Here's a thought -- if you have access to a number of circuit breakers of known value, you can use them to bracket in the value of your unknown breaker.
It goes something like this:
Say you have a 30 Amp breaker, put it in series with your unknown breaker [DUT] and ramp the current up by adding 1000W bulbs. If the 30 Amp breaker trips before your DUT breaker, then the DUT breaker is greater than 30 amps.
Next try a 40 Amp breaker in series with your DUT...then a 50 Amp... keep doing this until the DUT trips. At that point you know the DUT breaker has a trip point between the value of the final breaker and that of the previous breaker.
But, that approach may only be practical if you don't need to know the precise trip point.
OR, you can use this technique to get into the ballpark and then use some other method to hone in on the precise value (the advantage being that your test setup doesn't have to be as variable).
Also, in rebuttal to my previous comment regarding the efficacy of using lamps as test loads, if the voltage across the lamps is the same in both cases (i.e. when tested alone and when used in parallel with other lamps as a load) then the load they provide is predictable. But, at such high currents, more than likely, the voltage will decrease as lamps are added in parallel, in which case, not so predictable.
One way around that would be to use a variac to adjust the voltage across a single lamp down to the same voltage that is present across numerous lamps and then measure the current through that single lamp. Also, it would be wise to measure the current through each lamp, to make sure they're consistent.
Example: With three lamps in parallel, measure the voltage across the lamps. Then, connect a single lamp to a variac and adjust the variac voltage to the same voltage as was across the three lamp parallel combination, and then measure the current through that one lamp. This current, can then be multiplied by three to determine the three-lamp load current (assuming that all three lamps draw the same current).
Also, because line voltage can vary, make sure to monitor the voltage across the lamps at all times, and make adjustments (i.e. re-calibrate) as needed.
And, hopefully the labor involved in determining the trip point of these unknown breakers is worth the cost savings