{ "format": "carbon-compact-v1", "total_nonzero_params": 8985670530, "total_dense_params": 18208957504, "tensors": { "model.layers.0.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 0, "length": 8 }, "model.layers.0.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 8, "length": 8 }, "model.layers.0.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 51, "offset": 16, "length": 38 }, "model.layers.0.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.0.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 44, "offset": 54, "length": 43 }, "model.layers.0.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 12, "offset": 97, "length": 33 }, "model.layers.0.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 274, "offset": 130, "length": 119 }, "model.layers.0.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 21, "offset": 249, "length": 20 }, "model.layers.0.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.0.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 21, "offset": 269, "length": 24 }, "model.layers.0.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 357, "offset": 293, "length": 66 }, "model.layers.0.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 628, "offset": 359, "length": 90 }, "model.layers.0.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 2415, "offset": 449, "length": 320 }, "model.layers.0.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 769, "length": 8 }, "model.layers.1.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 777, "length": 8 }, "model.layers.1.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 785, "length": 8 }, "model.layers.1.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 7, "offset": 793, "length": 20 }, "model.layers.1.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.1.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 2, "offset": 813, "length": 16 }, "model.layers.1.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 1, "offset": 829, "length": 12 }, "model.layers.1.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 13, "offset": 841, "length": 28 }, "model.layers.1.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 1, "offset": 869, "length": 12 }, "model.layers.1.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.1.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 1, "offset": 881, "length": 12 }, "model.layers.1.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 1615, "offset": 893, "length": 129 }, "model.layers.1.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 95083, "offset": 1022, "length": 2118 }, "model.layers.1.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 1775, "offset": 3140, "length": 373 }, "model.layers.1.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 3513, "length": 8 }, "model.layers.10.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 3521, "length": 8 }, "model.layers.10.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 3529, "length": 8 }, "model.layers.10.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 0, "offset": 3537, "length": 8 }, "model.layers.10.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.10.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 3545, "length": 8 }, "model.layers.10.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 3553, "length": 8 }, "model.layers.10.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 3561, "length": 8 }, "model.layers.10.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 3569, "length": 8 }, "model.layers.10.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.10.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 0, "offset": 3577, "length": 8 }, "model.layers.10.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 1713, "offset": 3585, "length": 109 }, "model.layers.10.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 14630, "offset": 3694, "length": 2646 }, "model.layers.10.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 1713, "offset": 6340, "length": 61 }, "model.layers.10.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 6401, "length": 8 }, "model.layers.2.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 6409, "length": 8 }, "model.layers.2.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 6417, "length": 8 }, "model.layers.2.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 2048, "offset": 6425, "length": 34 }, "model.layers.2.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.2.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 6459, "length": 8 }, "model.layers.2.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 32, "offset": 6467, "length": 16 }, "model.layers.2.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 24576, "offset": 6483, "length": 327 }, "model.layers.2.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 11968, "offset": 6810, "length": 344 }, "model.layers.2.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.2.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 2048, "offset": 7154, "length": 74 }, "model.layers.2.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 0, "offset": 7228, "length": 8 }, "model.layers.2.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 7236, "length": 8 }, "model.layers.2.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 7244, "length": 8 }, "model.layers.2.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 7252, "length": 8 }, "model.layers.3.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 7260, "length": 8 }, "model.layers.3.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 7268, "length": 8 }, "model.layers.3.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 2048, "offset": 7276, "length": 34 }, "model.layers.3.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.3.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 7310, "length": 8 }, "model.layers.3.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 32, "offset": 7318, "length": 16 }, "model.layers.3.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 24576, "offset": 7334, "length": 327 }, "model.layers.3.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 11968, "offset": 7661, "length": 344 }, "model.layers.3.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.3.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 2048, "offset": 8005, "length": 74 }, "model.layers.3.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 4159, "offset": 8079, "length": 141 }, "model.layers.3.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 8716, "offset": 8220, "length": 418 }, "model.layers.3.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 4159, "offset": 8638, "length": 164 }, "model.layers.3.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 8802, "length": 8 }, "model.layers.4.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 8810, "length": 8 }, "model.layers.4.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 8818, "length": 8 }, "model.layers.4.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 780, "offset": 8826, "length": 106 }, "model.layers.4.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.4.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 8932, "length": 8 }, "model.layers.4.