{ "metadata": { "ParamSize": 898, "ParamBytes": 2946802496.0, "BitsPerParam": 7.478748540353469 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 28214784, "records": [ { "name": "llm.model.layers.36.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 0 }, { "name": "llm.model.layers.36.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 4608 }, { "name": "llm.model.layers.36.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6640128 }, { "name": "llm.model.layers.36.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7469568 }, { "name": "llm.model.layers.36.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8299008 }, { "name": "llm.model.layers.36.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21570048 }, { "name": "llm.model.layers.36.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23228928 }, { "name": "llm.model.layers.36.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24887808 }, { "name": "llm.model.layers.36.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 24892416 }, { "name": "llm.model.layers.36.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 27546624 }, { "name": "llm.model.layers.36.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 27878400 }, { "name": "llm.model.layers.37.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 28210176 } ], "md5sum": "f8e67c90b09aac2224d5904e63e58c0e" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.37.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.37.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.37.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.37.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.37.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.37.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.37.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.37.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "0396e3fb3c2e5390da2bd35437bd06c1" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.37.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.37.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.37.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.37.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.37.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.38.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.38.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.38.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.38.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.38.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.38.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.38.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.38.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "11a00c2997f26c892a74297db6fea4ac" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.38.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.38.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.38.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.38.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.38.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.38.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.39.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.39.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.39.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.39.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "ef0e9206b7a37cc3fbeac9010d160a8a" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 29882880, "records": [ { "name": "llm.model.layers.39.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.39.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.39.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.39.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.39.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.39.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.39.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.39.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.39.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.39.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.norm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 }, { "name": "resampler.attn.q_proj.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29869056 }, { "name": "resampler.attn.k_proj.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29873664 }, { "name": "resampler.attn.v_proj.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29878272 } ], "md5sum": "7daf6065d76a70a1f5b5f822f0c2ff6e" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 31855104, "records": [ { "name": "resampler.attn.q_proj.weight", "shape": [ 2304, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10616832, "byteOffset": 0 }, { "name": "resampler.attn.k_proj.weight", "shape": [ 2304, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10616832, "byteOffset": 10616832 }, { "name": "resampler.attn.v_proj.weight", "shape": [ 2304, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10616832, "byteOffset": 21233664 }, { "name": "resampler.attn.out_proj.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 31850496 } ], "md5sum": "2b1c44924657428b1fb1b61f37568346" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 31002624, "records": [ { "name": "resampler.attn.out_proj.weight", "shape": [ 2304, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10616832, "byteOffset": 0 }, { "name": "resampler.kv_proj.weight", "shape": [ 2304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5308416, "byteOffset": 10616832 }, { "name": "resampler.ln_kv.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 15925248 }, { "name": "resampler.ln_kv.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 15929856 }, { "name": "resampler.ln_post.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 15934464 }, { "name": "resampler.ln_post.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 15939072 }, { "name": "resampler.ln_q.bias", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 15943680 }, { "name": "resampler.ln_q.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 15948288 }, { "name": "resampler.pos_embed_k", "shape": [ 256, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1179648, "byteOffset": 15952896 }, { "name": "resampler.pos_embed", "shape": [ 64, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 294912, "byteOffset": 17132544 }, { "name": "resampler.proj", "shape": [ 2304, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10616832, "byteOffset": 17427456 }, { "name": "resampler.query", "shape": [ 64, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 294912, "byteOffset": 28044288 }, { "name": "vpm.blocks.0.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 28339200 }, { "name": "vpm.blocks.0.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 28341504 }, { "name": "vpm.blocks.0.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30995712 } ], "md5sum": "8c85b7bd0eead67f22ead18c9077ff53" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.0.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.0.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.0.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.0.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.0.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.0.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.0.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.0.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.0.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.1.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.1.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.1.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "148154ffdd9609f708ea350a81447b4e" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.1.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.1.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.1.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.1.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.1.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.1.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.1.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.1.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.1.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.10.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.10.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.10.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "ff874568c12fe5939349233756af547b" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.10.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.10.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.10.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.10.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.10.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.10.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.10.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.10.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.10.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.11.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.11.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.11.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "b2d9b7389d8659c55bb0170fedc6a7c6" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.11.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.11.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.11.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.11.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.11.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.11.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.11.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.11.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.11.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.12.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.12.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.12.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "0be2d0e1c41e3d8682563120ea67d256" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.12.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.12.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.12.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.12.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.12.