There are a few important steps that could impact the formation of biogenic amines. Most of them do not form during fermentation, i'm afraid.
The first is fermentation - fermentation can take up to 7 days, and can be done a variety of ways. One of the things that occurs during fermentation is the consumption of raw materials (i.e. sugars are consumed), which results in the increase of non-consumed components. This leads to a higher fat content, which is important for future processing. Unfermented cocoa beans, on their own, will have a fat content that is too low to enable the cocoa bent to be fluid once it's ground. Those processing it won't be able to pump it. Most free amino acids are in the pulp at this stage, which drains away. Pyrazines begin to form at this stage. Temperatures can get fairly high here (150F +) - since there is no definition of 'raw' - it becomes difficult to say if fermented beans are raw or not - however the general consensus in the raw community is that if temperatures exceed 118F, it's not raw. Since it's quite common for the ambient environmental temperatures where cocoa is grown to exceed 118F, never mind the temperatures during fermentation, most of the raw community wouldn't consider cocoa to be raw.
Next is drying - if you don't dry the beans, they'll have very high moistures and rot, or if low enough prevent rotting, they can still be high enough to give viscosity issues during further processing. Temperatures can be very high here, but most use the power of the sun. A complex cascade of internal chemical reactions is occurring here.
Next is roasting - very high temperatures. Here is where most of your amino changes are occurring, with lots of strecker degradation reactions occurring as a result of the thermogenic induction. This is also your primary micro kill step. It's reducing your moisture down to ~1% level so you can process it (again, it will be too viscous if you don't), and since raw cocoa beans have extraordinarily high microbiological counts (this is the reason they're not safe to consume pre-roasting), this is the step - and the only step - where those counts will be reduced to safe levels. Since flavor precursors are forming during fermentation (hence the pyrazine comment above), roasting of unfermented beans doesn't result in a flavor improvement.
For cocoa powder production - what happens next is the beans are ground up (turned into liquor) - where they become flowable. The grinding process itself generates heat (i'm unaware of any work that has been done to examine the changes in peptide profiles at this step), and then it goes to a very, very high pressure hydraulic press. Since this press requires the liquor to be flowable and fluid, if the beans were not fermented, the processor would not be able to do this. A very small amount of cocoa powder has it's butter extracted via supercritical fluid extraction, but that still requires the liquor to be fluid enough to pump it.
The net/net is - any commercial cocoa powder you purchase is highly likely to be fermented and roasted. Most biogenic amine formation is driven by thermogenic - not enzymatic - changes, and free amino acids (including tryptophan), are present in un-fermented cocoa to begin with. If you don't ferment the beans, the beans will be unprocessable from a viscosity standpoint; if you don't roast the beans, they will be unsafe to consume from a microbiological standpoint.