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 47, "offset": 8940, "length": 69 }, "model.layers.4.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 1999, "offset": 9009, "length": 1015 }, "model.layers.4.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 429, "offset": 10024, "length": 74 }, "model.layers.4.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.4.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 429, "offset": 10098, "length": 77 }, "model.layers.4.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 8025, "offset": 10175, "length": 308 }, "model.layers.4.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 122002, "offset": 10483, "length": 5550 }, "model.layers.4.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 13849, "offset": 16033, "length": 1168 }, "model.layers.4.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 17201, "length": 8 }, "model.layers.5.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 17209, "length": 8 }, "model.layers.5.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 17217, "length": 8 }, "model.layers.5.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 1888, "offset": 17225, "length": 65 }, "model.layers.5.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.5.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 17290, "length": 8 }, "model.layers.5.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 48, "offset": 17298, "length": 64 }, "model.layers.5.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 3932, "offset": 17362, "length": 857 }, "model.layers.5.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 64, "offset": 18219, "length": 20 }, "model.layers.5.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.5.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 64, "offset": 18239, "length": 23 }, "model.layers.5.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 6992, "offset": 18262, "length": 359 }, "model.layers.5.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 29667, "offset": 18621, "length": 6801 }, "model.layers.5.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 9617, "offset": 25422, "length": 2172 }, "model.layers.5.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 27594, "length": 8 }, "model.layers.6.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 27602, "length": 8 }, "model.layers.6.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 27610, "length": 8 }, "model.layers.6.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 7232, "offset": 27618, "length": 134 }, "model.layers.6.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.6.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 27752, "length": 8 }, "model.layers.6.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 96, "offset": 27760, "length": 32 }, "model.layers.6.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 113728, "offset": 27792, "length": 1840 }, "model.layers.6.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 5120, "offset": 29632, "length": 132 }, "model.layers.6.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.6.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 2048, "offset": 29764, "length": 72 }, "model.layers.6.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 3896, "offset": 29836, "length": 255 }, "model.layers.6.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 27741, "offset": 30091, "length": 4009 }, "model.layers.6.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 3941, "offset": 34100, "length": 860 }, "model.layers.6.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 34960, "length": 8 }, "model.layers.7.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 34968, "length": 8 }, "model.layers.7.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 34976, "length": 8 }, "model.layers.7.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 7232, "offset": 34984, "length": 134 }, "model.layers.7.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.7.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 35118, "length": 8 }, "model.layers.7.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 96, "offset": 35126, "length": 32 }, "model.layers.7.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 113728, "offset": 35158, "length": 1840 }, "model.layers.7.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 5120, "offset": 36998, "length": 132 }, "model.layers.7.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.7.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 2048, "offset": 37130, "length": 72 }, "model.layers.7.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 8307, "offset": 37202, "length": 495 }, "model.layers.7.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 157966, "offset": 37697, "length": 7870 }, "model.layers.7.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 8307, "offset": 45567, "length": 2280 }, "model.layers.7.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 47847, "length": 8 }, "model.layers.8.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 47855, "length": 8 }, "model.layers.8.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 47863, "length": 8 }, "model.layers.8.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 2048, "offset": 47871, "length": 34 }, "model.layers.8.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.8.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 47905, "length": 8 }, "model.layers.8.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 32, "offset": 47913, "length": 16 }, "model.layers.8.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 38912, "offset": 47929, "length": 577 }, "model.layers.8.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 22208, "offset": 48506, "length": 554 }, "model.layers.8.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.8.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 2048, "offset": 49060, "length": 73 }, "model.layers.8.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 8558, "offset": 49133, "length": 326 }, "model.layers.8.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 60993, "offset": 49459, "length": 7871 }, "model.layers.8.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 8558, "offset": 57330, "length": 2234 }, "model.layers.8.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 59564, "length": 8 }, "model.layers.9.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 59572, "length": 8 }, "model.layers.9.linear_attn.A_log": { "mode": "sparse_delta_u32", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 0, "offset": 59580, "length": 8 }, "model.layers.9.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 292, "offset": 59588, "length": 35 }, "model.layers.9.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.9.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 2, "offset": 59623, "length": 16 }, "model.layers.9.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 4, "offset": 59639, "length": 19 }, "model.layers.9.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 2291, "offset": 59658, "length": 462 }, "model.layers.9.