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.12.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.12.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.12.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.12.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.13.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.13.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.13.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "43ac6fdb63df609bf5c5b931ebfd02be" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.13.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.13.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.13.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.13.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.13.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.13.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.13.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.13.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.13.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.14.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.14.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.14.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "1b73e96ed7955451d7330c3e239a029d" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.14.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.14.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.14.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.14.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.14.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.14.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.14.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.14.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.14.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.15.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.15.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.15.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "90a63aa3cbe6aa65d8015c5d04c7df81" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.15.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.15.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.15.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.15.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.15.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.15.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.15.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.15.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.15.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.16.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.16.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.16.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "13deb7bba1d6cf863c267af36ed5c328" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.16.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.16.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.16.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.16.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.16.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.16.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.16.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.16.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.16.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.17.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.17.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.17.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "f57edbac216a7abc1c7245d760dce3ae" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.17.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.17.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.17.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.17.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.17.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.17.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.17.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.17.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.17.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.18.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.18.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.18.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "01a87a1e2b9786b1f4f7bc035d1803a8" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.18.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.18.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.18.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.18.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.18.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.18.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.18.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.18.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.18.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.19.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.19.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.19.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "d0f9d6766e8431dd36ad01804680ed57" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.19.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.19.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.19.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.19.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.19.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.19.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.19.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.19.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.19.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.2.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.2.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.2.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "c3e939ecc6766101890195f9306541a7" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.2.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.2.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.2.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.2.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.2.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.2.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.2.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.2.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.2.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.20.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.20.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.20.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "25ba703b48b99e3958a8c6e702260ac6" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.20.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.20.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.20.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.20.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.20.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.20.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.20.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.20.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.20.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.21.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.21.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.21.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "f978226ef7ffc676b7e378522c2f5d86" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.21.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.21.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.21.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.21.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.21.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.21.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.21.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.21.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.21.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.22.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.22.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.22.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "080440b5277f12b1a7a930179c07d844" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.22.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.22.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.22.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.22.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.22.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.22.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.22.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.22.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.22.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.23.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.23.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.23.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "92fd2ed2d8f130fa7b1f7b113e64f241" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.23.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.23.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.23.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.23.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.23.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.23.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.23.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.23.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.23.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.24.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.24.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.24.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "93cd0cdbc450f195c4e55677ae9ca295" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.24.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.24.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.24.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.24.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.24.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.24.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.24.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.24.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.24.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.25.