diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,4015 @@ +{ + "metadata": { + "ParamSize": 313, + "ParamBytes": 3439746048.0, + "BitsPerParam": 3.6133605651807272 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 218972160, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 152064, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 218972160, + "byteOffset": 0 + } + ], + "md5sum": "b4fc39160c5d13cf63beb04be6da6974" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "2d561d8b5153af77f4b177385621b960" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "a3b467ba19b3b237ab15f7d36d617394" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 30776320, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 152064, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 27371520, + "byteOffset": 0 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 27371520 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 27378688 + } + ], + "md5sum": "d5afab1c0f90905466d9d7551560db72" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "33ae4379385644dfae9eb6c1277ff4cf" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "dc2ac2d2fca2215d90f71654a21b3f29" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 30339072, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 6827008 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 11987968 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 12633088 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 12640256 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16037888 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 22857728 + }, + { + "name": "model.layers.20.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 22864896 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 22874112 + }, + { + "name": "model.layers.20.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 29509632 + } + ], + "md5sum": "51fb81e215bfb6d41d4800c35d06b55a" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "65681f9d8a4d803fc283abbba4bdf10f" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "4eacd52be0910b9068c40072ee98d18e" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "c8bbfe069e662e3ecf23d13219e9cdf8" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "3a1c0e93b64e4bc06ccf62d77dc084f7" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 26909696, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.21.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.21.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23504896 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 23512064 + } + ], + "md5sum": "16fe516adb4c90f3598111912d1fe552" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "97ebbb2accbc8cfb1e3aa3cec00acf07" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "5dadc5b949b864bd29f52d9ce6163b7f" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 30348288, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.22.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.22.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 20107264 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 20114432 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 23512064 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 30331904 + }, + { + "name": "model.layers.23.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 30339072 + } + ], + "md5sum": "8cdaf158e25684e3259f3a7d11885cd6" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "b8cb47f18acecf91f19d4686a57b734d" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "10fcc9cf4bfb9fe739f94bf70ddd8e1a" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 30977024, + "records": [ + { + "name": "model.layers.23.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 0 + }, + { + "name": "model.layers.23.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 6635520 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 7464960 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 12625920 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13271040 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 13278208 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16675840 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23495680 + }, + { + "name": "model.layers.24.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 23502848 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 23512064 + }, + { + "name": "model.layers.24.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 30147584 + } + ], + "md5sum": "2aecbeda2ca5b9a7568fd2efbf185372" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "615c85f8928381187355934b0d2f9279" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "ce51019d2157833ef0c49ae5fe84f015" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "65f4576995c60fafdbd50cfdb03d933b" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "33cd5958082609eb3641970450a0b51c" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 26909696, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.25.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.25.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23504896 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 23512064 + } + ], + "md5sum": "c8a206944aedb223a5cfd35b10e5ef28" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "1b13d05efd8f61428213a346587849c8" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "69a91cb98369a2f2f11ba241f3d4a8c7" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 30348288, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.26.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.26.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 20107264 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 20114432 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 23512064 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 30331904 + }, + { + "name": "model.layers.27.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 30339072 + } + ], + "md5sum": "be683f8ee58ab0897ff72f5bcad1cabf" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 218972160, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 152064, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 218972160, + "byteOffset": 0 + } + ], + "md5sum": "6eed88c33f805d5bfab2cda7a468fccd" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 27371520, + "records": [ + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 152064, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 27371520, + "byteOffset": 0 + } + ], + "md5sum": "4f4368e3531e08689eef3e74d4e44dcf" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "239c75055c4fb495658a4ac3d5a07376" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "68afb29c0aaf687cdfa0573457670c13" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 30984192, + "records": [ + { + "name": "model.layers.27.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 0 + }, + { + "name": "model.layers.27.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 6635520 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 7464960 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 12625920 + }, + { + "name": "model.norm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13271040 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13278208 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 13285376 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16683008 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23502848 + }, + { + "name": "model.layers.0.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 23510016 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 23519232 + }, + { + "name": "model.layers.0.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 30154752 + } + ], + "md5sum": "41bfee786a991f6b8f09daeb5412c9d0" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "b393618ce5f723676c06e84a03900e18" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "042fdc2cefa7587db1a3225ded6dc7d2" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "907475299861cec1e845a1975825a400" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "67ead573838c08de78f9bf64c4f12344" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 26909696, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.1.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.1.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23504896 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 23512064 + } + ], + "md5sum": "e3a8c8015714c2d2da82091357b7717b" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "939be3b3c97611a00376fb8214d26751" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "ee6777e13a7815d925ef649f4759edfa" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 30348288, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.10.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.10.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 20107264 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 20114432 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 23512064 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 30331904 + }, + { + "name": "model.layers.11.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 30339072 + } + ], + "md5sum": "39c0c9ce88151a51153565fd1e20fd00" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "888a47fd5ed83111e3fff42e1228bd58" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "031a1b06f9148f66a84a1fdd9bccfdf3" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 30977024, + "records": [ + { + "name": "model.layers.11.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 6635520 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 7464960 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 12625920 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13271040 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 13278208 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16675840 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23495680 + }, + { + "name": "model.layers.12.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 23502848 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 23512064 + }, + { + "name": "model.layers.12.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 30147584 + } + ], + "md5sum": "ed8eea457399a9795178d4c06435ad96" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "5727feb7dce7d8c825ebd62c3fef0a73" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "3650e7b90d9f058921baac2356fd30c0" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "32086e3e6b7d1ad88991b0bac36249c3" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "ad54adde49211ebbf90aae6eb860644d" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 26909696, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.13.