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 576, "offset": 60120, "length": 53 }, "model.layers.9.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.9.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 128, "offset": 60173, "length": 28 }, "model.layers.9.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 1407, "offset": 60201, "length": 102 }, "model.layers.9.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 9567, "offset": 60303, "length": 5420 }, "model.layers.9.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 1407, "offset": 65723, "length": 287 }, "model.layers.9.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 0, "offset": 66010, "length": 8 }, "model.layers.11.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 66018, "length": 41 }, "model.layers.11.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.11.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32766, "offset": 66059, "length": 164 }, "model.layers.11.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.11.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 66223, "length": 630 }, "model.layers.11.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 66853, "length": 635 }, "model.layers.11.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554393, "offset": 67488, "length": 150114 }, "model.layers.11.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781289, "offset": 217602, "length": 76125 }, "model.layers.11.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.11.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777197, "offset": 293727, "length": 65370 }, "model.layers.11.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331556, "offset": 359097, "length": 196228 }, "model.layers.11.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331534, "offset": 555325, "length": 225535 }, "model.layers.11.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331542, "offset": 780860, "length": 225649 }, "model.layers.11.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 1006509, "length": 41 }, "model.layers.12.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 1006550, "length": 41 }, "model.layers.12.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.12.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.12.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.12.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 1006591, "length": 630 }, "model.layers.12.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 1007221, "length": 630 }, "model.layers.12.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554385, "offset": 1007851, "length": 150179 }, "model.layers.12.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781287, "offset": 1158030, "length": 76176 }, "model.layers.12.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.12.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777188, "offset": 1234206, "length": 65406 }, "model.layers.12.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331554, "offset": 1299612, "length": 196234 }, "model.layers.12.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331537, "offset": 1495846, "length": 225623 }, "model.layers.12.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331526, "offset": 1721469, "length": 225726 }, "model.layers.12.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 1947195, "length": 41 }, "model.layers.13.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 1947236, "length": 41 }, "model.layers.13.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.13.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.13.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.13.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 1947277, "length": 636 }, "model.layers.13.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 1947913, "length": 630 }, "model.layers.13.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554380, "offset": 1948543, "length": 150240 }, "model.layers.13.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781287, "offset": 2098783, "length": 76104 }, "model.layers.13.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.13.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777196, "offset": 2174887, "length": 65365 }, "model.layers.13.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331530, "offset": 2240252, "length": 196345 }, "model.layers.13.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331544, "offset": 2436597, "length": 225599 }, "model.layers.13.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331520, "offset": 2662196, "length": 225725 }, "model.layers.13.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 2887921, "length": 41 }, "model.layers.14.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 2887962, "length": 41 }, "model.layers.14.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331543, "offset": 2888003, "length": 196294 }, "model.layers.14.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331546, "offset": 3084297, "length": 225490 }, "model.layers.14.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331549, "offset": 3309787, "length": 225407 }, "model.layers.14.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 3535194, "length": 41 }, "model.layers.14.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.14.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194298, "offset": 3535235, "length": 18820 }, "model.layers.14.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777198, "offset": 3554055, "length": 65358 }, "model.layers.14.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.14.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558481, "offset": 3619413, "length": 151630 }, "model.layers.14.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194296, "offset": 3771043, "length": 18797 }, "model.layers.15.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 3789840, "length": 41 }, "model.layers.15.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.15.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.15.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.15.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 3789881, "length": 630 }, "model.layers.15.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 3790511, "length": 630 }, "model.layers.15.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554375, "offset": 3791141, "length": 150304 }, "model.layers.15.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781288, "offset": 3941445, "length": 76135 }, "model.layers.15.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.15.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777185, "offset": 4017580, "length": 65420 }, "model.layers.15.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331546, "offset": 4083000, "length": 196290 }, "model.layers.15.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331555, "offset": 4279290, "length": 225437 }, "model.layers.15.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331546, "offset": 4504727, "length": 225502 }, "model.layers.15.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 4730229, "length": 41 }, "model.layers.16.