states_buf has stride pad128(state_dim) for CUTLASS alignment. The concat_ofi kernel was reading with stride state_dim → OOB for sample 1+. With state_dim=50, pad=128: sample 1 OFI at offset 128+42=170, but kernel read at 50+42=92 (sample 0 padding area). Silent data corruption. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>