$\sum_{n=1}^{\infty}(-1)^{n}(2^{-n}+n^{-1})$ converges to $-\frac13-\log_e(2)\approx -1.02648$
when read as the limit of the partial sums of $-\left(\frac1{2^1}+\frac11\right)+\left(\frac1{2^2}+\frac12\right)-\left(\frac1{2^3}+\frac13\right)+\left(\frac1{2^4}+\frac14\right)-\left(\frac1{2^5}+\frac15\right)+\left(\frac1{2^6}+\frac16\right)- \cdots $
but if you reorder, for example to two negative terms followed by one positive so still retaining all the terms in the infinite series, as in $-\left(\frac1{2^1}+\frac11\right)-\left(\frac1{2^3}+\frac13\right)+\left(\frac1{2^2}+\frac12\right)-\left(\frac1{2^5}+\frac15\right)-\left(\frac1{2^7}+\frac17\right) +\left(\frac1{2^4}+\frac14\right)-\cdots $ then the limit of the partial sums is closer to $-1.373$
Compare this with $\sum_{n=1}^{\infty}(-1)^{n}(2^{-n})$ which is absolutely convergent as $\sum_{n=1}^{\infty}|(-1)^{n}(2^{-n})|=1$ is finite: you then know that the limits of the partial sums of $-\frac1{2^1}+\frac1{2^2}-\frac1{2^3}+\frac1{2^4}-\frac1{2^5}+\frac1{2^6}-\cdots$ and of $-\frac1{2^1}-\frac1{2^3}+\frac1{2^2}-\frac1{2^5}-\frac1{2^7}+\frac1{2^4}-\cdots$ will converge to the same value, in fact $-\frac13$
This is the key point: if a series is absolutely convergent (i.e. the sum of the absolute values converges to a finite value) then the sum of the actual values will also be finite and will not depend on the order of the terms; if a series is conditionally convergent but not absolutely convergent then changing the order of the actual values can change the result and even make the magnitude of the limit of the partial sums arbitrarily large or infinite