{ "metadata": { "ParamSize": 615, "ParamBytes": 2183177216.0, "BitsPerParam": 4.061649986174953 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 335626240, "records": [ { "name": "language_model.model.embed_tokens.q_weight", "shape": [ 262208, 320 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 335626240, "byteOffset": 0 } ], "md5sum": "e885a0b51b1e7a5feccb63c2016b00f2" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 41953280, "records": [ { "name": "language_model.model.embed_tokens.q_scale", "shape": [ 262208, 80 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 41953280, "byteOffset": 0 } ], "md5sum": "4af7c226d701c5d1c3b8e4da1ade95b5" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "ff2806e383819d9403330d1a7cf6321d" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 26896384, "records": [ { "name": "language_model.model.layers.0.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 0 }, { "name": "language_model.model.layers.0.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 5120 }, { "name": "language_model.model.layers.0.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13112320 }, { "name": "language_model.model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14750720 }, { "name": "language_model.model.layers.0.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.0.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.0.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18037760 }, { "name": "language_model.model.layers.0.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18042880 }, { "name": "language_model.model.layers.0.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18043392 }, { "name": "language_model.model.layers.0.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19354112 }, { "name": "language_model.model.layers.0.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19517952 }, { "name": "language_model.model.layers.0.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22139392 }, { "name": "language_model.model.layers.0.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22467072 }, { "name": "language_model.model.layers.0.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22467584 }, { "name": "language_model.model.layers.0.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25089024 }, { "name": "language_model.model.layers.0.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25416704 }, { "name": "language_model.model.layers.0.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26727424 }, { "name": "language_model.model.layers.1.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26891264 } ], "md5sum": "cf8eceaadf971077f5cee7d05f31138f" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "db1cdced44d837d60d780d4792dd85bd" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.1.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.1.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.1.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.1.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.1.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.1.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.1.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.1.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.1.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.1.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.1.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.1.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.1.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.1.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.1.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.10.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "6506507b183de7cb3f4e5ec6dda10237" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "97a310c35be44980f5bbe81ae69ef0a1" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.10.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.10.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.10.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.10.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.10.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.10.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.10.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.10.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.10.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.10.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.10.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.10.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.10.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.10.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.10.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.11.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "49def4e1c35514cdae7afdaf9e70cab1" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "b4266630302084309cf455c67f330917" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.11.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.11.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.11.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.11.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.11.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.11.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.11.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.11.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.11.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.11.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.11.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.11.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.11.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.11.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.11.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.12.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "543218f6a2bba042e84d814bf4378e91" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "6db8b9c8281bf3d7bafab630c5d0ad8b" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.12.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.12.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.12.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.12.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.12.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.12.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.12.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.12.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.12.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.12.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.12.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.12.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.12.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.12.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.12.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.13.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "e461067686875027bcb48b88d4b8c820" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "60526967e0d7958bb194e328a9374247" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "ec61555f12ff830dc2222742ef50c73d" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 31638016, "records": [ { "name": "language_model.model.layers.13.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.13.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.13.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.13.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.13.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.13.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.13.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.13.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.13.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.13.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.13.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.13.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.13.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.13.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.13.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 26886144 }, { "name": "language_model.model.layers.14.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 30162944 }, { "name": "language_model.model.layers.14.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 30163456 }, { "name": "language_model.model.layers.14.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 31474176 } ], "md5sum": "b3772c1d77ee61b028426d475687b567" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "cd6ec2567475f1380f85cbf3825946b6" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 32790016, "records": [ { "name": "language_model.model.layers.14.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "language_model.model.layers.14.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 2621440 }, { "name": "language_model.model.layers.14.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 2949120 }, { "name": "language_model.model.layers.14.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 2949632 }, { "name": "language_model.model.layers.14.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 5571072 }, { "name": "language_model.model.layers.14.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 5898752 }, { "name": "language_model.model.layers.14.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 7209472 }, { "name": "language_model.model.layers.2.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 7373312 }, { "name": "language_model.model.layers.2.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 7378432 }, { "name": "language_model.model.layers.2.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 20485632 }, { "name": "language_model.model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 22124032 }, { "name": "language_model.model.layers.2.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 25400832 }, { "name": "language_model.model.layers.2.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 25405952 }, { "name": "language_model.model.layers.2.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 25411072 }, { "name": "language_model.model.layers.2.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 25416192 }, { "name": "language_model.model.layers.2.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25416704 }, { "name": "language_model.model.layers.2.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26727424 }, { "name": "language_model.model.layers.2.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 26891264 }, { "name": "language_model.model.layers.2.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 29512704 }, { "name": "language_model.model.layers.2.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29840384 }, { "name": "language_model.model.layers.2.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 29840896 }, { "name": "language_model.model.layers.2.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 32462336 } ], "md5sum": "7b3cf4253b4ebe97e0dde90d5f23fbaf" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "e902b7c82520e2bb28f21112ca74bdfb" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 28370944, "records": [ { "name": "language_model.model.layers.2.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 0 }, { "name": "language_model.model.layers.2.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 1310720 }, { "name": "language_model.model.layers.3.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 1474560 }, { "name": "language_model.model.layers.3.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 1479680 }, { "name": "language_model.model.layers.3.