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

    Neuronpedia

    INDEX
    Explanations

    stop thinking know

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     trabaja
    0.60
     esperan
    0.57
    বাহী
    0.55
    ˀ
    0.53
     trabaj
    0.52
     quiere
    0.52
     lançado
    0.52
     egip
    0.51
    ק
    0.51
     indígena
    0.51
    POSITIVE LOGITS
    .
    0.68
    astro
    0.46
    string
    0.44
    -
    0.41
     regulation
    0.40
    element
    0.39
    substrate
    0.39
    helping
    0.38
    ontrol
    0.38
     Valuation
    0.38
    Act Density 0.010%

    No Known Activations