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.25.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.25.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "cd8b3cb526ddd49ceff2a2896d0972af" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.25.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.25.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.25.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.25.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.25.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.25.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.25.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.25.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.25.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.3.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.3.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.3.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "439ac788d351c31e6f79c2295d29aa6f" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.3.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.3.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.3.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.3.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.3.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.3.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.3.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.3.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.3.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.4.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.4.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.4.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "d10909380d314a2183f5fc41159f00cb" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.4.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.4.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.4.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.4.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.4.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.4.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.4.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.4.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.4.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.5.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.5.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.5.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "27c258b5c7236a6c2350c509ca909a62" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.5.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.5.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.5.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.5.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.5.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.5.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.5.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.5.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.5.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.6.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.6.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.6.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "310e92af988bfd0044aaa1254816ca45" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.6.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.6.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.6.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.6.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.6.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.6.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.6.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.6.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.6.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.7.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.7.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.7.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "f5e5eb08f57b993dd2e9072bfeda8c91" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.7.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.7.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.7.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.7.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.7.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.7.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.7.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.7.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.7.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.8.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.8.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.8.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "c15edd322f242ec96a3111bf8e438169" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 30479008, "records": [ { "name": "vpm.blocks.8.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.8.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.8.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.8.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.8.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.8.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.8.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.8.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.8.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.blocks.9.attn.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.blocks.9.attn.proj.weight", "shape": [ 1152, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 27817888 }, { "name": "vpm.blocks.9.attn.qkv.bias", "shape": [ 3456 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 6912, "byteOffset": 30472096 } ], "md5sum": "7dcdf464555a600541ad94f5250da0e4" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 565645824, "records": [ { "name": "llm.model.embed_tokens.weight", "shape": [ 122753, 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 565645824, "byteOffset": 0 } ], "md5sum": "9c9e93be3d2b223f6cf7126456b43b3d" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 29771680, "records": [ { "name": "vpm.blocks.9.attn.qkv.weight", "shape": [ 3456, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "vpm.blocks.9.mlp.fc1.bias", "shape": [ 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 8608, "byteOffset": 7962624 }, { "name": "vpm.blocks.9.mlp.fc1.weight", "shape": [ 4304, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 7971232 }, { "name": "vpm.blocks.9.mlp.fc2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 17887648 }, { "name": "vpm.blocks.9.mlp.fc2.weight", "shape": [ 1152, 4304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9916416, "byteOffset": 17889952 }, { "name": "vpm.blocks.9.norm1.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27806368 }, { "name": "vpm.blocks.9.norm1.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27808672 }, { "name": "vpm.blocks.9.norm2.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27810976 }, { "name": "vpm.blocks.9.norm2.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27813280 }, { "name": "vpm.norm.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27815584 }, { "name": "vpm.norm.weight", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27817888 }, { "name": "vpm.patch_embed.proj.bias", "shape": [ 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2304, "byteOffset": 27820192 }, { "name": "vpm.patch_embed.proj.weight", "shape": [ 1152, 3, 14, 14 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1354752, "byteOffset": 27822496 }, { "name": "vpm.pos_embed", "shape": [ 1, 256, 1152 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 589824, "byteOffset": 29177248 }, { "name": "llm.model.layers.0.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29767072 } ], "md5sum": "2545431e3e4a7829e259a041a05fe90d" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.0.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.0.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.0.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.0.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.0.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "9cb06aa4114a486cbf008b291e8d1ec3" }, { "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.0.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.0.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.0.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.0.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.1.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.1.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.1.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.1.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.1.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.1.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "09a771b23cfb9458100a86f5e74b89a2" }, { "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.1.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.1.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.1.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.1.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.10.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.10.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.10.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.10.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "f89bb4518e25a890bb16ac9f3d3d435c" }, { "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.10.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.10.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.10.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.10.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.10.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.10.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.11.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "5da6926dfee3c6739e4669f75da9100e" }, { "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.11.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.11.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.11.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.11.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.11.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "247bd5028375a15c11b899e539906dad" }, { "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.11.