Machine learning applied to blockchain graphs offers extensive opportunities for advanced data analysis and application. However, the field's potential has been constrained by the lack of large-scale, cross-chain datasets that encompass hierarchical graph-level data. To address it, this paper introduces a novel dataset that provides detailed label information at the token level and integrates token-token interactions on multiple blockchain platforms. Specifically, we model transactions of each token as local graphs and the relationships between tokens as global graphs, collectively forming a Graphs of Graphs (GoG) system. This framework enables a deeper understanding of systemic structures and hierarchical interactions, essential for applications such as anomaly detection and token classification. We conduct a series of experiments demonstrating that this dataset delivers new insights and challenges for exploring GoG within the blockchain domain. Our work fosters further advancements and provides new opportunities in blockchain research and the graph community.