INDEX
    Explanations

    This neuron detects HTML references to site icons (e.g. “icon”/“shortcut icon” and favicon URLs).

    New Auto-Interp
    Negative Logits
     Aph
    -0.07
    ,private
    -0.06
     tablets
    -0.06
     hiçbir
    -0.06
     legends
    -0.06
     Hud
    -0.06
    ","#
    -0.06
     Alley
    -0.06
    ();
    ↵
    ↵
    -0.06
     ugl
    -0.06
    POSITIVE LOGITS
     favicon
    0.08
    atura
    0.07
     evacuate
    0.07
    /favicon
    0.07
     wirk
    0.07
    ρωση
    0.07
    aver
    0.07
     *.
    0.07
     investigating
    0.06
    mín
    0.06
    Act Density 0.001%

    No Known Activations