Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    shield whenever, potentially leading, debt rapidly, almost subsonic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    scape
    0.51
    date
    0.48
    destination
    0.45
    habitat
    0.45
    arch
    0.45
    data
    0.45
    steel
    0.44
    mer
    0.44
    tree
    0.44
    guide
    0.44
    POSITIVE LOGITS
     threaten
    0.54
     Cronin
    0.47
     threatens
    0.46
     kwargs
    0.46
     đau
    0.46
     novelists
    0.46
     değerlend
    0.45
     denk
    0.45
     lesión
    0.45
     quando
    0.45
    Act Density 0.003%

    No Known Activations