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

    Neuronpedia

    INDEX
    Explanations

    ai safety and responsibility

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    phases
    0.42
    ering
    0.41
     amelior
    0.41
    opting
    0.40
    max
    0.40
    her
    0.39
    erc
    0.39
    og
    0.38
    ables
    0.38
     pollute
    0.38
    POSITIVE LOGITS
    貶
    0.43
    жень
    0.42
     sợ
    0.41
     Resize
    0.41
    วัด
    0.39
     Sección
    0.39
    許多
    0.39
     Missionary
    0.39
    良く
    0.38
    桝
    0.38
    Act Density 0.000%

    No Known Activations