summaryrefslogtreecommitdiff
path: root/results/ep_d512/ep_s42.json
diff options
context:
space:
mode:
authorYurenHao0426 <Blackhao0426@gmail.com>2026-04-07 23:58:04 -0500
committerYurenHao0426 <Blackhao0426@gmail.com>2026-04-07 23:58:04 -0500
commitab1b783c7a4f3d586d082ba142d7c046453a310c (patch)
tree78a19b2baceea99f86f3608ce0d6e7f728649a78 /results/ep_d512/ep_s42.json
parentcbe851cf382a2af13037304afdd783214bad5c6b (diff)
CHECKLIST pitfall #6: layer-0 dominance is ResMLP-specific, not universal
Verified by extracting per-layer gamma_dfa from existing ViT-Mini snapshot JSON (3 seeds, final epoch). On ViT all 4 layers have per-layer cosine near zero (~0.001 with eps clamp); no layer dominates. Compare to ResMLP where layer 0 has +0.42 and layers 1-4 are essentially zero. The pitfall is real on ResMLP but the specific 'layer 0 dominates' framing doesn't generalize to ViT. Reframed as 'aggregation hides per-layer structure'; lesson is to always report per-layer values regardless of which architecture-specific pattern you might be hiding.
Diffstat (limited to 'results/ep_d512/ep_s42.json')
0 files changed, 0 insertions, 0 deletions