset": 080 }81 ],82 "md5sum": "607377f17782be27fff9c36ecab568a7"83 },84 {85 "dataPath": "params_shard_4.bin",86 "format": "raw-shard",87 "nbytes": 29360128,88 "records": [89 {90 "name": "model.layers.0.mlp.down_proj.q_weight",91 "shape": [92 1792,93 409694 ],95 "dtype": "uint32",96 "format": "f32-to-bf16",97 "nbytes": 29360128,98 "byteOffset": 099 }100 ],101 "md5sum": "7243bbb4827433d4e6106d8e27986e19"102 },103 {104 "dataPath": "params_shard_5.bin",105 "format": "raw-shard",106 "nbytes": 58720256,107 "records": [108 {109 "name": "model.layers.0.mlp.gate_up_proj.q_weight",110 "shape": [111 512,112 28672113 ],114 "dtype": "uint32",115 "format": "f32-to-bf16",116 "nbytes": 58720256,117 "byteOffset": 0118 }119 ],120 "md5sum": "7318ca39cddcd98456b832c18512a7c0"121 },122 {123 "dataPath": "params_shard_6.bin",124 "format": "raw-shard",125 "nbytes": 21962752,126 "records": [127 {128 "name": "lm_head.q_scale",129 "shape": [130 32,131 128256132 ],133 "dtype": "float16",134 "format": "f32-to-bf16",135 "nbytes": 8208384,136 "byteOffset": 0137 },138 {139 "name": "model.layers.31.input_layernorm.weight",140 "shape": [141 4096142 ],143 "dtype": "float16",144 "format": "f32-to-bf16",145 "nbytes": 8192,146 "byteOffset": 8208384147 },148 {149 "name": "model.layers.31.mlp.down_proj.q_scale",150 "shape": [151 112,152 4096153 ],154 "dtype": "float16",155 "format": "f32-to-bf16",156 "nbytes": 917504,157 "byteOffset": 8216576158 },159 {160 "name": "model.layers.31.mlp.gate_up_proj.q_scale",161 "shape": [162 32,163 28672164 ],165 "dtype": "float16",166 "format": "f32-to-bf16",167 "nbytes": 1835008,168 "byteOffset": 9134080169 },170 {171 "name": "model.layers.31.post_attention_layernorm.weight",172 "shape": [173 4096174 ],175 "dtype": "float16",176 "format": "f32-to-bf16",177 "nbytes": 8192,178 "byteOffset": 10969088179 },180 {181 "name": "model.norm.weight",182 "shape": [183 4096184 ],185 "dtype": "float16",186 "format": "f32-to-bf16",187 "nbytes": 8192,188 "byteOffset": 10977280189 },190 {191 "name": "model.embed_tokens.q_scale",192 "shape": [193 128256,194 32195 ],196 "dtype": "float16",197 "format": "f32-to-bf16",198 "nbytes": 8208384,199 "byteOffset": 10985472200 },201 {202 "name": "model.layers.0.input_layernorm.weight",203 "shape": [204 4096205 ],206 "dtype": "float16",207 "format": "f32-to-bf16",208 "nbytes": 8192,209 "byteOffset": 19193856210 },211 {212 "name": "model.layers.0.mlp.down_proj.q_scale",213 "shape": [214 112,215 4096216 ],217 "dtype": "float16",218 "format": "f32-to-bf16",219 "nbytes": 917504,220 "byteOffset": 19202048221 },222 {223 "name": "model.layers.0.mlp.gate_up_proj.q_scale",224 "shape": [225 32,226 28672227 ],228 "dtype": "float16",229 "format": "f32-to-bf16",230 "nbytes": 1835008,231 "byteOffset": 20119552232 },233 {234 "name": "model.layers.0.post_attention_layernorm.weight",235 "shape": [236 4096237 ],238 "dtype": "float16",239 "format": "f32-to-bf16",240 "nbytes": 8192,241 "byteOffset": 21954560242 }243 ],244 "md5sum": "8d84b4bf8292e0b48020ae75d4627622"245 },246 {247 "dataPath": "params_shard_7.bin",248 "format": "raw-shard",249 "nbytes": 29360128,250 "records": [251 {252 "name": "model.layers.1.mlp.down_proj.q_weight",253 "shape": [254 1792,255 4096256 ],257 "dtype": "uint32",258 "format": "f32-to-bf16",259 "nbytes": 29360128,260 "byteOffset": 0261 }262 ],263 "md5sum": "c0cc79e61de81c6e49956e1854003797"264 },265 {266 "dataPath": "params_shard_8.bin",267 "format": "raw-shard",268 "nbytes": 58720256,269 "records": [270 {271 "name": "model.layers.1.mlp.gate_up_proj.q_weight",272 "shape": [273 512,274 28672275 ],276 "dtype": "uint32",277 "format": "f32-to-bf16",278 "nbytes": 58720256,279 "byteOffset": 0280 }281 ],282 "md5sum": "26ac4c4096da6980c9f61e86a1d3a009"283 },284 {285 "dataPath": "params_shard_9.bin",286 "format": "raw-shard",287 "nbytes": 24395776,288 "records": [289 {290 "name": "model.layers.0.self_attn.qkv_proj.q_weight",291 "shape": [292 512,293 6144294 ],295 "dtype": "uint32",296 "format": "f32-to-bf16",297 "nbytes": 12582912,298 "byteOffset": 0299 },300 {301 "name": "model.layers.0.self_attn.qkv_proj.q_scale",302 "shape": [303 32,304 6144305 ],306 "dtype": "float16",307 "format": "f32-to-bf16",308 "nbytes": 393216,309 "byteOffset": 12582912310 },311 {312 "name": "model.layers.0.self_attn.o_proj.q_weight",313 "shape": [314 512,315 4096316 ],317 "dtype": "uint32",318 "format": "f32-to-bf16",319 "nbytes": 8388608,320 "byteOffset": 12976128321 },322 {323 "name": "model.layers.0.self_attn.o_proj.q_scale",324 "shape": [325 32,326 4096327 ],328 "dtype": "float16",329 "format": "f32-to-bf16",330 "nbytes": 262144,331 "byteOffset": 21364736332 },333 {334 "name": "model.layers.1.input_layernorm.weight",335 "shape": [336 4096337 ],338 "dtype": "float16",339 "format": "f32-to-bf16",340 "nbytes": 8192,341 "byteOffset": 21626880342 },343 {344 "name": "model.layers.1.mlp.down_proj.q_scale",345 "shape": [346 112,347 4096348 ],349 "dtype": "float16",350 "format": "f32-to-bf16",351 "nbytes": 917504,352 "byteOffset": 21635072353 },354 {355 "name": "model.layers.1.mlp.gate_up_proj.q_scale",356 "shape": [357 32,358 28672359 ],360 "dtype": "float16",361 "format": "f32-to-bf16",362 "nbytes": 1835008,363 "byteOffset": 22552576364 },365 {366 "name": "model.layers.1.post_attention_layernorm.weight",367 "shape": [368 4096369 ],370 "dtype": "float16",371 "format": "f32-to-bf16",372 "nbytes": 8192,373 "byteOffset": 24387584374 }375 ],376 "md5sum": "9e90e43675446c51f2c58a9186ef7b04"377 },378 {379 "dataPath": "params_shard_10.bin",380 "format": "raw-shard",381 "nbytes": 29360128,382 "records": [383 {384 "name": "model.layers.2.mlp.down_proj.q_weight",385 "shape": [386 1792,387 4096388 ],389 "dtype": "uint32",390 "format": "f32-to-bf16",391 "nbytes": 29360128,392 "byteOffset": 0393 }394 ],395 "md5sum": "1d070fa7a5804938f2a8f8cbc474f672"396 },397 {398 "dataPath": "params_shard_11.bin",399 "format": "raw-shard",400 "nbytes": 58720256,401 "records": [402 {403 "name": "model.layers.2.mlp.gate_up_proj.q_weight",404 "shape": [405 512,406 28672407 ],408 "dtype": "uint32",409 "format": "f32-to-bf16",410 "nbytes": 58720256,411 "byteOffset": 0412 }413 ],414 "md5sum": "5f7adb2cbce4fb050b421b6b8ca23767"415 },416 {417 "dataPath": "params_shard_12.bin",418 "format": "raw-shard",419 "nbytes": 24395776,420 "records": [421 {422 "name": "model.layers.1.self_attn.qkv_proj.q_weight",423 "shape": [424 512,425 6144426 ],427 "dtype": "uint32",428 "format": "f32-to-bf16",429 "nbytes": 12582912,430 "byteOffset": 0431 },432 {433 "name": "model.layers.1.self_attn.qkv_proj.q_scale",434 "shape": [435 32,436 6144437 ],438 "dtype": "float16",439 "format": "f32-to-bf16",440 "nbytes": 393216,441 "byteOffset": 12582912442 },443 {444 "name": "model.layers.1.self_attn.o_proj.q_weight",445 "shape": [446 512,447 4096448 ],449 "dtype": "uint32",450 "format": "f32-to-bf16",451 "nbytes": 8388608,452 "byteOffset": 12976128453 },454 {455 "name": "model.layers.1.self_attn.o_proj.q_scale",456 "shape": [457 32,458 4096459 ],460 "dtype": "float16",461 "format": "f32-to-bf16",462 "nbytes": 262144,463 "byteOffset": 21364736464 },465 {466 "name": "model.layers.2.input_layernorm.weight",467 "shape": [468 4096469 ],470 "dtype": "float16",471 "format": "f32-to-bf16",472 "nbytes": 8192,473 "byteOffset": 21626880474 },475 {476 "name": "model.layers.2.mlp.down_proj.q_scale",477 "shape": [478 112,479 4096480 ],481 "dtype": "float16",482 "format": "f32-to-bf16",483 "nbytes": 917504,484 "byteOffset": 21635072485 },486 {487 "name": "model.layers.2.mlp.gate_up_proj.q_scale",488 "shape": [489 32,490 28672491 ],492 "dtype": "float16",493 "format": "f32-to-bf16",494 "nbytes": 1835008,495 "byteOffset": 22552576496 },497 {498 "name": "model.layers.2.post_attention_layernorm.weight",499 "shape": [500 4096501 ],502 "dtype": "float16",503 "format": "f32-to-bf16",504 "nbytes": 8192,505 "byteOffset": 24387584506 }507 ],508 "md5sum": "a20d692709a226246d33c62a45574eed"509 },510 {511 "dataPath": "params_shard_13.bin",512 "format": "raw-shard",513 "nbytes": 29360128,514 "records": [515 {516 "name": "model.layers.3.mlp.down_proj.q_weight",517 "shape": [518 1792,519 4096520 ],521 "dtype": "uint32",522 "format": "f32-to-bf16",523 "nbytes": 29360128,524 "byteOffset": 0525 }526 ],527 "md5sum": "ab72611bf2bcee21fdb26bc0b1308486"528 },529 {530 "dataPath": "params_shard_14.bin",531 "format": "raw-shard",532 "nbytes": 58720256,533 "records": [534 {535 "name": "model.layers.3.mlp.gate_up_proj.q_weight",536 "shape": [537 512,538 28672539 ],540 "dtype": "uint32",541 "format": "f32-to-bf16",542 "nbytes": 58720256,543 "byteOffset": 0544 }545 ],546 "md5sum": "17bc5c8a93422268fdae9bf1abcac526"547 },548 {549 "dataPath": "params_shard_15.bin",550 "format": "raw-shard",551 "nbytes": 24395776,552 "records": [553 {554 "name": "model.layers.2.self_attn.qkv_proj.q_weight",555 "shape": [556 512,557 6144558 ],559 "dtype": "uint32",560 "format": "f32-to-bf16",561 "nbytes": 12582912,562 "byteOffset": 0563 },564 {565 "name": "model.layers.2.self_attn.qkv_proj.q_scale",566 "shape": [567 32,568 6144569 ],570 "dtype": "float16",571 "format": "f32-to-bf16",572 "nbytes": 393216,573 "byteOffset": 12582912574 },575 {576 "name": "model.layers.2.self_attn.o_proj.q_weight",577 "shape": [578 512,579 4096580 ],581 "dtype": "uint32",582 "format": "f32-to-bf16",583 "nbytes": 8388608,584 "byteOffset": 12976128585 },586 {587 "name": "model.layers.2.self_attn.o_proj.q_scale",588 "shape": [589 32,590 4096591 ],592 "dtype": "float16",593 "format": "f32-to-bf16",594 "nbytes": 262144,595 "byteOffset": 21364736596 },597 {598 "name": "model.layers.3.input_layernorm.weight",599 "shape": [600 4096601 ],602 "dtype": "float16",603 "format": "f32-to-bf16",604 "nbytes": 8192,605 "byteOffset": 21626880606 },607 {608 "name": "model.layers.3.mlp.down_proj.q_scale",609 "shape": [610 112,611 4096612 ],613 "dtype": "float16",614 "format": "f32-to-bf16",615 "nbytes": 917504,616 "byteOffset": 21635072617 },618 {619 "name": "model.layers.3.mlp.gate_up_proj.q_scale",620 "shape": [621 32,622 28672623 ],624 "dtype": "float16",625 "format": "f32-to-bf16",626 "nbytes": 1835008,627 "byteOffset": 22552576628 },629 {630 "name": "model.layers.3.post_attention_layernorm.weight",631 "shape": [632 4096633 ],634 "dtype": "float16",635 "format": "f32-to-bf16",636 "nbytes": 8192,637 "byteOffset": 24387584638 }639 ],640 "md5sum": "5471c4a4c35aae66366e740a36dabafe"641 },642 {643 "dataPath": "params_shard_16.bin",644 "format": "raw-shard",645 "nbytes": 29360128,646 "records": [647 {648 "name": "model.layers.4.mlp.down_proj.q_weight",649 "shape": [650 1792,651 4096652 ],653 "dtype": "uint32",654 "format": "f32-to-bf16",655 "nbytes": 29360128,656 "byteOffset": 0657 }658 ],659 "md5sum": "d31530fd8966876b7e518bd67d3cf561"660 },661 {662 "dataPath": "params_shard_17.bin",663 "format": "raw-shard",664 "nbytes": 58720256,665 "records": [666 {667 "name": "model.layers.4.mlp.gate_up_proj.q_weight",668 "shape": [669 512,670 28672671 ],672 "dtype": "uint32",673 "format": "f32-to-bf16",674 "nbytes": 58720256,675 "byteOffset": 0676 }677 ],678 "md5sum": "37b3526132f05dcfba8cdfdc91a12c05"679 },680 {681 "dataPath": "params_shard_18.bin",682 "format": "raw-shard",683 "nbytes": 24395776,684 "records": [685 {686 "name": "model.layers.3.self_attn.qkv_proj.q_weight",687 "shape": [688 512,689 6144690 ],691 "dtype": "uint32",692 "format": "f32-to-bf16",693 "nbytes": 12582912,694 "byteOffset": 0695 },696 {697 "name": "model.layers.3.self_attn.qkv_proj.q_scale",698 "shape": [699 32,700 6144701 ],702 "dtype": "float16",703 "format": "f32-to-bf16",704 "nbytes": 393216,705 "byteOffset": 12582912706 },707 {708 "name": "model.layers.3.self_attn.o_proj.q_weight",709 "shape": [710 512,711 4096712 ],713 "dtype": "uint32",714 "format": "f32-to-bf16",715 "nbytes": 8388608,716 "byteOffset": 12976128717 },718 {719 "name": "model.layers.3.self_attn.o_proj.q_scale",720 "shape": [721 32,722 4096723 ],724 "dtype": "float16",725 "format": "f32-to-bf16",726 "nbytes": 262144,727 "byteOffset": 21364736728 },729 {730 "name": "model.layers.4.input_layernorm.weight",731 "shape": [732 4096733 ],734 "dtype": "float16",735 "format": "f32-to-bf16",736 "nbytes": 8192,737 "byteOffset": 21626880738 },739 {740 "name": "model.layers.4.mlp.down_proj.q_scale",741 "shape": [742 112,743 4096744 ],745 "dtype": "float16",746 "format": "f32-to-bf16",747 "nbytes": 917504,748 "byteOffset": 21635072749 },750 {751 "name": "model.layers.4.mlp.gate_up_proj.q_scale",752 "shape": [753 32,754 28672755 ],756 "dtype": "float16",757 "format": "f32-to-bf16",758 "nbytes": 1835008,759 "byteOffset": 22552576760 },761 {762 "name": "model.layers.4.post_attention_layernorm.weight",763 "shape": [764 4096765 ],766 "dtype": "float16",767 "format": "f32-to-bf16",768 "nbytes": 8192,769 "byteOffset": 24387584770 }771 ],772 "md5sum": "50516261632bb66ca22f5652994fd04a"773 },774 {775 "dataPath": "params_shard_19.bin",776 "format": "raw-shard",777 "nbytes": 29360128,778 "records": [779 {780 "name": "model.layers.5.mlp.down_proj.q_weight",781 "shape": [782 1792,783 4096784 ],785 "dtype": "uint32",786 "format": "f32-to-bf16",787 "nbytes": 29360128,788 "byteOffset": 0789 }790 ],791 "md5sum": "1e0534a3f1335ee938e2204ddd900b43"792 },793 {794 "dataPath": "params_shard_20.bin",795 "format": "raw-shard",796 "nbytes": 58720256,797 "records": [798 {799 "name": "model.layers.5.mlp.gate_up_proj.q_weight",800 "shape": [801 512,802 28672803 ],804 "dtype": "uint32",805 "format": "f32-to-bf16",806 "nbytes": 58720256,807 "byteOffset": 0808 }809 ],810 "md5sum": "1cb6842bf82bd7b679020cac1b4f2173"811 },812 {813 "dataPath": "params_shard_21.bin",814 "format": "raw-shard",815 "nbytes": 24395776,816 "records": [817 {818 "name": "model.layers.4.self_attn.qkv_proj.q_weight",819 "shape": [820 512,821 6144822 ],823 "dtype": "uint32",824 "format": "f32-to-bf16",825 "nbytes": 12582912,826 "byteOffset": 0827 },828 {829 "name": "model.layers.4.self_attn.qkv_proj.q_scale",830 "shape": [831 32,832 6144833 ],834 "dtype": "float16",835 "format": "f32-to-bf16",836 "nbytes": 393216,837 "byteOffset": 12582912838 },839 {840 "name": "model.layers.4.self_attn.o_proj.q_weight",841 "shape": [842 512,843 4096844 ],845 "dtype": "uint32",846 "format": "f32-to-bf16",847 "nbytes": 8388608,848 "byteOffset": 12976128849 },850 {851 "name": "model.layers.4.self_attn.o_proj.q_scale",852 "shape": [853 32,854 4096855 ],856 "dtype": "float16",857 "format": "f32-to-bf16",858 "nbytes": 262144,859 "byteOffset": 21364736860 },861 {862 "name": "model.layers.5.input_layernorm.weight",863 "shape": [864 4096865 ],866 "dtype": "float16",867 "format": "f32-to-bf16",868 "nbytes": 8192,869 "byteOffset": 21626880870 },871 {872 "name": "model.layers.5.mlp.down_proj.q_scale",873 "shape": [874 112,875 4096876 ],877 "dtype": "float16",878 "format": "f32-to-bf16",879 "nbytes": 917504,880 "byteOffset": 21635072881 },882 {883 "name": "model.layers.5.mlp.gate_up_proj.q_scale",884 "shape": [885 32,886 28672887 ],888 "dtype": "float16",889 "format": "f32-to-bf16",890 "nbytes": 1835008,891 "byteOffset": 22552576892 },893 {894 "name": "model.layers.5.post_attention_layernorm.weight",895 "shape": [896 4096897 ],898 "dtype": "float16",899 "format": "f32-to-bf16",900 "nbytes": 8192,901 "byteOffset": 24387584902 }903 ],904 "md5sum": "0b72fefff9cf352bd245761c6d1045ae"905 },906 {907 "dataPath": "params_shard_22.bin",908 "format": "raw-shard",909 "nbytes": 29360128,910 "records": [911 {912 "name": "model.layers.6.mlp.down_proj.q_weight",913 "shape": [914 1792,915 4096916 ],917 "dtype": "uint32",918 "format": "f32-to-bf16",919 "nbytes": 29360128,920 "byteOffset": 0921 }922 ],923 "md5sum": "b7e0aa2d4157f4538218a713538d132c"924 },925 {926 "dataPath": "params_shard_23.bin",927 "format": "raw-shard",928 "nbytes": 58720256,929 "records": [930 {931 "name": "model.layers.6.mlp.gate_up_proj.q_weight",932 "shape": [933 512,934 28672935 ],936 "dtype": "uint32",937 "format": "f32-to-bf16",938 "nbytes": 58720256,939 "byteOffset": 0940 }941 ],942 "md5sum": "d87b43bd62f597bb04662b968f60e50c"943 },944 {945 "dataPath": "params_shard_24.bin",946 "format": "raw-shard",947 "nbytes": 24395776,948 "records": [949 {950 "name": "model.layers.5.self_attn.qkv_proj.q_weight",951 "shape": [952 512,953 6144954 ],955 "dtype": "uint32",956 "format": "f32-to-bf16",957 "nbytes": 12582912,958 "byteOffset": 0959 },960 {961 "name": "model.layers.5.self_attn.qkv_proj.q_scale",962 "shape": [963 32,964 6144965 ],966 "dtype": "float16",967 "format": "f32-to-bf16",968 "nbytes": 393216,969 "byteOffset": 12582912970 },971 {972 "name": "model.layers.5.self_attn.o_proj.q_weight",973 "shape": [974 512,975 4096976 ],977 "dtype": "uint32",978 "format": "f32-to-bf16",979 "nbytes": 8388608,980 "byteOffset": 12976128981 },982 {983 "name": "model.layers.5.self_attn.o_proj.q_scale",984 "shape": [985 32,986 4096987 ],988 "dtype": "float16",989 "format": "f32-to-bf16",990 "nbytes": 262144,991 "byteOffset": 21364736992 },993 {994 "name": "model.layers.6.input_layernorm.weight",995 "shape": [996 4096997 ],998 "dtype": "float16",999 "format": "f32-to-bf16",1000 "nbytes": 8192,1001 "byteOffset": 216268801002 },1003 {1004 "name": "model.layers.6.mlp.down_proj.q_scale",1005 "shape": [1006 112,1007 40961008 ],1009 "dtype": "float16",1010 "format": "f32-to-bf16",1011 "nbytes": 917504,1012 "byteOffset": 216350721013 },1014 {1015 "name": "model.layers.6.mlp.gate_up_proj.q_scale",1016 "shape": [1017 32,1018 286721019 ],1020 "dtype": "float16",1021 "format": "f32-to-bf16",1022 "nbytes": 1835008,1023 "byteOffset": 225525761024 },1025 {1026 "name": "model.layers.6.post_attention_layernorm.weight",1027 "shape": [1028 40961029 ],1030 "dtype": "float16",1031 "format": "f32-to-bf16",1032 "nbytes": 8192,1033 "byteOffset": 243875841034 }1035 ],1036 "md5sum": "168b4ceddf3b5a6422c4f858d806172b"1037 },1038 {1039 "dataPath": "params_shard_25.bin",1040 "format": "raw-shard",1041 "nbytes": 29360128,1042 "records": [1043 {1044 "name": "model.layers.7.mlp.down_proj.q_weight",1045 "shape": [1046 1792,1047 