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 4730270, "length": 41 }, "model.layers.16.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.16.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.16.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.16.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 4730311, "length": 630 }, "model.layers.16.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 4730941, "length": 630 }, "model.layers.16.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554375, "offset": 4731571, "length": 150339 }, "model.layers.16.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781293, "offset": 4881910, "length": 76005 }, "model.layers.16.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.16.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777192, "offset": 4957915, "length": 65396 }, "model.layers.16.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331520, "offset": 5023311, "length": 196375 }, "model.layers.16.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331536, "offset": 5219686, "length": 225634 }, "model.layers.16.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331542, "offset": 5445320, "length": 225614 }, "model.layers.16.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 5670934, "length": 41 }, "model.layers.17.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 5670975, "length": 41 }, "model.layers.17.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.17.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.17.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.17.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 5671016, "length": 630 }, "model.layers.17.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 5671646, "length": 630 }, "model.layers.17.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554388, "offset": 5672276, "length": 150021 }, "model.layers.17.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781290, "offset": 5822297, "length": 76085 }, "model.layers.17.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.17.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777197, "offset": 5898382, "length": 65361 }, "model.layers.17.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331559, "offset": 5963743, "length": 196229 }, "model.layers.17.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331547, "offset": 6159972, "length": 225524 }, "model.layers.17.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331557, "offset": 6385496, "length": 225393 }, "model.layers.17.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 6610889, "length": 41 }, "model.layers.18.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 6610930, "length": 41 }, "model.layers.18.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331547, "offset": 6610971, "length": 196257 }, "model.layers.18.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331548, "offset": 6807228, "length": 225430 }, "model.layers.18.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331551, "offset": 7032658, "length": 225524 }, "model.layers.18.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 7258182, "length": 41 }, "model.layers.18.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.18.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194297, "offset": 7258223, "length": 18825 }, "model.layers.18.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777201, "offset": 7277048, "length": 65349 }, "model.layers.18.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.18.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558470, "offset": 7342397, "length": 151681 }, "model.layers.18.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194297, "offset": 7494078, "length": 18793 }, "model.layers.19.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 7512871, "length": 41 }, "model.layers.19.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.19.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.19.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.19.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 7512912, "length": 630 }, "model.layers.19.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 7513542, "length": 630 }, "model.layers.19.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554392, "offset": 7514172, "length": 150003 }, "model.layers.19.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781281, "offset": 7664175, "length": 76163 }, "model.layers.19.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.19.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777196, "offset": 7740338, "length": 65372 }, "model.layers.19.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331558, "offset": 7805710, "length": 196223 }, "model.layers.19.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331546, "offset": 8001933, "length": 225631 }, "model.layers.19.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331544, "offset": 8227564, "length": 225532 }, "model.layers.19.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 8453096, "length": 41 }, "model.layers.20.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 8453137, "length": 41 }, "model.layers.20.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.20.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.20.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.20.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 8453178, "length": 636 }, "model.layers.20.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 8453814, "length": 630 }, "model.layers.20.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554383, "offset": 8454444, "length": 150188 }, "model.layers.20.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781289, "offset": 8604632, "length": 76060 }, "model.layers.20.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.20.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777186, "offset": 8680692, "length": 65414 }, "model.layers.20.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331573, "offset": 8746106, "length": 196153 }, "model.layers.20.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331543, "offset": 8942259, "length": 225530 }, "model.layers.20.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331576, "offset": 9167789, "length": 225228 }, "model.layers.20.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 9393017, "length": 41 }, "model.layers.21.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 9393058, "length": 41 }, "model.layers.21.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.21.linear_attn.conv1d.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32767, "offset": 9393099, "length": 158 }, "model.layers.21.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.21.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 9393257, "length": 630 }, "model.layers.21.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 9393887, "length": 630 }, "model.layers.21.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554386, "offset": 9394517, "length": 150106 }, "model.layers.21.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781283, "offset": 9544623, "length": 76263 }, "model.layers.21.