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

    Neuronpedia

    INDEX
    Explanations

    instance volume configuration

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    erten
    0.80
    isholm
    0.75
     cù
    0.75
     wasteland
    0.72
    IsDir
    0.72
    xb
    0.70
     کمار
    0.69
    ชร์
    0.69
    érir
    0.69
     காவல்துற
    0.69
    POSITIVE LOGITS
    专
    0.65
     tourn
    0.61
     boycot
    0.61
     نز
    0.61
     siêu
    0.60
     પ્રતિ
    0.60
     boycott
    0.59
    Wait
    0.59
     मार्
    0.59
    зал
    0.57
    Act Density 0.003%

    No Known Activations