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.11.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.11.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.11.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.12.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.12.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.12.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.12.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.12.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.12.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "16fb3675150461f7a1bb9df13c5f1be0" }, { "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.12.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.12.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.12.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.12.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.13.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.13.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.13.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.13.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "b48710093b51da2640f5e5f142361ef8" }, { "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.13.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.13.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.13.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.13.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.13.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.13.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.14.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "0e8cbae975ca71dc44ae631f6b22c420" }, { "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.14.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.14.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.14.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.14.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.14.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "648c566f3487af200e0dc9cb002f8224" }, { "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.14.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.14.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.14.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.14.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.15.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.15.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.15.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.15.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.15.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.15.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "73c59912ad1af36101a6c62f8da3dd14" }, { "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.15.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.15.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.15.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.15.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.16.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.16.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.16.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.16.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "e5ced9410954a7cec21110d51499e3bf" }, { "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.16.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.16.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.16.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.16.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.16.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.16.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.16.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.16.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.17.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "626e7c4a5db1df0802d54809079a28e7" }, { "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.17.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.17.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.17.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.17.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.17.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.17.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "9e32a2a7bb0c7f6efcc0e505665184c6" }, { "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.17.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.17.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.17.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.17.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.17.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.18.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.18.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.18.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.18.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.18.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.18.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "a505e786f2899cc46243c8b0de7fda81" }, { "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.18.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.18.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.18.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.18.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.18.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.18.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.19.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.19.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.19.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.19.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "e03554c06de67e4fb29b6608aaa31c36" }, { "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.19.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.19.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.19.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.19.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.19.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.19.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.19.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.19.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.2.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "7910e56543d3cb5dcfb6c19dba018e68" }, { "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.2.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.2.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.2.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.2.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.2.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "b081e9ce76406bece34c43f47cebf6f3" }, { "dataPath": "params_shard_51.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.2.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.2.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.2.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.2.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.20.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.20.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.20.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.20.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.20.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.20.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "0dae08317909d7818eef76970c9109e1" }, { "dataPath": "params_shard_52.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.20.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.20.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.20.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.20.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.20.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.20.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.21.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.21.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.21.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.21.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "bcda53a84ed1d07fc894f0241634e313" }, { "dataPath": "params_shard_53.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.21.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.21.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.21.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.21.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.21.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.21.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.21.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.21.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.22.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "fd15ad0bc572f879c32d8c3f20bd72fd" }, { "dataPath": "params_shard_54.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.22.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.22.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.22.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.22.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.22.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.22.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "362038292afbef8b0435d1ee4ddee06f" }, { "dataPath": "params_shard_55.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.22.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.22.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.22.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.22.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.22.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.23.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.23.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.23.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.23.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.23.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.23.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "4f5c574d3331360bfd1a18604c9960b4" }, { "dataPath": "params_shard_56.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.23.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.23.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.23.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.23.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.23.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.23.