I am new to Dynare. The linearized version of the model I want to simulate has some equations with variables with big ratios (of non linearized variables), for example:

No, that does not work this way. Y and y_t are different objects. Y is the steady state level of output, while y_t is the percentage deviation of output from its steady state, i.e.

Usually you need to compute the steady state values and define them as parameters. Have 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. There you can find an example.