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

    Neuronpedia

    INDEX
    Explanations

    say "percent" problems

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    abol
    -0.08
    gende
    -0.07
    berta
    -0.07
    ems
    -0.07
    bos
    -0.06
    .Messages
    -0.06
     helm
    -0.06
    ensis
    -0.06
    .ph
    -0.06
     CB
    -0.06
    POSITIVE LOGITS
     поводу
    0.10
     fotoana
    0.10
     זמן
    0.10
     часу
    0.10
     kurzen
    0.09
     години
    0.09
    ирад
    0.09
     बजे
    0.09
     अवधि
    0.09
     хватает
    0.09
    Act Density 0.011%

    No Known Activations