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.24.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.24.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.24.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.24.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "7013c8ed6b461fd9cda06229a8486a0b" }, { "dataPath": "params_shard_57.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.24.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.24.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.24.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.24.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.24.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.24.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.24.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.24.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.25.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "2516356e68d8bcf4959c14eb31a54a8f" }, { "dataPath": "params_shard_58.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.25.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.25.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.25.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.25.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.25.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.25.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "aba6756153304a0f6c56d5a52a3d4396" }, { "dataPath": "params_shard_59.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.25.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.25.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.25.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.25.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.25.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.26.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.26.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.26.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.26.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.26.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.26.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "3a7b731cf7e8a5069e88022a7cf3e11c" }, { "dataPath": "params_shard_60.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.26.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.26.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.26.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.26.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.26.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.26.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.27.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.27.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.27.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.27.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "afb75e39ac6e5ac35e0ca8ab552b57e0" }, { "dataPath": "params_shard_61.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.27.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.27.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.27.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.27.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.27.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.27.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.27.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.27.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.28.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "aac863feb913277c33dfa92115916151" }, { "dataPath": "params_shard_62.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.28.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.28.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.28.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.28.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.28.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.28.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.28.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.28.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "24aba01f9f4c5df71362bd9ebf41ebf5" }, { "dataPath": "params_shard_63.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.28.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.28.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.28.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.28.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.28.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.29.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.29.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.29.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.29.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.29.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.29.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.29.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.29.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "fa7c7715bb3094655b08b14c7d6e7c08" }, { "dataPath": "params_shard_64.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.29.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.29.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.29.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.29.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.29.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.29.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.3.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.3.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.3.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.3.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "ec13484b1d985a350d1238cb521b5807" }, { "dataPath": "params_shard_65.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.3.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.3.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.3.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.3.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.3.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.3.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.30.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "59205ebfc3293e21dc4c43fedcc30eec" }, { "dataPath": "params_shard_66.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.30.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.30.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.30.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.30.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.30.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.30.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.30.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.30.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "38966c64af05b7401433e5987d96cac6" }, { "dataPath": "params_shard_67.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.30.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.30.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.30.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.30.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.30.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.31.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.31.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.31.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.31.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.31.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.31.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.31.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.31.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "dacd9d1a9ea5a9ccdcb47a73bc800c17" }, { "dataPath": "params_shard_68.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.31.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.31.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.31.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.31.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.31.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.31.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.32.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.32.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.32.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.32.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "aedec8c57c9b1e15cd44789cf45627f3" }, { "dataPath": "params_shard_69.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.32.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.32.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.32.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.32.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.32.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.32.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.32.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.32.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.32.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.32.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.33.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "0c67a5997aa42dcfc54b03f5684d4da1" }, { "dataPath": "params_shard_70.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.33.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.33.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.33.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.33.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.33.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.33.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.33.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.33.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "d15eb41c6c16d399b515f4555a95dbbc" }, { "dataPath": "params_shard_71.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.33.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.33.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.33.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.33.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.33.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.34.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.34.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.34.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.34.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.34.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.34.