Accurate subseasonal predictions of high surface air temperature (SAT) and heat wave events 10–30 days in advance are crucial for mitigating the risks of extreme weather; however, they pose a challenge for current operational models. In this research, we implemented a convolutional neural network (CNN)-based deep learning model to leverage the modulations in China's SAT by precursor signals across various timescales to enhance predictions of future SAT and heat wave events. Our CNN model demonstrated superior ability in capturing the evolution of SAT anomalies and the occurrence of heat wave events with forecast lead times beyond 20 days, compared with that of the operational models of the China Meteorological Administration and European Centre for Medium-Range Weather Forecasts. Explainability analysis highlighted that subseasonal SAT predictability in China is primarily driven by large-scale intraseasonal perturbations from both lower- and higher-latitude regions of Eurasia, as well as interannual variability. Rather than focusing solely on specific timescale components, our findings suggest that considering interactions across multiple timescales could enhance subseasonal predictability.