Hi! Where could I find instructions how to write a separate steady-state file for a model?

Take a look at the NK_baseline.mod in the Dynare examples folder. The header and footer should not be altered. It says where you can enter your own equations.
However, if you are doing nothing fancy like calling a solver, go for a steady_state_model block. It will automatically create a steady state file. Take a look at Pfeifer(2013): “A Guide to Specifying Observation Equations for the Estimation of DSGE Models” sites.google.com/site/pfeiferecon/Pfeifer_2013_Observation_Equations.pdf for examples.

thank you for quick reply. But I have checked with this source before creating this post.

All I changed was putting exp() for variables and compute initial values using equations in steady state model part.

I was wondering if it is correct after using exp () for some of equations in your source :

1.log(tb)= (d) -(d(+1))(g)/(r) ----> (tb)= exp(d) - exp(d(+1))exp(g)/(exp(r));
2.log(g/gbar)=rho_g
log(g(-1)/gbar)+eps_g; ---->(g) = (1-rho_g)
(gbar)+g(-1)*rho_g+ e_g;
3.log(tb_y) = log(tb)/y;---->(tb_y) = (tb)/exp(y);
4.g_y= y/y(-1)*g(-1);---->(g_y) =(y)-y(-1)+(g(-1));

Please do not cross-post. See my reply at Impossible to find the steady state:Cicco, Pancrazi, and Uribe (2010) - #4 by jpfeifer