The present work details further information regarding a new benchmark to be introduced to the international community, for dealing with neutronic code validation in the frame of the analysis of severe accidents in fast reactors leading to core degradation and material relocation. This specific benchmark is based on further analysis of selected experiments performed at the Schnelle Null-Energie-Anordnung Karlsruhe (SNEAK). The SNEAK-12B core was loaded with plutonium fuel to better represent future fast systems and the experiments considered fuel relocation and redistribution of structural material. In this paper, the experimental results are analyzed by state of the art tools, Serpent-2 Monte Carlo and the ERANOS code for deterministic calculations. The paper presents a full sensitivity and uncertainty analysis based on the JEFF-3.1.1 and the associated covariance data COMAC-V01, which is performed in order to gain deeper insight into the governing phenomena related to geometrical changes of the core. A comparison of the propagated uncertainties between Serpent and ERANOS is made. The uncertainty propagation vary from code to code, and strongly disagree in most cases of axial fuel relocation. This is evident for small reactivity variation ( 1 cent), where the difference in the propagated uncertainties obtained from the two codes is vividly visible. The analysis provide valuable information on uncertainty propagation in a system where the overall material balance is not modified, and contributes to the design of future experiments. This work is done within the frame of new core design capacities and innovative experimental programs to be implemented in Zero Power Reactors, such as the ZEPHYR project led independently by CEA.