amygdala: default to index 0 for v2 deep manifest (layers 62, 63)
v2 retraining (readout_v2_paired) fixed the broken clusters — anger, sexual, high_pos, and social_pos all flipped from anti-clustered to positively clustered at deep layers. Validation showed layers 62 and 63 give the best signal; paring the serve-side manifest down to just those two keeps response size tight (~2 KB/token) while keeping the A/B option between the two strongest layers. Co-Authored-By: Proof of Concept <poc@bcachefs.org>
This commit is contained in:
parent
50d5b3f6e1
commit
b8714e8b3a
1 changed files with 7 additions and 7 deletions
|
|
@ -54,13 +54,13 @@ enum DisplayMode {
|
|||
impl AmygdalaScreen {
|
||||
pub fn new() -> Self {
|
||||
Self {
|
||||
// Default to layer 62 — validation across all 64 layers
|
||||
// showed 58 has high cohesion but also high cross-cluster
|
||||
// contamination (fear-state bleeds into sadness, etc.),
|
||||
// while 60-62 are cleaner discriminators. With the deep
|
||||
// manifest (layers 58, 60, 62, 63), index 2 = layer 62.
|
||||
// Bounded down to the actual layer count at render time.
|
||||
selected_layer: 2,
|
||||
// Default to layer 62 — clean cross-cluster discrimination
|
||||
// with good within-cluster cohesion. With the v2 deep
|
||||
// manifest (layers 62, 63), index 0 = layer 62 and
|
||||
// index 1 = layer 63 (sharper but noisier on some
|
||||
// dimensions). Bounded down to actual layer count at
|
||||
// render time.
|
||||
selected_layer: 0,
|
||||
mode: DisplayMode::MeanRecent,
|
||||
display_indices: Vec::new(),
|
||||
normalize: true,
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue