First I would compare the data sheets to see the difference.
Both chips use Active Low Inputs and for the chip you are using the logic is:
Input High = >2.0 Volts
Input Low = <0.8 Volts
R1 through R9 are 560K pull up resistors so depending on the conductivity of the water in the tank between any probe and common if the resistance was 560K that would only get you about 2.5 Volts, still not anywhere close to low enough. I have no idea what the conductivity of your water is? Fill the tank above a few probes and measure the input voltages at the input pins. If the pins do not go below 0.8 volts then it won't work. Circuits like this are also very poor designs for measuring water tank levels because the probes will begin to corrode with minerals from the water. Anyway, you need to know the conductivity of the water before anything else. I am guessing that if everything is correctly wired the problem is the low inputs aren't low enough. You can try replacing R1 through R9 with between 1 Meg and as high as 10 Meg Ohm resistors.
SIDE NOTE: Measuring the tap water in my house as it comes in from the supply with probes spaced about 25mm the water resistance is right about 700K Ohms (700,000 Ohms). Using that circuit the best I could hope for would be about 2.777 Volts into the chip. Even using the original chip I would need to get below about 2.5 Volts for an active low. That circuit shows up all over the Internet and circuits like that simply don't work most of the time. They are a poor design.
Ron