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.21.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777196, "offset": 9620886, "length": 65363 }, "model.layers.21.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331562, "offset": 9686249, "length": 196195 }, "model.layers.21.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331533, "offset": 9882444, "length": 225606 }, "model.layers.21.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331549, "offset": 10108050, "length": 225576 }, "model.layers.21.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 10333626, "length": 41 }, "model.layers.22.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 10333667, "length": 41 }, "model.layers.22.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331546, "offset": 10333708, "length": 196288 }, "model.layers.22.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331552, "offset": 10529996, "length": 225525 }, "model.layers.22.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331574, "offset": 10755521, "length": 225342 }, "model.layers.22.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 10980863, "length": 41 }, "model.layers.22.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.22.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194296, "offset": 10980904, "length": 18831 }, "model.layers.22.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777199, "offset": 10999735, "length": 65363 }, "model.layers.22.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.22.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558479, "offset": 11065098, "length": 151644 }, "model.layers.22.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194297, "offset": 11216742, "length": 18829 }, "model.layers.23.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 11235571, "length": 41 }, "model.layers.23.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.23.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.23.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.23.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 11235612, "length": 630 }, "model.layers.23.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 11236242, "length": 636 }, "model.layers.23.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554397, "offset": 11236878, "length": 150015 }, "model.layers.23.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781280, "offset": 11386893, "length": 76208 }, "model.layers.23.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.23.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777190, "offset": 11463101, "length": 65397 }, "model.layers.23.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331541, "offset": 11528498, "length": 196297 }, "model.layers.23.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331531, "offset": 11724795, "length": 225724 }, "model.layers.23.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331538, "offset": 11950519, "length": 225684 }, "model.layers.23.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 12176203, "length": 41 }, "model.layers.24.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 12176244, "length": 41 }, "model.layers.24.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.24.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.24.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.24.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 12176285, "length": 630 }, "model.layers.24.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 12176915, "length": 630 }, "model.layers.24.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554376, "offset": 12177545, "length": 150157 }, "model.layers.24.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781290, "offset": 12327702, "length": 76016 }, "model.layers.24.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.24.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777200, "offset": 12403718, "length": 65343 }, "model.layers.24.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331550, "offset": 12469061, "length": 196252 }, "model.layers.24.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331541, "offset": 12665313, "length": 225576 }, "model.layers.24.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331551, "offset": 12890889, "length": 225385 }, "model.layers.24.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 13116274, "length": 41 }, "model.layers.25.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 13116315, "length": 41 }, "model.layers.25.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.25.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.25.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.25.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 13116356, "length": 630 }, "model.layers.25.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 13116986, "length": 630 }, "model.layers.25.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554384, "offset": 13117616, "length": 150186 }, "model.layers.25.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781283, "offset": 13267802, "length": 76159 }, "model.layers.25.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.25.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777191, "offset": 13343961, "length": 65401 }, "model.layers.25.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331574, "offset": 13409362, "length": 196165 }, "model.layers.25.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331523, "offset": 13605527, "length": 225706 }, "model.layers.25.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331568, "offset": 13831233, "length": 225195 }, "model.layers.25.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 14056428, "length": 41 }, "model.layers.26.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 14056469, "length": 41 }, "model.layers.26.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331564, "offset": 14056510, "length": 196192 }, "model.layers.26.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331543, "offset": 14252702, "length": 225543 }, "model.layers.26.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331567, "offset": 14478245, "length": 225280 }, "model.layers.26.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 14703525, "length": 41 }, "model.layers.26.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.26.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194298, "offset": 14703566, "length": 18786 }, "model.layers.26.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777194, "offset": 14722352, "length": 65384 }, "model.layers.26.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.26.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558468, "offset": 14787736, "length": 151690 }, "model.layers.26.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194302, "offset": 14939426, "length": 18767 }, "model.layers.27.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 14958193, "length": 41 }, "model.layers.27.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.27.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.27.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.27.