AI.info

Alex Pretko

Explore Alex Pretko on AI.info.

• Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference