From 1118b7457c261de36ead6103503c00c321c75f9b Mon Sep 17 00:00:00 2001 From: YurenHao0426 Date: Sun, 14 Jun 2026 20:32:31 -0500 Subject: Depth-utility ladder: trainable-block sweep (BP/FA/DFA) on ResMLP CIFAR-10 Appendix experiment triangulating the depth-utility diagnostic (D3) by varying the number of trainable residual blocks k (last-k trainable, first L-k frozen at init; embed/LN/head always trained). - d=256 L=4 and d=512 L=2, 3 seeds, recipe identical to the main audit. - BP climbs monotonically (+22-23pp); DFA peaks at the frozen baseline (k=0) and declines once any deep block is trained; FA shows partial/no net depth utility. - Cross-checks reproduce existing anchors (BP 0.617, DFA 0.301, FA 0.402, frozen 0.349). - frozen_init_identity_check quantifies frozen stack as a near-norm-preserving random feature map (per-block ||f||/||h||~0.10, stack cos 0.981), explaining the above-chance k=0 rung. Co-Authored-By: Claude Opus 4.8 (1M context) --- results/depth_ladder/depth_ladder.png | Bin 0 -> 89453 bytes 1 file changed, 0 insertions(+), 0 deletions(-) create mode 100644 results/depth_ladder/depth_ladder.png (limited to 'results/depth_ladder/depth_ladder.png') diff --git a/results/depth_ladder/depth_ladder.png b/results/depth_ladder/depth_ladder.png new file mode 100644 index 0000000..5fd1f81 Binary files /dev/null and b/results/depth_ladder/depth_ladder.png differ -- cgit v1.2.3