Add synthetic FA trajectory validation

1 files changed, 22 insertions, 0 deletions

diff --git a/scripts/README.md b/scripts/README.md
index f24c117..42633d0 100644
--- a/scripts/README.md
+++ b/scripts/README.md
@@ -111,3 +111,25 @@ and:
 \]
 
 Outputs are written under `outputs/functional_capacity_overlap/`.
+
+## Synthetic MLP FA/BP Trajectories
+
+Run:
+
+```bash
+python scripts/trajectory_mlp_fa.py --samples 128 --hidden-widths 24 24 --steps 80 --lr 0.02 --eval-every 10 --feedback-runs 3 --data-seed 3 --init-seed 4 --feedback-seed-start 50 --plot
+```
+
+This trains one BP baseline and several FA runs from the same initial weights on
+a synthetic regression task. At each checkpoint, the script records:
+
+- training loss;
+- full-model cosine between the BP gradient and the FA surrogate gradient at the FA weights;
+- layerwise \(Q_l=\cos^2(W_{l+1}^{\top},B_l)\).
+
+Outputs are written under `outputs/trajectory_mlp_fa/`:
+
+- `summary.csv`
+- `trajectories.csv`
+- `layer_metrics.csv`
+- diagnostic plots when `--plot` is set.