Give her all she can take!
Short answer -- watch internodes and don't let them stretch. By the time the serrated, lobed leaves get larger, it'll probably handle full intensity. If they get too tight, that's too much light. if they stretch, that's too little light.
Why/how and ramblings.
Basically, this can be a little different for every plant, but you can use the same method and get consistent results.
Observe internode length. Internode is the part of plant between 2 nodes... it will 'stretch' if the plant is not provided enough light.
So, it's easy to tell when not giving enough light. How tight do you want the nodes? Again, needs a little trial and error. Plants have different growth patters, so there is no one-size-fits-all solution here.
Too much light -- plant will wilt... all over. not just the petioles and leaves. Extreme issues will result in pale spots forming near top in most intense areas of light. Before that, internode can be almost non existent. If you see 4+ leaves all originating from nearly the same spot and you don't have a mutation, it's very likely too much light. The nodes have grown too tight and everything will be on top of each other, which will be good for mold and other microbial life, maybe even pests too?
So, obeserve how the spacing of nodes works out, adjust any light.
if you have a cheap lux meter, thery are fine to use under any 1 light (same spectrum, cct, et al.). Don't take readings from 1 light and apply it to a different light unless it is exactly the same down to diodes. But, you can measure and see the lux value and the next plant will be very similar. you can set light height or dimmer, then observe for any adjustments based on what the plant does.
Subtle differences will exist. You are not far away from full-bore.
Ambient CO2 levels will impact how much you can give a plant. So will VPD (temp + RH%), but light intensity and co2 is the determining factor for optimal VPD. This is why how much a plant can take is variable...
So, be wary of anyone giving exact recommendations without 100% confidence. This is only meant to be used as a ballpark concept given all the factors that matter and that we rarely control them all in the same way.