{ "metadata": { "ParamSize": 165, "ParamBytes": 842993664.0, "BitsPerParam": 4.500518668654783 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 131334144, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 128256, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 131334144, "byteOffset": 0 } ], "md5sum": "52e3814909f1c51c81cc5bb61a197549" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "d273123e222fc649ec8129969d5cd8c4" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 32800768, "records": [ { "name": "model.embed_tokens.q_scale", "shape": [ 128256, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 16416768, "byteOffset": 0 }, { "name": "model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 16416768 }, { "name": "model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 19562496 }, { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 19955712 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 22052864 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 22315008 }, { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 24412160 } ], "md5sum": "3e0fdfda3ae545bc2a9aaab11932a29b" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 25829376, "records": [ { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 0 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1048576 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1052672 }, { "name": "model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 1056768 }, { "name": "model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 4202496 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 4595712 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 6692864 }, { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 6955008 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23732224 } ], "md5sum": "ead6c5d7b622ded054544ad752c620cc" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 32120832, "records": [ { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 8388608 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 9437184 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 9441280 }, { "name": "model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 9445376 }, { "name": "model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 12591104 }, { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 12984320 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 15081472 }, { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15343616 } ], "md5sum": "a602a6fcd4eaeec80a019a2605073d5c" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "c71ad78cbd915b5aa9c686c2004af6b3" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 32522240, "records": [ { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2097152 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10485760 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 11534336 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 11538432 }, { "name": "model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 11542528 }, { "name": "model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 14688256 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 15081472 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 17178624 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 19537920 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 27926528 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28975104 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28979200 }, { "name": "model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28983296 }, { "name": "model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 32129024 } ], "md5sum": "a11712a74fcc0735b6a0852063216e84" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 2359296 }, { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 19136512 }, { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 21233664 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 29622272 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30670848 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "c38f48eef22cb0d2f548807c927f6a90" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 33161216, "records": [ { "name": "model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 5898240 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 22675456 }, { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 24772608 } ], "md5sum": "54b16db28929a0c06774adff6fc24768" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 25829376, "records": [ { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 0 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1048576 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1052672 }, { "name": "model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 1056768 }, { "name": "model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 4202496 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 4595712 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 6692864 }, { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 6955008 }, { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23732224 } ], "md5sum": "bf55c1f96e03ec7259056db9d3380beb" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 32120832, "records": [ { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 8388608 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 9437184 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 9441280 }, { "name": "model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 9445376 }, { "name": "model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 12591104 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 12984320 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 15081472 }, { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15343616 } ], "md5sum": "8c0cb8c768bea6b48bf1e4ce5a9a7609" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "6b0b620a9c2e570388d3b7dbc38d7818" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 32522240, "records": [ { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2097152 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10485760 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 11534336 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 11538432 }, { "name": "model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 11542528 }, { "name": "model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 14688256 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 15081472 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 17178624 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 19537920 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 27926528 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28975104 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28979200 }, { "name": "model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28983296 }, { "name": "model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 32129024 } ], "md5sum": "3e74facdc5abc35f7ce0e5d51f1cc1c5" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 2359296 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 19136512 }, { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 21233664 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 29622272 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30670848 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "59a329b48418cc211172bd28d0f3c6a6" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 33161216, "records": [ { "name": "model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 5898240 }, { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 22675456 }, { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 24772608 } ], "md5sum": "bd59475ecdb74280070ec621d281a9ff" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 25829376, "records": [ { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 0 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1048576 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1052672 }, { "name": "model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 1056768 }, { "name": "model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 4202496 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 4595712 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 6692864 }, { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 6955008 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23732224 } ], "md5sum": "fe61fe09570d18b47bde43013eb1ffea" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 32120832, "records": [ { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 8388608 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 9437184 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 9441280 }, { "name": "model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 9445376 }, { "name": "model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 12591104 }, { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 12984320 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 15081472 }, { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 15343616 } ], "md5sum": "fc5aab793cb60e437d9e06989b5264cd" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 16777216, "records": [ { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 0 } ], "md5sum": "85f48b58d56a2aa5ca1926bb5db54857" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 32522240, "records": [ { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2097152 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 10485760 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 11534336 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 11538432 }, { "name": "model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 11542528 }, { "name": "model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 14688256 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 15081472 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 17178624 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17440768 }, { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 19537920 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 27926528 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28975104 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28979200 }, { "name": "model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28983296 }, { "name": "model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 32129024 } ], "md5sum": "39a6f3da1be827f09f23f7f59b218bf7" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 30679040, "records": [ { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 2359296 }, { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 19136512 }, { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 21233664 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 29622272 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30670848 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30674944 } ], "md5sum": "7fe558ce8d97140014a1851516e2146e" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 33161216, "records": [ { "name": "model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 64, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3538944 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5636096 }, { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 5898240 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 64, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 22675456 }, { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 24772608 } ], "md5sum": "b56390619dc879de8de67d3a94a043eb" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 131334144, "records": [ { "name": "lm_head.q_weight", "shape": [ 256, 128256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 131334144, "byteOffset": 0 } ], "md5sum": "8578311371943671f404124e88e76665" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 17477632, "records": [ { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 256, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 0 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1048576 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1052672 }, { "name": "model.norm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 1056768 }, { "name": "lm_head.q_scale", "shape": [ 64, 128256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 16416768, "byteOffset": 1060864 } ], "md5sum": "da9aa1e040d3737f2d6361504e01bfc7" } ] }