faeval.git/experiments/blend_mechanism_dissection.py, branch master Unnamed repository; edit this file 'description' to name the repository. Add Phase 10A.5: blend gain is implicit regularization, not learned credit 2026-03-26T21:27:53+00:00 YurenHao0426 Blackhao0426@gmail.com 2026-03-26T21:27:53+00:00 610e1169e19378cccd2d9b92a588c24dca7f3df7 Dissection of 6 branches from same DFA checkpoint: - blend_random_frozen: 12.6% (CATASTROPHIC — frozen noise destroys training) - blend_random_trainable: 32.2% (+1.2% — trainable network helps) - blend_shuffled_trainable: 32.5% (+1.4% — even wrong targets work!) - blend_gaussian_noise: 30.8% (neutral) - scaled_DFA_norm_match: 31.0% (neutral) The gain comes from implicit regularization through a co-optimized auxiliary network, NOT from learned credit quality. Phase 9A's +1.5% was an optimization dynamics effect, not evidence of useful credit assignment. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> 
Dissection of 6 branches from same DFA checkpoint:
- blend_random_frozen: 12.6% (CATASTROPHIC — frozen noise destroys training)
- blend_random_trainable: 32.2% (+1.2% — trainable network helps)
- blend_shuffled_trainable: 32.5% (+1.4% — even wrong targets work!)
- blend_gaussian_noise: 30.8% (neutral)
- scaled_DFA_norm_match: 31.0% (neutral)

The gain comes from implicit regularization through a co-optimized auxiliary
network, NOT from learned credit quality. Phase 9A's +1.5% was an optimization
dynamics effect, not evidence of useful credit assignment.

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