Go into the options and set the CPU power limit to 253w. This will set long/short term power to max of Intel limit of 253w. Otherwise you're going to pull upwards of 350w+++ and throttle like crazy. 
 
Then undervolt with an offset of -50 to -100 depending on silicon quality you got. Maybe more, maybe less. I also use +50%/more vdroop setting. 
 
 

A couple notes.
 
You have CPU Multiplier Control set on "PerCore", which tries to set fixed speeds on specific cores. I believe it tries to figure out what your "best" cores are and put the highest multiplier on those. In your case, I'd be surprised if any of your cores can't do 6GHz, so I'd opt for the "RatioLimit" option and then just set 60 on the first two and 57 on the rest. This will set it to do any cores up to 6GHz as long as the work-load is on 2 cores or less.
 
13900k, but you get the idea:

 
As for your temperatures, what load were you running during your HWiNFO screenshot? The variance in temperature core-to-core is very high. I don't even see that if I'm running Cinebench single core tests. This is usually indicative of a bad mount (bad thermal paste coverage, uneven pressure of your block, or sometimes really bad IHS solder inside the CPU. The only other thing I can think of that may cause this sort of thing is if the voltages are just completely out of whack. What voltage settings do you have right now? What settings did you try to lower Vcore? Did you set an override and see if a fixed voltage would work? You should see measurable differences on VIDs and package power by doing that. Of course going too low will lead to instability, but I've tried the global adaptive offsets and didn't see any change at all even on 13900k (which is supposed to work in this BIOS lol), so I don't think that'll work for you.
 
Also, as a good thing to try, I agree with Talon on the power limit setting.

Setting power limit to 253 resolved the throttling, but not out of the woods yet. I agree that the core temp differences are too high. This was happening as well when I was first testing with the 240mm AIO, I just didn't think it could keep up. It's been the same with this 420mm, and I've remounted it multiple times now to see if it was just a bad mount on my part. I was running it at 1.410 vcore override btw.
 
The issue I'm running into now is that I'm still getting a power limit exceeded across all P and E cores. I even lowered the max power to 251 to see and still happening. Cores clock down quite a bit. Here's a screenshot of the limits that were hit during the test.

 
 
EDIT: Here's a screenshot of my CPU Configuration settings fwiw

1.41 seems too high to me. I would think you could override down in the 1.3-1.35 region, some people are running even lower, but they have the proper microcode and stuff. The other option is to try and set voltage on auto and mess with the v/F points, but if you go too low (which is a mystery number because evga doesn't expose actual voltages), I think it ignores it and just runs auto, which is too high. I can generally set and test mine down to the point of instability though, but it seems very coarse in the adjustment for me, plus it seems like it's still doing a global offset the way I have it set up, even though it's not supposed to... Which is the part of this bios I think is still broken even for 13900k.

Anyway, if you can get it to drop the voltages a bit, it'd be interesting to see if it still hits crazy Temps on a couple cores (especially if it's the same cores every time) because that may confirm it's the solder TIM in the IHS. If you take the block off again, take a picture so we can see how the paste looks

Moved the vcore to 1.35 override, temps didn't cross 80 on any of the cores, but still power limit throttling. At idle the cores are jumping up to the right numbers, but they throttle down to around 3ghz on the p-cores after about 5 seconds of Cinebench. Something is still causing it to power limit throttle. 
 
Do you think any of those limits I posted in the screenshot above give any ideas? The specific limits were...
IA: Package-Level RAPL/PBM PL2,PL3
IA: Max Turbo Limit
GT: Max VR Voltage, ICCmax, PL4
RING: Max VR Voltage, ICCmax, PL4
 
I'm planning on trying a different 14900k once it comes in the mail in the next day or so. I'll take a picture of the thermal paste once I do.

I have adaptive boost and TVB disabled (I left TVB voltage optimizations on auto). I think you'll need more than 251W to get to full speed. You got it to stop thermal throttling because it's power throttling with the limit you set. You may want to try 280W, and then keep dialing the override down if it's stable. Just be careful, as it will stop being stable during 6GHz boosts before it stops being stable at all core loads when using override, especially if it's throttling and dropping clock speeds you will have issues opening a web browser or running single core bench but all core will be stable lol

Something weird is happening here. I've upped the power limit to 280 (and disabled Adaptive Boost and TVB), and I'm still getting power limit exceeded and clocks just keep on downclocking. The core temps themselves are fine now, those are under control now that I have a limit put in place.
 
Here's what's also weird. I was still getting power limit exceeded even before I put an actual power limit in place. If you look at my screenshot from my first post today, you can see at the bottom how I'm both thermal throttling AND power throttling. SOMETHING weird is happening here.

Well, the most likely culprit is that the board isn't reading or setting V/F levels correctly due to the wrong microcode (no BIOS support from EVGA). I don't have time today, but maybe someone can run their 13900k and see if they see the same flags thrown in hwinfo. Also, it's possible that there are fast pulses that hit the power or thermal limit that just aren't being caught. You can try to increase the monitor speed, but it only goes so fast. That said, I'd be curious what your other 14900k does.

Anyway, if your Temps are under control with a lower override set, I'd start ramping up the power and see if you can get to a point where you're not dropping clocks and the temperatures are still bearable.

Forgot to take a picture of the paste, but have a few more updates.
 
The new 14900k is in and temps are a bit better with a little less variances in core temps. Unfortunately still power throttling. I've tried upping the power limit to 300, 350, and 400. Starts to thermal throttle again at 400, but I did notice that the cores didn't downclock as much at 400 compared to the lower limits. I was getting around 5ghz as opposed to 3-4ghz when testing lower limits. So it's TRYING to boost to the appropriate clocks.
 
I'm not convinced it's a cooling issue at this point. To Talon's point, 400w is very high so I'm not surprised I would throttle again at that level. 350w was actually fine, never crossed 90c on any cores. I'll keep trying to lower the vcore just for laughs to see if I can get temps lower, but if at 400w it's still thinking there isn't enough power and downclocks, then I'm not gonna have much wiggle room between upping the power limit and reducing vcore.
 
Where in the bios is the monitoring speed you mention?
 
Appreciate all your help with this. Odds are I'm just facing bios support issues, but regardless, thank you for your patience and help.

If the voltage is on auto and you run flat out no power limits, I'm betting it'll try and pull at least 390W, maybe more if not thermal throttling first. That said, the whole point of 14900k was that it's binned to not need that much. So most people I've seen running override voltages are much lower. I would expect it to thermal throttle if it's trying to pull over 325W on your setup. So if you get the voltages low enough and are still stable, you shouldn't have to even set a power limit. I'd keep trying, but you may just find a performance limit for your board until a BIOS update with a new microcode comes out (if it does) and it may be worth it to try and sell this one to get one that supports your CPU.