DFA + residual-branch penalty test: seed=456, lam=0.0001 ep 0: ||h_L||=8.890e+00 ||g_2||=9.971e-04 acc=0.0928 ep 1: ||h_L||=3.954e+03 ||g_2||=2.489e-06 eval_acc=0.2988 test_acc=0.3043 ep 10: ||h_L||=1.636e+04 ||g_2||=8.429e-07 eval_acc=0.3506 test_acc=0.3490 ep 20: ||h_L||=2.166e+04 ||g_2||=7.285e-07 eval_acc=0.3721 test_acc=0.3601 ep 30: ||h_L||=2.285e+04 ||g_2||=7.001e-07 eval_acc=0.3711 test_acc=0.3597 FINAL test acc: 0.3597 Compare to: DFA-vanilla (3-seed mean): 0.308 DFA-shallow (3-seed mean): 0.349 DFA-frozen (3-seed mean): 0.349 BP-trainable (3-seed mean): 0.609 Saved results/dfa_pen_short/dfa_pen_lam0.0001_s456.json Saved results/dfa_pen_short/dfa_pen_lam0.0001_s456.pt