faeval.git/results/confirmatory/cross_state_distance.csv, branch master Unnamed repository; edit this file 'description' to name the repository. Add EP cross-state distance for CIFAR + verify CNN summary 2026-04-03T15:58:22+00:00 YurenHao0426 Blackhao0426@gmail.com 2026-04-03T15:58:22+00:00 83b3e04e6856b9d9d6f3e75b6e955e5b78f9c787 EP CIFAR d_BP: L0=2.0×, L4=26.7× (much closer to BP than DFA=162×/2.5M×) EP synthetic: no checkpoints saved (ep_synthetic.py didn't save .pt) CNN summary: 20 rows confirmed correct Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> 
EP CIFAR d_BP: L0=2.0×, L4=26.7× (much closer to BP than DFA=162×/2.5M×)
EP synthetic: no checkpoints saved (ep_synthetic.py didn't save .pt)
CNN summary: 20 rows confirmed correct

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add cross-method hidden state distance vs BP 2026-04-03T15:34:33+00:00 YurenHao0426 Blackhao0426@gmail.com 2026-04-03T15:34:33+00:00 ae368dafa311946c5a57b07c4c2d3c78d92e48a4 Non-BP methods produce radically different representations: DFA L0: 162×, L4: 2.5M× relative to BP hidden norms SB L0: 3.2×, L4: 1.1M× CB L0: 59×, L4: 1.4M× (BP vs itself = 0) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> 
Non-BP methods produce radically different representations:
DFA L0: 162×, L4: 2.5M× relative to BP hidden norms
SB L0: 3.2×, L4: 1.1M×
CB L0: 59×, L4: 1.4M×
(BP vs itself = 0)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>