Step 1: Simplify \(s_n\).
For \(n\) odd:
\[
s_n = \frac{1}{2^n}\cdot \frac{4^n - 1}{3} \approx \frac{1}{3}\left(2^n - 2^{-n}\right).
\]
For \(n\) even:
\[
s_n = \frac{1}{2^n}\cdot \frac{4^{n/2 - 1} - 1}{3} \approx \frac{1}{3}\left(2^{n-2} - 2^{-n}\right).
\]
Step 2: Observe the behavior for large \(n\).
As \(n \to \infty,\) the dominant term alternates between approximately
\(\frac{1}{3} \times 2^{-2}\) and \(\frac{1}{3} \times 2^{0}\),
yielding two distinct limiting subsequences for even and odd \(n.\)
Step 3: Average over terms.
The sequence \(\sigma_m = \frac{1}{m}\sum s_n\) inherits these oscillations;
thus, \(\sigma_m\) has two distinct limit points corresponding to even and odd averaging limits.
Step 4: Conclusion.
Therefore, the number of limit points of \(\{\sigma_m\}\) is \(\boxed{2}.\)