I tend to think of it as a big ball of pure state. The field has both wavelike behaviors and quantized interactions. And even then, a field is only a model of the underlying state. This is why we can meaningfully talk of different fields independently even though there's only one underlying state.