In your equation,
[tex]\sigma_{\mu\nu}[/tex]
means the mu'th-nu'th component of the tensor (or matrix) sigma. When you have an expression like
[tex]\sigma_{\mu\nu}F^{\mu\nu}[/tex],
Einstein's summing convention is implied - that is, you should sum over repeated indices, in this case mu and nu, from zero to three. It is a kind of "dot product" between the matrices sigma and F. Typically, you will need to know [tex]\sigma_{\mu\nu}[/tex] for all mu and nu to actually calculate this. The difference between upper and lower indices is that (depending on convention), for a four-vector,
[tex]f^{\mu} = g^{\mu\nu}f_{\nu}[/tex]
where g is the 4x4 matrix that has zero in all positions when you're not on the diagonal, and it has 1 in its first diagonal position and -1 in the last three positions. Thus, [tex]f^0 = f_0[/tex], and [tex]f^i = -f_i[/tex] for i = 1, 2 or 3. For a matrix, we would then write
[tex]\sigma^{\mu\nu} = g^{\alpha\mu}g^{\beta\nu}\sigma_{\alpha\beta}[/tex]
It's not very simple, but this is standard notation in relativity, so if you get the hang of this, a lot of stuff becomes easier..