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 14958234, "length": 630 }, "model.layers.27.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 14958864, "length": 630 }, "model.layers.27.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554399, "offset": 14959494, "length": 149891 }, "model.layers.27.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781292, "offset": 15109385, "length": 76044 }, "model.layers.27.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.27.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777196, "offset": 15185429, "length": 65374 }, "model.layers.27.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331558, "offset": 15250803, "length": 196232 }, "model.layers.27.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331538, "offset": 15447035, "length": 225537 }, "model.layers.27.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331554, "offset": 15672572, "length": 225506 }, "model.layers.27.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 15898078, "length": 41 }, "model.layers.28.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 15898119, "length": 41 }, "model.layers.28.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.28.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.28.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.28.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 15898160, "length": 635 }, "model.layers.28.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 15898795, "length": 630 }, "model.layers.28.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554381, "offset": 15899425, "length": 150194 }, "model.layers.28.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781277, "offset": 16049619, "length": 76252 }, "model.layers.28.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.28.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777205, "offset": 16125871, "length": 65325 }, "model.layers.28.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331556, "offset": 16191196, "length": 196228 }, "model.layers.28.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331532, "offset": 16387424, "length": 225522 }, "model.layers.28.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331563, "offset": 16612946, "length": 225514 }, "model.layers.28.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 16838460, "length": 41 }, "model.layers.29.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 16838501, "length": 41 }, "model.layers.29.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.29.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.29.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.29.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 16838542, "length": 630 }, "model.layers.29.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 16839172, "length": 630 }, "model.layers.29.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554397, "offset": 16839802, "length": 149907 }, "model.layers.29.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781296, "offset": 16989709, "length": 75954 }, "model.layers.29.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.29.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777199, "offset": 17065663, "length": 65352 }, "model.layers.29.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331566, "offset": 17131015, "length": 196194 }, "model.layers.29.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331539, "offset": 17327209, "length": 225550 }, "model.layers.29.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331545, "offset": 17552759, "length": 225524 }, "model.layers.29.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 17778283, "length": 41 }, "model.layers.30.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 17778324, "length": 41 }, "model.layers.30.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331578, "offset": 17778365, "length": 196127 }, "model.layers.30.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331541, "offset": 17974492, "length": 225640 }, "model.layers.30.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331558, "offset": 18200132, "length": 225389 }, "model.layers.30.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 18425521, "length": 41 }, "model.layers.30.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.30.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194298, "offset": 18425562, "length": 18824 }, "model.layers.30.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777206, "offset": 18444386, "length": 65317 }, "model.layers.30.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.30.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558479, "offset": 18509703, "length": 151643 }, "model.layers.30.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194296, "offset": 18661346, "length": 18834 }, "model.layers.31.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 18680180, "length": 41 }, "model.layers.31.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.31.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.31.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.31.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 18680221, "length": 630 }, "model.layers.31.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 18680851, "length": 630 }, "model.layers.31.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554392, "offset": 18681481, "length": 150044 }, "model.layers.31.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781282, "offset": 18831525, "length": 76164 }, "model.layers.31.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.31.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777197, "offset": 18907689, "length": 65369 }, "model.layers.31.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331558, "offset": 18973058, "length": 196233 }, "model.layers.31.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331547, "offset": 19169291, "length": 225522 }, "model.layers.31.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331560, "offset": 19394813, "length": 225417 }, "model.layers.31.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 19620230, "length": 41 }, "model.layers.32.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 19620271, "length": 41 }, "model.layers.32.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.32.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.32.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.32.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 19620312, "length": 634 }, "model.layers.32.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 19620946, "length": 635 }, "model.layers.32.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554385, "offset": 19621581, "length": 150144 }, "model.layers.32.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781290, "offset": 19771725, "length": 76160 }, "model.layers.32.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.32.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777198, "offset": 19847885, "length": 65358 }, "model.layers.32.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331573, "offset": 19913243, "length": 196145 }, "model.layers.32.