Hi Yael,
That call works for me OK and retrieves and displays the Hebrew correctly. Can you give some more information on which version of OpenRefine you are using and on which platform (Windows/Linux/Mac)?
One option with mis-encoded data is to use the `reinterpret` transformation - this will try to reinterpret the data using an encoding scheme.
e.g. value.reinterpret("utf-8")
You could see if that is a solution, although it would be good (from the point of view of solving the underlying problem) to understand why it comes through incorrectly for you and correctly for me
Best wishes,
Owen