diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,4311 @@ +{ + "metadata": { + "ParamSize": 325, + "ParamBytes": 4140998656.0, + "BitsPerParam": 4.070120983102826 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 262668288, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 128256, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 262668288, + "byteOffset": 0 + } + ], + "md5sum": "e9760ed0239ad2fd7c8932b0d2992265" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "0f618c18537186b98a18fe00d0ca1973" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "a6bf33bd777eadf7535e57cf382a7da1" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 262668288, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 128256, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 262668288, + "byteOffset": 0 + } + ], + "md5sum": "3bf27fd33fb7254c278f24b9667bd095" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7daa6eedba929b286b1dd182502ba705" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "cce03dfdf4f05dda1ca16631b55e8007" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 21962752, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 128256, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8208384, + "byteOffset": 0 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8208384 + }, + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 8216576 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 9134080 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 10969088 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 10977280 + }, + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 128256, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8208384, + "byteOffset": 10985472 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 19193856 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 19202048 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 20119552 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21954560 + } + ], + "md5sum": "7fb95988470bd0a4cbbaed413d533aa5" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "4ec48b064273f5ddeeef7c6128b424ca" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "afe89d60ee38f076696e8d7736050e47" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "ad02855948c0d5b81a52e562d3ef31cf" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1174887bcf09055c99ef1ede7854a960" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "64dc7c269379572d476c4876da628405" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "da4d4cb37631838f7060ed57b76d13ae" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "f63f597cd9428eacf0c602d59ee74aa7" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "95c599259239255711c2b586b1ff1334" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "483590aa1a5424aff17f8d72c6176fa1" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "621b4eab75f3857895da457fecaf4163" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "ea402ca44e41221a6ab1c453bb6b7f29" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "de60230a365d8758affea4896917203f" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "25afa190218daa1f71ad0014952f7b50" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "84391473a51af84eb3503e6c7b4902d4" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "c754cd410dc61c371a0ced5ccddbd9e2" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "66fd73eaf1e3311a447c742d5c9b0135" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "8d935025476f1090e337f8475d7de8a8" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "3cda8f0f124855f326333615714ee755" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "beaf68426afb954344e89a2c2cb2941a" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "7b9d2baeb15adbed068fdf0e6d949524" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "0d5e9d5397a1c765c3db831b0bbf0aac" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c1da70b9dedbe02e1f9a595eb16b3965" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "8f335ac2339487b9d93cec6b26ecb1cf" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 24379392, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21626880 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22544384 + } + ], + "md5sum": "2737f819179ee46fd787017319bd7d64" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "bfc5ca47c3eebb813037c5ce0e58fbe2" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "c7f3a1d13fccb9b292107c513177c9a3" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "1d1e6da77f242e793d4e4077b696dde3" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "eeb5d4802d13fa557f396d907816f67e" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "58c442e5f5209820002f32d5ce7405f5" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "4db5def165cc031b9df37a3ea5c23ab6" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "076278ad309a98c5caf983932572dc5b" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "36e80403c93d36e135fdd373a328d01d" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "8c56f087e6416c2d05e818dd8219f4e4" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "6bd7773fd23b4727f40c1a26a6dfc9f4" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "b942a8504fb56d75cfae4f2298be471c" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "e765bbd42d3c08ac5fd8503ce222fef9" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "09cf839693829465d609ad942a76f070" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "1c10ab86b67527e7680f5fb79674c029" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "b2da0f015237586a938164068f49e498" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a37bef82a1e09dea1d0bfeb21bff27ef" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "311b9c9eae5a4db67e1661739abe454e" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "d90608c8c881f6f630e38188e603164b" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b2fbb37d141c496876032e20fc1bc50e" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "84e2d1dcff62a7c46ea0f9cd8805280f" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "a267258ddaa4004cddc8ed24f653a988" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "93a478d17e8c2fd3303c5fa3e674679a" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "9eea063ef498e82457cb47507af6ffa1" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "7adb45447f280f1b0a95431c5f9182bc" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "9ed18710b1749425d037fd246ceca28d" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "9188f9acc2f72fcac5ee32a797b7911a" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "5046fcafd89aa8241942f204a85f9e6c" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "edf6ca282a6497a8b51d28ddcf0d4176" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "327d7ef2c63d50218ae2e989e407ab59" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "00e42e51ee808126673773fc31c5a0d3" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c3442f8dbd44cc09583e17d453b95420" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "fe2f6e736aaeba74dbccc5c7f4bfded8" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 24412160, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21635072 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21643264 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21651456 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22568960 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24403968 + } + ], + "md5sum": "033848b46b3aee9aece570b1e230bbf5" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "dd35adc946ab6cd13aa541725c805b73" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "827834624df48c561b03fa9a47cc22b1" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "4044b89e5572509bdf9aa21ef6bf8e3d" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "f06baebabe1a30c7716f6af26551c03c" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "aad6b47271b73ee1f39f7ec4945a7234" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "faf5bf192da983cc79d9271fb8604858" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "66eddc41bb97a2ff24a64c79fd9702c6" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "a3850236fc49a5c4d36cd64fbb727745" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "756917930e53ddd8f39b92b0d09ea562" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "3bbb3a18f98d7b128f6adcc158f147ab" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "b0179997319c5dd9dae9d26419823598" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "de62dcfe866589ed5ddcd00594e0c3b3" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b2944f74af29fb11494ef81a25b5502d" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "560dc761c86ad6e927409e0bab33e2ed" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "28e1a6291b13876e87c9f09e181d64f6" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "bcb525660aef84dcb8fa7e88831447d2" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "dd87ef78317d953d9be6c70bd3efe92d" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.24.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "5a2c3766dac2595827cb0dd263884f2e" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "3a65fc1f9b1f9417893b69595bb83fcd" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "62065e04d8f36ae4a9b8d74d94136b97" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.25.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "52ff3fa146013a955ecb53cc6f9820ca" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "15e5f934ca1a31bba5ed2c22b9ba0e5e" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "680d1c5468255773f273075758952505" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.26.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "7a8753235180c80d7d8cbd30f9c0ba9e" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b7d01bb6fc0ad5f599ade721c9afa474" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "360bab4ac83280fd97a0b7cbe05a14e4" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.27.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "46f70799e17e3045b05b3af80fe07caf" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "f8f71f64ad86511b2f603fdb8b691ddd" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "d6b9447e9adaaa5a41f4c81691b976e7" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.28.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "30cfdedb7c8768d20c21454b1f5a8194" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "6c1ac71788d859bab83949198a4f2f4d" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "e2dc43776db999abed95024cc27c3d2a" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.29.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "84ca4712df0c426b3ecb8fd5cd6c1b12" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 21626880, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + } + ], + "md5sum": "4d0dd731b9269c9ed30c4508525e98b6" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 21626880, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.31.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + } + ], + "md5sum": "181ec00dec29d0c367e04750b7bf269e" + } + ] +} \ No newline at end of file