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

    Neuronpedia

    INDEX
    Explanations

    helping them succeed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ส
    0.50
    ي
    0.50
    হ
    0.48
    ко
    0.47
    dding
    0.46
    ח
    0.46
    请求
    0.46
    দ
    0.44
    ด
    0.44
     participan
    0.44
    POSITIVE LOGITS
     канце
    0.46
     facture
    0.45
     dogma
    0.44
     Taxation
    0.43
     Lant
    0.42
     doubt
    0.41
     confess
    0.41
     fact
    0.41
     Looks
    0.40
    {
    0.40
    Act Density 0.002%

    No Known Activations