DFA + residual-branch penalty test: seed=42, lam=0.1 ep 0: ||h_L||=8.893e+00 ||g_2||=9.934e-04 acc=0.1152 ep 1: ||h_L||=7.313e+02 ||g_2||=1.234e-05 eval_acc=0.3330 test_acc=0.3271 ep 10: ||h_L||=6.932e+03 ||g_2||=2.318e-06 eval_acc=0.3467 test_acc=0.3430 ep 20: ||h_L||=1.083e+04 ||g_2||=1.721e-06 eval_acc=0.3564 test_acc=0.3531 ep 30: ||h_L||=1.157e+04 ||g_2||=1.634e-06 eval_acc=0.3594 test_acc=0.3486 FINAL test acc: 0.3486 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_lam0.1_s42.json Saved results/dfa_pen_short/dfa_pen_lam0.1_s42.pt