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

    Neuronpedia

    INDEX
    Explanations

    I recognize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     privacy
    0.93
     phrases
    0.92
     always
    0.89
     politely
    0.85
     requests
    0.84
    anyakan
    0.82
     queries
    0.81
    请求
    0.81
     dialogue
    0.79
     polite
    0.79
    POSITIVE LOGITS
     parcialmente
    1.03
     dennoch
    0.90
     debacle
    0.89
     Partial
    0.86
     succumb
    0.84
     succumbed
    0.84
     компен
    0.83
     partiellement
    0.81
     توڑ
    0.79
     enraged
    0.79
    Act Density 0.934%

    No Known Activations