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 14586880 }, { "name": "language_model.model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 16225280 }, { "name": "language_model.model.layers.3.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 19502080 }, { "name": "language_model.model.layers.3.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 19507200 }, { "name": "language_model.model.layers.3.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 19512320 }, { "name": "language_model.model.layers.3.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 19517440 }, { "name": "language_model.model.layers.3.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 19517952 }, { "name": "language_model.model.layers.3.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 20828672 }, { "name": "language_model.model.layers.3.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 20992512 }, { "name": "language_model.model.layers.3.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 23613952 }, { "name": "language_model.model.layers.3.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 23941632 }, { "name": "language_model.model.layers.3.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 23942144 }, { "name": "language_model.model.layers.3.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26563584 }, { "name": "language_model.model.layers.3.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 26891264 }, { "name": "language_model.model.layers.3.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 28201984 }, { "name": "language_model.model.layers.4.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 28365824 } ], "md5sum": "85eff60390130348136de27734635f2d" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "7e3d655051aa1db448800caa29daf565" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.4.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.4.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.4.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.4.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.4.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.4.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.4.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.4.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.4.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.4.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.4.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.4.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.4.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.4.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.4.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.5.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "511a31d6cd521786f2e06c09e3407e79" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "0cbd4236a5e63e1a39033489458243ce" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.5.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.5.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.5.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.5.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.5.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.5.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.5.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.5.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.5.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.5.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.5.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.5.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.5.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.5.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.5.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.6.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "4a60393bd827dcfef0d1f9ce3fa1a1ba" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "791cfa5979805822d71aaa8773aefc60" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.6.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.6.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.6.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.6.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.6.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.6.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.6.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.6.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.6.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.6.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.6.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.6.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.6.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.6.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.6.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.7.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "04b732ff9891a17c8fd9bc9df2982f7c" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "465527f2eebaf7349971bb96f90a474f" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.7.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.7.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.7.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.7.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.7.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.7.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.7.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.7.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.7.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.7.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.7.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.7.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.7.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.7.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.7.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.8.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "681a2ea653c23498c2ac0fa9ca98600c" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "0ae22ecff1afe72780c290cfe3671e55" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.8.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.8.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.8.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.8.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.8.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.8.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.8.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.8.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.8.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.8.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.8.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.8.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.8.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.8.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.8.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.9.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "62d2609622da6faa2bb225d4002a743e" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "e8cd24f8bd88ac7dd4c0be24bd202391" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.9.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.9.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.9.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.9.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.9.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.9.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.9.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.9.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.9.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.9.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.9.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.9.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.9.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.9.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.9.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.14.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "f89b58935a02554810da6769615228f1" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "3017dfa5f964f20db5586966341b22f0" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 32804352, "records": [ { "name": "language_model.model.layers.14.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.14.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.14.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 14745600 }, { "name": "language_model.model.layers.14.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 14750720 }, { "name": "language_model.model.layers.14.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 14755840 }, { "name": "language_model.model.layers.15.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 14760960 }, { "name": "language_model.model.layers.15.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 14766080 }, { "name": "language_model.model.layers.15.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 27873280 }, { "name": "language_model.model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 29511680 }, { "name": "language_model.model.layers.15.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 32788480 }, { "name": "language_model.model.layers.15.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 32793600 }, { "name": "language_model.model.layers.15.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 32798720 }, { "name": "language_model.model.layers.15.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 32803840 } ], "md5sum": "0fe5ff79f7e93fed086d36b90bc2cad3" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "9cd8a44eb200ea7a5507309970f4693f" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 31315456, "records": [ { "name": "language_model.model.layers.15.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 0 }, { "name": "language_model.model.layers.15.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 1310720 }, { "name": "language_model.model.layers.15.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 1474560 }, { "name": "language_model.model.layers.15.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 4096000 }, { "name": "language_model.model.layers.15.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 4423680 }, { "name": "language_model.model.layers.15.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 4424192 }, { "name": "language_model.model.layers.15.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 7045632 }, { "name": "language_model.model.layers.15.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 7373312 }, { "name": "language_model.model.layers.15.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 8684032 }, { "name": "language_model.model.layers.16.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 8847872 }, { "name": "language_model.model.layers.16.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 8852992 }, { "name": "language_model.model.layers.16.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 21960192 }, { "name": "language_model.model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 23598592 }, { "name": "language_model.model.layers.16.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26875392 }, { "name": "language_model.model.layers.16.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26880512 }, { "name": "language_model.model.layers.16.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26885632 }, { "name": "language_model.model.layers.16.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 26890752 }, { "name": "language_model.model.layers.16.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 26891264 }, { "name": "language_model.model.layers.16.