text-generation-inference

f1f98e36 - Add support for Marlin 2:4 sparsity (#2102)

Add support for Marlin 2:4 sparsity (#2102) This change adds support for 2:4 sparsity when using Marlin quantization. The 2:4 kernel is used when: * The quantizer is `marlin`; * the quantizer checkpoint format is `marlin_24`. Fixes #2098.