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

    Neuronpedia

    INDEX
    Explanations

    setting configuration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    itattu
    0.40
     Wail
    0.39
    頸
    0.38
     ely
    0.38
    ParentElement
    0.38
     드
    0.37
    ineux
    0.37
    မိ
    0.36
     samme
    0.36
     mam
    0.35
    POSITIVE LOGITS
     unset
    0.61
    disable
    0.57
    設定
    0.57
     setDefault
    0.55
    Setting
    0.54
    の設定
    0.54
    setConfig
    0.52
    set
    0.52
    Disable
    0.52
     设置
    0.52
    Act Density 0.020%

    No Known Activations