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

    Neuronpedia

    INDEX
    Explanations

    conditional and error handling code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    aconda
    0.38
    腎
    0.37
    oniazid
    0.36
     bringt
    0.36
    sonic
    0.36
    getView
    0.36
    verw
    0.36
     Wohnung
    0.36
    ünden
    0.35
    Դ
    0.35
    POSITIVE LOGITS
     an
    0.40
     Ethel
    0.40
     https
    0.38
     Merit
    0.38
     deepening
    0.38
     bereavement
    0.38
     그리
    0.38
     sadness
    0.37
     Similarly
    0.37
     anyone
    0.37
    Act Density 0.004%

    No Known Activations