{ "metadata": { "ParamSize": 471, "ParamBytes": 562628864.0, "BitsPerParam": 4.5015443041247964 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 150994944, "records": [ { "name": "language_model.model.embed_tokens.q_weight", "shape": [ 262144, 144 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 150994944, "byteOffset": 0 } ], "md5sum": "b9049304503b42ea51e15ef46e388e32" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 33150976, "records": [ { "name": "language_model.model.embed_tokens.q_scale", "shape": [ 262144, 36 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18874368, "byteOffset": 0 }, { "name": "language_model.model.layers.0.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 18874368 }, { "name": "language_model.model.layers.0.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 18876672 }, { "name": "language_model.model.layers.0.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 22857984 }, { "name": "language_model.model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 23355648 }, { "name": "language_model.model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 31318272 }, { "name": "language_model.model.layers.0.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 32313600 }, { "name": "language_model.model.layers.0.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 32315904 }, { "name": "language_model.model.layers.0.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 32318208 }, { "name": "language_model.model.layers.0.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 32320512 }, { "name": "language_model.model.layers.0.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 32321024 }, { "name": "language_model.model.layers.0.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 32468480 }, { "name": "language_model.model.layers.0.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 32486912 }, { "name": "language_model.model.layers.0.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 33076736 }, { "name": "language_model.model.layers.0.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 33150464 } ], "md5sum": "8222e64e873cfca0c12f742aa86350b9" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 31043840, "records": [ { "name": "language_model.model.layers.0.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 0 }, { "name": "language_model.model.layers.0.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 589824 }, { "name": "language_model.model.layers.0.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 663552 }, { "name": "language_model.model.layers.0.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 811008 }, { "name": "language_model.model.layers.1.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 829440 }, { "name": "language_model.model.layers.1.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 831744 }, { "name": "language_model.model.layers.1.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 4813056 }, { "name": "language_model.model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 5310720 }, { "name": "language_model.model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 13273344 }, { "name": "language_model.model.layers.1.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 14268672 }, { "name": "language_model.model.layers.1.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 14270976 }, { "name": "language_model.model.layers.1.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 14273280 }, { "name": "language_model.model.layers.1.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14275584 }, { "name": "language_model.model.layers.1.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14276096 }, { "name": "language_model.model.layers.1.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 14423552 }, { "name": "language_model.model.layers.1.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14441984 }, { "name": "language_model.model.layers.1.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 15031808 }, { "name": "language_model.model.layers.1.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 15105536 }, { "name": "language_model.model.layers.1.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 15106048 }, { "name": "language_model.model.layers.1.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 15695872 }, { "name": "language_model.model.layers.1.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 15769600 }, { "name": "language_model.model.layers.1.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15917056 }, { "name": "language_model.model.layers.10.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15935488 }, { "name": "language_model.model.layers.10.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15937792 }, { "name": "language_model.model.layers.10.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19919104 }, { "name": "language_model.model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 20416768 }, { "name": "language_model.model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 28379392 }, { "name": "language_model.model.layers.10.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 29374720 }, { "name": "language_model.model.layers.10.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 29377024 }, { "name": "language_model.model.layers.10.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 29379328 }, { "name": "language_model.model.layers.10.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29381632 }, { "name": "language_model.model.layers.10.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 29382144 }, { "name": "language_model.model.layers.10.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 29529600 }, { "name": "language_model.model.layers.10.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29548032 }, { "name": "language_model.model.layers.10.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 30137856 }, { "name": "language_model.model.layers.10.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 30211584 }, { "name": "language_model.model.layers.10.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 30212096 }, { "name": "language_model.model.layers.10.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 30801920 }, { "name": "language_model.model.layers.10.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30875648 }, { "name": "language_model.model.layers.10.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 31023104 }, { "name": "language_model.model.layers.11.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 31041536 } ], "md5sum": "d80d17e3cf9bdc7872a21166a6bd045f" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.11.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.11.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.11.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.11.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.11.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.11.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.11.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.11.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.11.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.11.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.11.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.11.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.11.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.11.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.11.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.12.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.12.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.12.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.12.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.12.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.12.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.12.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.12.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.12.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.12.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.12.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.12.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.12.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.12.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.12.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.12.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.13.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "087e489856704febf155260519b7e51f" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.13.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.13.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.13.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.13.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.13.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.13.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.13.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.13.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.13.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.13.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.13.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.13.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.13.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.13.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.13.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.14.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.14.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.14.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.14.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.14.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.14.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.14.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.14.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.14.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.14.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.14.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.14.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.14.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.14.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.14.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.14.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.15.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "1de64244d86f8202fab36996a4d8b9a8" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.15.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.15.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.15.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.15.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.15.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.15.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.15.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.15.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.15.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.15.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.15.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.15.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.15.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.15.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.15.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.16.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.16.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.16.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.16.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.16.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.16.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.16.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.16.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.16.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.16.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.16.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.16.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.16.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.16.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.16.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.16.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.17.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "95e32d7d7ed80a09cace2511627e2ec1" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.17.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.17.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.17.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.17.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.17.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.17.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.17.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.17.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.17.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.17.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.17.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.17.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.17.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.17.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.17.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.18.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.18.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.18.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.18.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.18.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.18.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.18.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.18.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.18.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.18.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.18.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.18.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.18.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.18.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.18.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.18.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.19.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "89a474addbec463f7f7001d2dda9cbe5" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.19.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.19.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.19.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.19.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.19.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.19.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.19.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.19.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.19.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.19.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.19.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.19.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.19.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.19.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.19.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.2.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.2.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.2.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.2.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.2.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.2.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.2.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.2.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.2.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.2.