Exploring Quantization for Efficient Pre-Training of Transformer Language Models

Open in new window