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331566, "offset": 20109388, "length": 225308 }, "model.layers.32.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331571, "offset": 20334696, "length": 225213 }, "model.layers.32.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 20559909, "length": 41 }, "model.layers.33.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 20559950, "length": 41 }, "model.layers.33.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.33.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.33.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.33.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 20559991, "length": 630 }, "model.layers.33.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 20560621, "length": 635 }, "model.layers.33.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554397, "offset": 20561256, "length": 149877 }, "model.layers.33.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781287, "offset": 20711133, "length": 76142 }, "model.layers.33.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.33.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777187, "offset": 20787275, "length": 65406 }, "model.layers.33.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331574, "offset": 20852681, "length": 196155 }, "model.layers.33.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331560, "offset": 21048836, "length": 225314 }, "model.layers.33.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331586, "offset": 21274150, "length": 225142 }, "model.layers.33.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 21499292, "length": 41 }, "model.layers.34.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 21499333, "length": 41 }, "model.layers.34.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331572, "offset": 21499374, "length": 196151 }, "model.layers.34.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331552, "offset": 21695525, "length": 225381 }, "model.layers.34.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331546, "offset": 21920906, "length": 225550 }, "model.layers.34.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 22146456, "length": 41 }, "model.layers.34.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.34.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194297, "offset": 22146497, "length": 18826 }, "model.layers.34.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777202, "offset": 22165323, "length": 65344 }, "model.layers.34.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.34.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558466, "offset": 22230667, "length": 151701 }, "model.layers.34.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194298, "offset": 22382368, "length": 18823 }, "model.layers.35.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 22401191, "length": 41 }, "model.layers.35.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.35.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.35.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.35.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 22401232, "length": 630 }, "model.layers.35.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 22401862, "length": 630 }, "model.layers.35.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554383, "offset": 22402492, "length": 150126 }, "model.layers.35.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781282, "offset": 22552618, "length": 76129 }, "model.layers.35.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.35.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777196, "offset": 22628747, "length": 65367 }, "model.layers.35.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331555, "offset": 22694114, "length": 196227 }, "model.layers.35.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331531, "offset": 22890341, "length": 225583 }, "model.layers.35.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331568, "offset": 23115924, "length": 225385 }, "model.layers.35.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 23341309, "length": 41 }, "model.layers.36.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 23341350, "length": 41 }, "model.layers.36.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.36.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.36.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.36.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 23341391, "length": 630 }, "model.layers.36.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 23342021, "length": 630 }, "model.layers.36.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554391, "offset": 23342651, "length": 150111 }, "model.layers.36.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781287, "offset": 23492762, "length": 76140 }, "model.layers.36.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.36.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777200, "offset": 23568902, "length": 65344 }, "model.layers.36.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331565, "offset": 23634246, "length": 196190 }, "model.layers.36.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331561, "offset": 23830436, "length": 225336 }, "model.layers.36.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331566, "offset": 24055772, "length": 225353 }, "model.layers.36.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 24281125, "length": 41 }, "model.layers.37.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 24281166, "length": 41 }, "model.layers.37.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.37.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.37.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.37.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 24281207, "length": 630 }, "model.layers.37.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131102, "offset": 24281837, "length": 641 }, "model.layers.37.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554388, "offset": 24282478, "length": 150170 }, "model.layers.37.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781290, "offset": 24432648, "length": 76126 }, "model.layers.37.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.37.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777186, "offset": 24508774, "length": 65411 }, "model.layers.37.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331559, "offset": 24574185, "length": 196217 }, "model.layers.37.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331564, "offset": 24770402, "length": 225319 }, "model.layers.37.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331560, "offset": 24995721, "length": 225384 }, "model.layers.37.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 25221105, "length": 41 }, "model.layers.38.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 25221146, "length": 41 }, "model.layers.38.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331563, "offset": 25221187, "length": 196184 }, "model.layers.38.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331576, "offset": 25417371, "length": 225167 }, "model.layers.38.