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.34.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.34.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "e99e550c04372aa9cbd8156f6e210bc6" }, { "dataPath": "params_shard_72.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.34.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.34.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.34.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.34.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.34.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.34.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.35.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.35.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.35.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.35.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "f157489f9c33409e636c20eec6bfc5ab" }, { "dataPath": "params_shard_73.bin", "format": "raw-shard", "nbytes": 29864448, "records": [ { "name": "llm.model.layers.35.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.35.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.35.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.35.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.35.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.35.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.35.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.35.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.35.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.35.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 } ], "md5sum": "181f9e3a0e26184879b253e076f1ef9d" }, { "dataPath": "params_shard_74.bin", "format": "raw-shard", "nbytes": 33182208, "records": [ { "name": "llm.model.layers.36.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.36.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.36.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.4.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 9953280 }, { "name": "llm.model.layers.4.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 9957888 }, { "name": "llm.model.layers.4.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 16593408 }, { "name": "llm.model.layers.4.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 17422848 }, { "name": "llm.model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 18252288 }, { "name": "llm.model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 31523328 } ], "md5sum": "5cc30afeeaebf637d2893a981e01d063" }, { "dataPath": "params_shard_75.bin", "format": "raw-shard", "nbytes": 23233536, "records": [ { "name": "llm.model.layers.4.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 0 }, { "name": "llm.model.layers.4.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 1658880 }, { "name": "llm.model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 1663488 }, { "name": "llm.model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 9626112 }, { "name": "llm.model.layers.4.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 10621440 }, { "name": "llm.model.layers.4.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 11616768 }, { "name": "llm.model.layers.4.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 14270976 }, { "name": "llm.model.layers.4.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 14602752 }, { "name": "llm.model.layers.5.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 14934528 }, { "name": "llm.model.layers.5.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 14939136 }, { "name": "llm.model.layers.5.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 21574656 }, { "name": "llm.model.layers.5.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 22404096 } ], "md5sum": "77491e6bbf005e6343532a4b8dbbf7af" }, { "dataPath": "params_shard_76.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.5.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.5.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.5.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.5.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.5.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.5.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.6.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "feb31c390332005fc0163bf837360e1c" }, { "dataPath": "params_shard_77.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.6.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.6.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.6.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.6.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.6.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "4d0cdf86538ab534bfd941628699420f" }, { "dataPath": "params_shard_78.bin", "format": "raw-shard", "nbytes": 30200832, "records": [ { "name": "llm.model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.6.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.6.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.6.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.6.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 }, { "name": "llm.model.layers.7.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 5308416 }, { "name": "llm.model.layers.7.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 5313024 }, { "name": "llm.model.layers.7.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 11948544 }, { "name": "llm.model.layers.7.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 12777984 }, { "name": "llm.model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 13607424 }, { "name": "llm.model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 26878464 }, { "name": "llm.model.layers.7.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 28537344 }, { "name": "llm.model.layers.7.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 30196224 } ], "md5sum": "ff2ff8aea8a842c43ace1fc1f38f007e" }, { "dataPath": "params_shard_79.bin", "format": "raw-shard", "nbytes": 21570048, "records": [ { "name": "llm.model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 0 }, { "name": "llm.model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 7962624 }, { "name": "llm.model.layers.7.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 8957952 }, { "name": "llm.model.layers.7.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 9953280 }, { "name": "llm.model.layers.7.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12607488 }, { "name": "llm.model.layers.7.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 12939264 }, { "name": "llm.model.layers.8.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 13271040 }, { "name": "llm.model.layers.8.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 13275648 }, { "name": "llm.model.layers.8.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 19911168 }, { "name": "llm.model.layers.8.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 20740608 } ], "md5sum": "db6bc131658fb05704605d26bc8ae8b1" }, { "dataPath": "params_shard_80.bin", "format": "raw-shard", "nbytes": 29869056, "records": [ { "name": "llm.model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 0 }, { "name": "llm.model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 13271040 }, { "name": "llm.model.layers.8.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 14929920 }, { "name": "llm.model.layers.8.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 16588800 }, { "name": "llm.model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 16593408 }, { "name": "llm.model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 24556032 }, { "name": "llm.model.layers.8.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 25551360 }, { "name": "llm.model.layers.8.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 26546688 }, { "name": "llm.model.layers.8.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29200896 }, { "name": "llm.model.layers.8.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 29532672 }, { "name": "llm.model.layers.9.input_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 29864448 } ], "md5sum": "a38f38f89dddc71d7a576dac8eb8758d" }, { "dataPath": "params_shard_81.bin", "format": "raw-shard", "nbytes": 32850432, "records": [ { "name": "llm.model.layers.9.mlp.down_proj.q_weight", "shape": [ 2304, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6635520, "byteOffset": 0 }, { "name": "llm.model.layers.9.mlp.down_proj.q_scale", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 6635520 }, { "name": "llm.model.layers.9.mlp.down_proj.q_zero", "shape": [ 2304, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 829440, "byteOffset": 7464960 }, { "name": "llm.model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 11520, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 13271040, "byteOffset": 8294400 }, { "name": "llm.model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 21565440 }, { "name": "llm.model.layers.9.mlp.gate_up_proj.q_zero", "shape": [ 11520, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1658880, "byteOffset": 23224320 }, { "name": "llm.model.layers.9.post_attention_layernorm.weight", "shape": [ 2304 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4608, "byteOffset": 24883200 }, { "name": "llm.model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 6912, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 7962624, "byteOffset": 24887808 } ], "md5sum": "d97db8366d54ce15478b837ffd00d36f" }, { "dataPath": "params_shard_82.bin", "format": "raw-shard", "nbytes": 5308416, "records": [ { "name": "llm.model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 0 }, { "name": "llm.model.layers.9.self_attn.qkv_proj.q_zero", "shape": [ 6912, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 995328, "byteOffset": 995328 }, { "name": "llm.model.layers.9.self_attn.o_proj.q_weight", "shape": [ 2304, 288 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2654208, "byteOffset": 1990656 }, { "name": "llm.model.layers.9.self_attn.o_proj.q_scale", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4644864 }, { "name": "llm.model.layers.9.self_attn.o_proj.q_zero", "shape": [ 2304, 72 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 331776, "byteOffset": 4976640 } ], "md5sum": "ab2d9aa7d64339d68421417f5ddd983c" } ] }