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.13.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23504896 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 23512064 + } + ], + "md5sum": "cdd0c0962aba4e58bb8132868efedc64" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "e4498d56c4d053483957eee5f939911c" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "848af1e43e4813ca4f6788a98464db03" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 30348288, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.14.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.14.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 20107264 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 20114432 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 23512064 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 30331904 + }, + { + "name": "model.layers.15.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 30339072 + } + ], + "md5sum": "920c863af668e2b339a2d9fd1feb0b1c" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "4fcb8a1363c917a338591d6eb5960c31" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "c818acd9849724ee3a5444015b2b6a3f" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 30977024, + "records": [ + { + "name": "model.layers.15.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 0 + }, + { + "name": "model.layers.15.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 6635520 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 7464960 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 12625920 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13271040 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 13278208 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16675840 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23495680 + }, + { + "name": "model.layers.16.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 23502848 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 23512064 + }, + { + "name": "model.layers.16.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 30147584 + } + ], + "md5sum": "45e7be01822be63c11795abd05221ed2" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "602f216e1760de5a7eca801cc5eab687" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "7b09065354e9b243b197460e9bfddcd4" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "3c9e065a3e9cd4bb1b24ff82906246b3" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "208922f688d0dd64819ad54b7dcfdf01" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 26909696, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.17.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.17.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23504896 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 23512064 + } + ], + "md5sum": "3b2725f9be3bdfade2d53d104b7cdd84" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "781dde45a2054689005d6d6892d75676" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "fa77c72865126460156eb14b020798b7" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 30986240, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.18.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.18.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.19.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 20107264 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 20116480 + }, + { + "name": "model.layers.19.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 26752000 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 27581440 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 27588608 + } + ], + "md5sum": "9d4b24f756e1475fd17b7a89f1cf4ab5" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "4fd332b49d620e42dc04f7315e2d260e" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "7c287b9a13e4083a3d2f676880be1c75" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 30348288, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.2.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.2.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 20107264 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 20114432 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 23512064 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 30331904 + }, + { + "name": "model.layers.3.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 30339072 + } + ], + "md5sum": "a41dfe804aa359fcf5bc831b0812a8a2" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "88d98853742945b160c70bb1646dafb8" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "ee226420f0671dc1b0d7349d69f43d40" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 30977024, + "records": [ + { + "name": "model.layers.3.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 0 + }, + { + "name": "model.layers.3.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 6635520 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 7464960 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 12625920 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13271040 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 13278208 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16675840 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23495680 + }, + { + "name": "model.layers.4.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 23502848 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 23512064 + }, + { + "name": "model.layers.4.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 30147584 + } + ], + "md5sum": "d30aa67e9745914b9a027662125873e0" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "bfcfd1893e04b52d348b3ca8683be209" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "e547b8d8d8425d2c3d08fb1caa6947cd" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "40f7480fd1dbe5eee62014b85fbbcb2c" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "aa7372e642e12dbb5aecf98e421e2b16" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 26909696, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.5.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.5.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23504896 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 23512064 + } + ], + "md5sum": "e13e6bbf32c2c41dcd351883a95747d8" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "86d6137dc9e935b8beea2f515c5d921d" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "c3b16dee726e2be74ead249fecce9ccd" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 30348288, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 0 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 6819840 + }, + { + "name": "model.layers.6.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 6827008 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 6836224 + }, + { + "name": "model.layers.6.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 13471744 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 14301184 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 19462144 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 20107264 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 20114432 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 23512064 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 30331904 + }, + { + "name": "model.layers.7.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 30339072 + } + ], + "md5sum": "9f24b9451c58b830230514814b91c59a" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 27181056, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 0 + } + ], + "md5sum": "68750e70312a5c7ec9f2964a6f826da8" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "c0e4f0965d619ebab3269ec24c3fc820" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 30977024, + "records": [ + { + "name": "model.layers.7.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 0 + }, + { + "name": "model.layers.7.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 6635520 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 7464960 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 12625920 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 13271040 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 13278208 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 16675840 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 23495680 + }, + { + "name": "model.layers.8.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 23502848 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 23512064 + }, + { + "name": "model.layers.8.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 30147584 + } + ], + "md5sum": "39fea896add6b9d50fe46251438b1ebc" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 32994304, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 5160960 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 5806080 + }, + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 3584, + 1896 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 27181056, + "byteOffset": 5813248 + } + ], + "md5sum": "f2b0c6d89b3f015dc58d6b18b66c29b9" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 54558720, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 37888, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 54558720, + "byteOffset": 0 + } + ], + "md5sum": "84b1d749ef66443542f5f619de402ab3" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 23504896, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 3584, + 474 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 3397632, + "byteOffset": 0 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 37888, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6819840, + "byteOffset": 3397632 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 3584 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 7168, + "byteOffset": 10217472 + }, + { + "name": "model.layers.9.self_attn.c_attn.bias", + "shape": [ + 4608 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 9216, + "byteOffset": 10224640 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_weight", + "shape": [ + 4608, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6635520, + "byteOffset": 10233856 + }, + { + "name": "model.layers.9.self_attn.c_attn.q_scale", + "shape": [ + 4608, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 829440, + "byteOffset": 16869376 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 3584, + 360 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5160960, + "byteOffset": 17698816 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 3584, + 90 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 645120, + "byteOffset": 22859776 + } + ], + "md5sum": "0926df2f75e607591fe1c668a8942536" + } + ] +} \ No newline at end of file