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 28201984 }, { "name": "language_model.model.layers.16.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 28365824 }, { "name": "language_model.model.layers.16.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 30987264 }, { "name": "language_model.model.layers.16.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 31314944 } ], "md5sum": "556f4de61deda2c4737112fbdecf28f0" }, { "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "994c6f0cf4ab4b943e9298088ce14bd4" }, { "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 31320064, "records": [ { "name": "language_model.model.layers.16.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "language_model.model.layers.16.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 2621440 }, { "name": "language_model.model.layers.16.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 2949120 }, { "name": "language_model.model.layers.16.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 4259840 }, { "name": "language_model.model.layers.17.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 4423680 }, { "name": "language_model.model.layers.17.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 4428800 }, { "name": "language_model.model.layers.17.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 17536000 }, { "name": "language_model.model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 19174400 }, { "name": "language_model.model.layers.17.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 22451200 }, { "name": "language_model.model.layers.17.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 22456320 }, { "name": "language_model.model.layers.17.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 22461440 }, { "name": "language_model.model.layers.17.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22466560 }, { "name": "language_model.model.layers.17.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 22467072 }, { "name": "language_model.model.layers.17.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 23777792 }, { "name": "language_model.model.layers.17.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 23941632 }, { "name": "language_model.model.layers.17.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26563072 }, { "name": "language_model.model.layers.17.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 26890752 }, { "name": "language_model.model.layers.17.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 26891264 }, { "name": "language_model.model.layers.17.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 29512704 }, { "name": "language_model.model.layers.17.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 29840384 }, { "name": "language_model.model.layers.17.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 31151104 }, { "name": "language_model.model.layers.18.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 31314944 } ], "md5sum": "51ed1b6c3dca91e6ed350ab5fba81507" }, { "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "958c288e002f58fcc1dc19444702006b" }, { "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.18.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.18.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.18.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.18.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.18.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.18.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.18.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.18.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.18.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.18.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.18.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.18.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.18.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.18.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.18.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.19.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "62501849e96250a1cbf4d2a88537067d" }, { "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "ab665606e7ef0789c000defccabd537e" }, { "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.19.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.19.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.19.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.19.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.19.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.19.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.19.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.19.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.19.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.19.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.19.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.19.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.19.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.19.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.19.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.20.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "0ed51c4dde9c3d69ec5924aa67fda487" }, { "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "e7dfdb1fa56e38490c5491bd34fae54d" }, { "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.20.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.20.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.20.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.20.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.20.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.20.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.20.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.20.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.20.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.20.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.20.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.20.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.20.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.20.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.20.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.21.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "1bc2587e62ac8e834d6aab767f84439c" }, { "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "8591c145707d3bbcdcb96a43c0811243" }, { "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.21.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.21.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.21.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.21.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.21.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.21.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.21.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.21.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.21.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.21.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.21.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.21.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.21.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.21.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.21.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.22.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "c8ed4d995b7e64e7f8db860879012a73" }, { "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "caa732ce4655443a5b7feeb08f7db14d" }, { "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.22.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.22.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.22.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.22.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.22.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.22.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.22.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.22.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.22.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.22.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.22.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.22.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.22.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.22.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.22.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.23.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "0577b40953eed7d50a4c5ee725220389" }, { "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "583db2c60fdf02954bfe96f55602c423" }, { "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.23.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.23.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.23.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.23.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.23.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.23.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.23.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.23.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.23.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.23.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.23.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.23.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.23.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.23.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.23.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.24.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "f77ae927de2d7ab4ae194813bb509ab8" }, { "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "275ff93683dbd4cbc6464d7b6dc000f6" }, { "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.24.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.24.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.24.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.24.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.24.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.24.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.24.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.24.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.24.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.24.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.24.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.24.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.24.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.24.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.24.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.25.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "9de71188e878b23c4a24ed5850baba0b" }, { "dataPath": "params_shard_51.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "0b22dd3286c255c55701a523044e6c48" }, { "dataPath": "params_shard_52.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.25.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.25.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.25.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.25.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.25.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.25.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.25.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.25.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.25.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.25.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.25.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.25.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.25.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.25.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.25.