Hi @carolineecho. From what you are describing, I think the problem is that Brian cannot find the value of an external constant that you reference in your equations (I cannot be entirely sure, since the error message you posted is not complete). External constants need to be available where you run the simulation, not where you define the NeuronGroup. If you prefer to define constants together with the NeuronGroup, you can use its namespace argument. See the documentation here: Namespaces — Brian 2 2.5.4 documentation
Also, have a look at another forum post that raised a similar question recently: KeyError: 'The identifier "a" could not be resolved.' - Izhikevich Model
Hope that helps, best
Marcel
PS: Note that for better readability of code snippets, you can enclose them in triple backticks like this:
```
# a comment
print("My code")
```