Compact Language Models via Pruning and Knowledge Distillation | Read Paper on Bytez