Find виртуальный GPU resource allocation on server

I have been provided в виртуальный server, which runs Ubuntu 14.04. гm accessing it using PuTTY. The server should have в NVIDIA GRID K2 GPU, but гm having some issues using it in Python/C ++ with the Caffe library. Особенный In, I get the errors modprobe: FATAL: Module nvidia not found and Check failed: error == cudaSuccess (38 vs. 0) no CUDA-capable device is detected.

гd like to check that the виртуальный GPU is actually allocated to the machine. How хан I do that using the command line?