RuntimeError: The size of tensor a (10) must match the size of tensor b (3) at non-singleton dimension 1

Here is my model:

<bound method Module.state_dict of model(
(hidden_layers): ModuleList(
(0): Linear(in_features=3, out_features=10, bias=True)
(1): Linear(in_features=10, out_features=10, bias=True)
)
(output_layers): ModuleList(
(0): Linear(in_features=10, out_features=4, bias=True)
(1): Linear(in_features=10, out_features=4, bias=True)
)
)>

I think your model input has a wrong dimension: the model expects 3 (…x3), but you give it 10 (…x10)…As the message says

More details about this would be useful to help you

Thanks for reply,
when first example of 1*3 is passed through first hidden layer it works,
below is weight matrix 3 * 10 of first hidden layer:
(Pytorch output):
tensor([[ 0.4778, 0.4947, -0.4435],
[-0.0535, 0.5000, 0.1650],
[-0.5120, 0.5192, -0.0072],
[-0.1666, 0.0537, 0.5399],
[-0.5655, -0.4967, 0.0043],
[ 0.1430, 0.0842, -0.4497],
[-0.1533, 0.2130, -0.4762],
[-0.2144, -0.3397, -0.1646],
[-0.1230, -0.3268, 0.4827],
[-0.4620, -0.3071, 0.3321]], dtype=torch.float64)

when second example of same batch is passed then it throws error of size mismatching.Below is the weight matrix of first hidden layer after updating

tensor([[-1.9652e-05, -5.2278e-04, 5.1098e-05],
[ 6.4976e-06, 1.7285e-04, -1.6894e-05],
[ 3.9244e-06, 1.0440e-04, -1.0204e-05],
[-9.4330e-06, -2.5093e-04, 2.4527e-05],
[ 8.6443e-06, 2.2995e-04, -2.2476e-05],
[ 4.9573e-06, 1.3187e-04, -1.2889e-05],
[-9.0200e-07, -2.3995e-05, 2.3453e-06],
[-9.8638e-06, -2.6239e-04, 2.5647e-05],
[ 8.2014e-06, 2.1817e-04, -2.1324e-05],
[-8.1189e-06, -2.1598e-04, 2.1110e-05]], dtype=torch.float64)

Dimensions are same for the second example also how could it be possible ??