Part I: Models, workload state, and GPU hardware
These chapters establish the work and its storage. Detailed diagnosis follows the execution mechanisms and the first controlled measurement in Part II.
Chapters in this part
- 1. Model operations and tensors: Follow text/token IDs through the model; read tensor shape, dtype and storage descriptors.
- 2. Workload phases and live memory: Trace a plain training/generation cycle before identifying which objects coexist.
- 3. GPU hardware and memory: Locate arithmetic and bytes in the device and distinguish capacity from transfer rate.
