Fetching the paper…

TensorGPT: Efficient Compression of Large Language Models based on Tensor-Train Decomposition · Around