INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    Accuracy
    -0.08
     in
    -0.07
     illustrative
    -0.07
     démarches
    -0.07
    Appro
    -0.07
    524
    -0.07
     impost
    -0.07
     nader
    -0.07
    GA
    -0.07
    Emb
    -0.07
    POSITIVE LOGITS
    0.09
     laughter
    0.09
    0.09
     COLORS
    0.08
     basin
    0.08
    gın
    0.08
    0.08
    Brown
    0.08
     bina
    0.08
    0.08
    Act Density 0.010%

    No Known Activations