As in this post, you may have a scaling problem in the Jacobian matrix of your model evaluated at the deterministic steady state. The smaller value in this jacobian matrix is of order 1e-5 while the bigger is of order 1e3. In this case dgges in not able to compute the generalized schur decomposition. I think that the problem is related to the steady state value of b2. This value is too big compared to the other steady state values. I don’t know the model, but shouldn’t you have a constraint between exp(b1) and exp(b2) ?
Best,
Stéphane.