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

    Neuronpedia

    INDEX
    Explanations

    clarifying questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    問
    0.85
     Asking
    0.82
     asking
    0.81
    Asking
    0.81
     問
    0.75
     ask
    0.75
    Ask
    0.73
     Ask
    0.70
     bertanya
    0.69
    问
    0.69
    POSITIVE LOGITS
     clarifying
    0.87
     follow
    0.73
     followup
    0.63
     probing
    0.63
    follow
    0.58
     pertinent
    0.55
     insightful
    0.54
     qualifying
    0.54
     Follow
    0.53
     logistical
    0.53
    Act Density 0.086%

    No Known Activations