transform_weight broken for uint1 #311

import bitblas
import torch

M = 1
NK = 32

matmul_config = bitblas.MatmulConfig(
    M=M,  # M dimension
    N=NK,  # N dimension
    K=NK,  # K dimension
    A_dtype="float16",  # activation A dtype
    W_dtype="uint1",  # weight W dtype
    accum_dtype="float32",  # accumulation dtype
    out_dtype="float16",  # output dtype
    layout="nt",  # matrix layout, "nt" indicates the layout of A is non-transpose and the layout of W is transpose
    with_bias=False,  # bias
    # configs for weight only quantization
    group_size=None,  # setting for grouped quantization
    with_scaling=False,  # setting for scaling factor
    with_zeros=False,  # setting for zeros
    zeros_mode=None,  # setting for how to calculating zeros
)
matmul = bitblas.Matmul(config=matmul_config)

t1 = torch.zeros((NK,NK), dtype=torch.uint8).cuda()
t1[0,0:8]+= 1

out1 = matmul.transform_weight(t1)

t2=t1.clone()
t2[0,8]+=1
out2 = matmul.transform_weight(t2)

print(torch.equal(out1, out2)) # This should not be true, but it is
print(torch.equal(t1, t2))