We consider the tails of probability density function (PDF) for the ve
locity that satisfies Burgers equation driven by a Gaussian large-scal
e force. The saddle-point approximation is employed in the path integr
al so that the calculation of the PDF tails boils down to finding the
special field-force configuration (instanton) that realizes the extrem
um of probability. For the PDFs of velocity and its derivatives u((k))
= partial derivative(x)(k)u, the general formula is found: In P(\u((k
))\) proportional to -(\u((k))\/Re-k)(3/(k+1)).