40961048 ],1049 "dtype": "uint32",1050 "format": "f32-to-bf16",1051 "nbytes": 29360128,1052 "byteOffset": 01053 }1054 ],1055 "md5sum": "44d91c05db86c646c0259049cf7dfe10"1056 },1057 {1058 "dataPath": "params_shard_26.bin",1059 "format": "raw-shard",1060 "nbytes": 58720256,1061 "records": [1062 {1063 "name": "model.layers.7.mlp.gate_up_proj.q_weight",1064 "shape": [1065 512,1066 286721067 ],1068 "dtype": "uint32",1069 "format": "f32-to-bf16",1070 "nbytes": 58720256,1071 "byteOffset": 01072 }1073 ],1074 "md5sum": "5357c7122720b56b9cb4e8dd654f73e4"1075 },1076 {1077 "dataPath": "params_shard_27.bin",1078 "format": "raw-shard",1079 "nbytes": 24395776,1080 "records": [1081 {1082 "name": "model.layers.6.self_attn.qkv_proj.q_weight",1083 "shape": [1084 512,1085 61441086 ],1087 "dtype": "uint32",1088 "format": "f32-to-bf16",1089 "nbytes": 12582912,1090 "byteOffset": 01091 },1092 {1093 "name": "model.layers.6.self_attn.qkv_proj.q_scale",1094 "shape": [1095 32,1096 61441097 ],1098 "dtype": "float16",1099 "format": "f32-to-bf16",1100 "nbytes": 393216,1101 "byteOffset": 125829121102 },1103 {1104 "name": "model.layers.6.self_attn.o_proj.q_weight",1105 "shape": [1106 512,1107 40961108 ],1109 "dtype": "uint32",1110 "format": "f32-to-bf16",1111 "nbytes": 8388608,1112 "byteOffset": 129761281113 },1114 {1115 "name": "model.layers.6.self_attn.o_proj.q_scale",1116 "shape": [1117 32,1118 40961119 ],1120 "dtype": "float16",1121 "format": "f32-to-bf16",1122 "nbytes": 262144,1123 "byteOffset": 213647361124 },1125 {1126 "name": "model.layers.7.input_layernorm.weight",1127 "shape": [1128 40961129 ],1130 "dtype": "float16",1131 "format": "f32-to-bf16",1132 "nbytes": 8192,1133 "byteOffset": 216268801134 },1135 {1136 "name": "model.layers.7.mlp.down_proj.q_scale",1137 "shape": [1138 112,1139 40961140 ],1141 "dtype": "float16",1142 "format": "f32-to-bf16",1143 "nbytes": 917504,1144 "byteOffset": 216350721145 },1146 {1147 "name": "model.layers.7.mlp.gate_up_proj.q_scale",1148 "shape": [1149 32,1150 286721151 ],1152 "dtype": "float16",1153 "format": "f32-to-bf16",1154 "nbytes": 1835008,1155 "byteOffset": 225525761156 },1157 {1158 "name": "model.layers.7.post_attention_layernorm.weight",1159 "shape": [1160 40961161 ],1162 "dtype": "float16",1163 "format": "f32-to-bf16",1164 "nbytes": 8192,1165 "byteOffset": 243875841166 }1167 ],1168 "md5sum": "c170a56fbf5f694aa84807e9dd225b3c"1169 },1170 {1171 "dataPath": "params_shard_28.bin",1172 "format": "raw-shard",1173 "nbytes": 29360128,1174 "records": [1175 {1176 "name": "model.layers.8.mlp.down_proj.q_weight",1177 "shape": [1178 1792,1179 40961180 ],1181 "dtype": "uint32",1182 "format": "f32-to-bf16",1183 "nbytes": 29360128,1184 "byteOffset": 01185 }1186 ],1187 "md5sum": "423df162ff3d0988dbdb2d4c10477592"1188 },1189 {1190 "dataPath": "params_shard_29.bin",1191 "format": "raw-shard",1192 "nbytes": 58720256,1193 "records": [1194 {1195 "name": "model.layers.8.mlp.gate_up_proj.q_weight",1196 "shape": [1197 512,1198 286721199 ],1200 "dtype": "uint32",