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.2.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.2.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.2.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.2.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.2.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.2.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.20.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "972000954263307de1b089af6880a691" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.20.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.20.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.20.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.20.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.20.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.20.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.20.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.20.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.20.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.20.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.20.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.20.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.20.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.20.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.20.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.21.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.21.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.21.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.21.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.21.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.21.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.21.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.21.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.21.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.21.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.21.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.21.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.21.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.21.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.21.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.21.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.22.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "d167a35cf8fe1fde2255ec79221cc423" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.22.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.22.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.22.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.22.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.22.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.22.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.22.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.22.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.22.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.22.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.22.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.22.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.22.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.22.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.22.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.23.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.23.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.23.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.23.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.23.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.23.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.23.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.23.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.23.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.23.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.23.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.23.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.23.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.23.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.23.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.23.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.24.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "037509f7bfc15a98bffd2a1a5157cd9d" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.24.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.24.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.24.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.24.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.24.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.24.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.24.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.24.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.24.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.24.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.24.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.24.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.24.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.24.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.24.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.25.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.25.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.25.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.25.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.25.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.25.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.25.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.25.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.25.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.25.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.25.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.25.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.25.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.25.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.25.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.25.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.3.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "f77b60459fa1fd75411df1dcb2faeb18" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.3.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.3.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.3.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.3.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.3.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.3.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.3.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.3.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.3.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.3.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.3.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.3.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.3.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.3.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.3.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.4.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.4.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.4.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.4.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.4.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.4.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.4.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.4.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.4.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.4.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.4.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.4.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.4.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.4.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.4.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.4.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.5.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "567fc23d09bae362fe7138aaddf7a56a" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.5.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.5.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.5.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.5.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.5.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.5.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.5.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.5.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.5.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.5.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.5.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.5.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.5.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.5.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.5.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.6.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.6.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.6.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.6.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.6.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.6.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.6.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.6.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.6.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.6.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.6.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.6.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.6.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.6.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.6.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.6.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.7.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "d1bac4417edb9ae1bf61de6cbcdc7029" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 30212096, "records": [ { "name": "language_model.model.layers.7.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.7.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.7.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.7.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.7.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.7.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.7.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.7.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.7.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.7.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.7.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.7.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.7.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.7.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.7.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.layers.8.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 }, { "name": "language_model.model.layers.8.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 15106048 }, { "name": "language_model.model.layers.8.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 19087360 }, { "name": "language_model.model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 19585024 }, { "name": "language_model.model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 27547648 }, { "name": "language_model.model.layers.8.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28542976 }, { "name": "language_model.model.layers.8.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28545280 }, { "name": "language_model.model.layers.8.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28547584 }, { "name": "language_model.model.layers.8.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 28549888 }, { "name": "language_model.model.layers.8.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 28550400 }, { "name": "language_model.model.layers.8.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 28697856 }, { "name": "language_model.model.layers.8.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 28716288 }, { "name": "language_model.model.layers.8.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29306112 }, { "name": "language_model.model.layers.8.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29379840 }, { "name": "language_model.model.layers.8.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29380352 }, { "name": "language_model.model.layers.8.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 29970176 }, { "name": "language_model.model.layers.8.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 30043904 }, { "name": "language_model.model.layers.8.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 30191360 }, { "name": "language_model.model.layers.9.input_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 30209792 } ], "md5sum": "79cc0eb0ab5f46e6e9b2b9fa94c05a5b" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 15106048, "records": [ { "name": "language_model.model.layers.9.mlp.down_proj.q_weight", "shape": [ 864, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3981312, "byteOffset": 0 }, { "name": "language_model.model.layers.9.mlp.down_proj.q_scale", "shape": [ 216, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 497664, "byteOffset": 3981312 }, { "name": "language_model.model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 144, 13824 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 4478976 }, { "name": "language_model.model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 36, 13824 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 12441600 }, { "name": "language_model.model.layers.9.post_attention_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13436928 }, { "name": "language_model.model.layers.9.post_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13439232 }, { "name": "language_model.model.layers.9.pre_feedforward_layernorm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 13441536 }, { "name": "language_model.model.layers.9.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 13443840 }, { "name": "language_model.model.layers.9.self_attn.k_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 13444352 }, { "name": "language_model.model.layers.9.self_attn.k_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 13591808 }, { "name": "language_model.model.layers.9.self_attn.o_proj.q_weight", "shape": [ 128, 1152 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 13610240 }, { "name": "language_model.model.layers.9.self_attn.o_proj.q_scale", "shape": [ 32, 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14200064 }, { "name": "language_model.model.layers.9.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 14273792 }, { "name": "language_model.model.layers.9.self_attn.q_proj.q_weight", "shape": [ 144, 1024 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 14274304 }, { "name": "language_model.model.layers.9.self_attn.q_proj.q_scale", "shape": [ 36, 1024 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 73728, "byteOffset": 14864128 }, { "name": "language_model.model.layers.9.self_attn.v_proj.q_weight", "shape": [ 144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 147456, "byteOffset": 14937856 }, { "name": "language_model.model.layers.9.self_attn.v_proj.q_scale", "shape": [ 36, 256 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 18432, "byteOffset": 15085312 }, { "name": "language_model.model.norm.weight", "shape": [ 1152 ], "dtype": "bfloat16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 15103744 } ], "md5sum": "87f82e57abefee78570018129c2110a6" } ] }