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.26.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "e8befa926e8646a43d44898cffa7d3ff" }, { "dataPath": "params_shard_53.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "d4c8a2f0259d2c15f9925b32e5a9c30e" }, { "dataPath": "params_shard_54.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.26.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.26.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.26.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.26.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.26.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.26.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.26.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.26.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.26.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.26.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.26.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.26.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.26.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.26.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.26.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.27.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "0f9068430040fc87860228bfd9d30fa9" }, { "dataPath": "params_shard_55.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "d8e508266d9ea67ce4c4a4d47f79a586" }, { "dataPath": "params_shard_56.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.27.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.27.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.27.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.27.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.27.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.27.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.27.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.27.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.27.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.27.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.27.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.27.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.27.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.27.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.27.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.28.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "e6089e5b1a0b21a2a98a5b5ffda7f2a1" }, { "dataPath": "params_shard_57.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.28.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "3b613761389cc45042b80244edb06156" }, { "dataPath": "params_shard_58.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.28.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.28.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.28.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.28.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.28.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.28.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.28.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.28.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.28.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.28.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.28.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.28.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.28.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.28.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.28.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.28.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.29.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "ef135370b5f47285e4ef6140eda280ce" }, { "dataPath": "params_shard_59.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.29.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "34ac190e0693dc1726562bfce72e549f" }, { "dataPath": "params_shard_60.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.29.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.29.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.29.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.29.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.29.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.29.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.29.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.29.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.29.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.29.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.29.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.29.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.29.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.29.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.29.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.29.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.30.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "c3c85fd20c927606877b25c9bc37409a" }, { "dataPath": "params_shard_61.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.30.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "9aecf564a3f148dc2b760adfea440a5b" }, { "dataPath": "params_shard_62.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.30.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.30.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.30.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.30.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.30.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.30.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.30.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.30.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.30.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.30.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.30.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.30.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.30.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.30.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.30.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.30.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.31.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "4183c0d687c1f554359fafad68a710d8" }, { "dataPath": "params_shard_63.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.31.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "3b6a896db5fa4e9e40f7e7193bb6574e" }, { "dataPath": "params_shard_64.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.31.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.31.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.31.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.31.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.31.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.31.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.31.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.31.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.31.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.31.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.31.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.31.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.31.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.31.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.31.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.31.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.32.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "15505407e5266933c939a2f768a43090" }, { "dataPath": "params_shard_65.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.32.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "88c8b5e3cba7b380a807c64e11eb9126" }, { "dataPath": "params_shard_66.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.32.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.32.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.32.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.32.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.32.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.32.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.32.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.32.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.32.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.32.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.32.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.32.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.32.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.32.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.32.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.32.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.layers.33.input_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "4d546ffd26dd65e111f29d7f4630c007" }, { "dataPath": "params_shard_67.bin", "format": "raw-shard", "nbytes": 26214400, "records": [ { "name": "language_model.model.layers.33.mlp.gate_up_proj.q_weight", "shape": [ 320, 20480 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 26214400, "byteOffset": 0 } ], "md5sum": "bab5e99b95168d80e4b9706b22c734fb" }, { "dataPath": "params_shard_68.bin", "format": "raw-shard", "nbytes": 26891264, "records": [ { "name": "language_model.model.layers.33.mlp.down_proj.q_weight", "shape": [ 1280, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13107200, "byteOffset": 0 }, { "name": "language_model.model.layers.33.mlp.down_proj.q_scale", "shape": [ 320, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 1638400, "byteOffset": 13107200 }, { "name": "language_model.model.layers.33.mlp.gate_up_proj.q_scale", "shape": [ 80, 20480 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 3276800, "byteOffset": 14745600 }, { "name": "language_model.model.layers.33.post_attention_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18022400 }, { "name": "language_model.model.layers.33.post_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18027520 }, { "name": "language_model.model.layers.33.pre_feedforward_layernorm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 18032640 }, { "name": "language_model.model.layers.33.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 18037760 }, { "name": "language_model.model.layers.33.self_attn.k_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 18038272 }, { "name": "language_model.model.layers.33.self_attn.k_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 19348992 }, { "name": "language_model.model.layers.33.self_attn.o_proj.q_weight", "shape": [ 256, 2560 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 19512832 }, { "name": "language_model.model.layers.33.self_attn.o_proj.q_scale", "shape": [ 64, 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 22134272 }, { "name": "language_model.model.layers.33.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 22461952 }, { "name": "language_model.model.layers.33.self_attn.q_proj.q_weight", "shape": [ 320, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 22462464 }, { "name": "language_model.model.layers.33.self_attn.q_proj.q_scale", "shape": [ 80, 2048 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 25083904 }, { "name": "language_model.model.layers.33.self_attn.v_proj.q_weight", "shape": [ 320, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1310720, "byteOffset": 25411584 }, { "name": "language_model.model.layers.33.self_attn.v_proj.q_scale", "shape": [ 80, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 163840, "byteOffset": 26722304 }, { "name": "language_model.model.norm.weight", "shape": [ 2560 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 5120, "byteOffset": 26886144 } ], "md5sum": "6cbb3f39c7375f59f52ebbd25e85d8b7" } ] }