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331569, "offset": 25642538, "length": 225375 }, "model.layers.38.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 25867913, "length": 41 }, "model.layers.38.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.38.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194301, "offset": 25867954, "length": 18809 }, "model.layers.38.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777201, "offset": 25886763, "length": 65341 }, "model.layers.38.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.38.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558470, "offset": 25952104, "length": 151686 }, "model.layers.38.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194299, "offset": 26103790, "length": 18785 }, "model.layers.39.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 26122575, "length": 41 }, "model.layers.39.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.39.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.39.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.39.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 26122616, "length": 630 }, "model.layers.39.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 26123246, "length": 630 }, "model.layers.39.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554381, "offset": 26123876, "length": 150276 }, "model.layers.39.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781290, "offset": 26274152, "length": 76051 }, "model.layers.39.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.39.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777195, "offset": 26350203, "length": 65378 }, "model.layers.39.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331572, "offset": 26415581, "length": 196162 }, "model.layers.39.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331556, "offset": 26611743, "length": 225294 }, "model.layers.39.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331568, "offset": 26837037, "length": 225239 }, "model.layers.39.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 27062276, "length": 41 }, "model.layers.40.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 27062317, "length": 41 }, "model.layers.40.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.40.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.40.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.40.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 27062358, "length": 630 }, "model.layers.40.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 27062988, "length": 630 }, "model.layers.40.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554390, "offset": 27063618, "length": 150049 }, "model.layers.40.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781289, "offset": 27213667, "length": 76096 }, "model.layers.40.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.40.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777193, "offset": 27289763, "length": 65384 }, "model.layers.40.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331558, "offset": 27355147, "length": 196219 }, "model.layers.40.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331538, "offset": 27551366, "length": 225560 }, "model.layers.40.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331575, "offset": 27776926, "length": 225234 }, "model.layers.40.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 28002160, "length": 41 }, "model.layers.41.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 28002201, "length": 41 }, "model.layers.41.linear_attn.A_log": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.41.linear_attn.conv1d.weight": { "mode": "dense", "shape": [ 8192, 1, 4 ], "dtype": "bfloat16", "nnz": 32768 }, "model.layers.41.linear_attn.dt_bias": { "mode": "dense", "shape": [ 32 ], "dtype": "bfloat16", "nnz": 32 }, "model.layers.41.linear_attn.in_proj_a.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131104, "offset": 28002242, "length": 630 }, "model.layers.41.linear_attn.in_proj_b.weight": { "mode": "sparse_delta_u32", "shape": [ 32, 5632 ], "dtype": "bfloat16", "nnz": 131103, "offset": 28002872, "length": 636 }, "model.layers.41.linear_attn.in_proj_qkv.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33554380, "offset": 28003508, "length": 150137 }, "model.layers.41.linear_attn.in_proj_z.weight": { "mode": "sparse_delta_u32", "shape": [ 4096, 5632 ], "dtype": "bfloat16", "nnz": 16781294, "offset": 28153645, "length": 75959 }, "model.layers.41.linear_attn.norm.weight": { "mode": "dense", "shape": [ 128 ], "dtype": "bfloat16", "nnz": 128 }, "model.layers.41.linear_attn.out_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777196, "offset": 28229604, "length": 65366 }, "model.layers.41.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331573, "offset": 28294970, "length": 196170 }, "model.layers.41.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331565, "offset": 28491140, "length": 225210 }, "model.layers.41.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331595, "offset": 28716350, "length": 224926 }, "model.layers.41.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 28941276, "length": 41 }, "model.layers.42.input_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 28941317, "length": 41 }, "model.layers.42.mlp.down_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 12288 ], "dtype": "bfloat16", "nnz": 50331600, "offset": 28941358, "length": 196022 }, "model.layers.42.mlp.gate_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331590, "offset": 29137380, "length": 224973 }, "model.layers.42.mlp.up_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 12288, 5632 ], "dtype": "bfloat16", "nnz": 50331593, "offset": 29362353, "length": 225040 }, "model.layers.42.post_attention_layernorm.weight": { "mode": "sparse_delta_u32", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 4096, "offset": 29587393, "length": 41 }, "model.layers.42.self_attn.k_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.42.self_attn.k_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194296, "offset": 29587434, "length": 18761 }, "model.layers.42.self_attn.o_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 5632, 4096 ], "dtype": "bfloat16", "nnz": 16777208, "offset": 29606195, "length": 65306 }, "model.layers.42.self_attn.q_norm.weight": { "mode": "dense", "shape": [ 256 ], "dtype": "bfloat16", "nnz": 256 }, "model.layers.42.self_attn.q_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 8192, 5632 ], "dtype": "bfloat16", "nnz": 33558472, "offset": 29671501, "length": 151662 }, "model.layers.42.self_attn.v_proj.weight": { "mode": "sparse_delta_u32", "shape": [ 1024, 5632 ], "dtype": "bfloat16", "nnz": 4194303, "offset": 29823163, "length": 18727 }, "model.norm.weight": { "mode": "dense", "shape": [ 5632 ], "dtype": "bfloat16", "nnz": 5632 }, "model.embed_tokens.weight": { "mode": "sparse_delta_u32", "shape": [ 478656, 5632 ], "dtype": "bfloat16", "nnz": 1042497871, "offset": 29841890, "length": 11390146 }, "lm_head.weight": { "mode": "sparse_delta_u32", "shape": [ 478656, 5632 ], "dtype": "bfloat16", "nnz": 1022438491, "offset": 41232036, "length": 4678381 } } }