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

    Neuronpedia

    INDEX
    Explanations

    potential value variations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.46
    dued
    0.44
    given
    0.42
    cion
    0.41
    rous
    0.40
    cellaneous
    0.40
    reve
    0.39
    ERCIAL
    0.39
    身
    0.39
     rigu
    0.38
    POSITIVE LOGITS
     którym
    0.53
     که
    0.52
     që
    0.52
     Öl
    0.51
    kę
    0.50
     ambayo
    0.50
     renovación
    0.50
     Lobkowicz
    0.48
     fatta
    0.48
     благодар
    0.48
    Act Density 0.001%

    No Known Activations