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

    Neuronpedia

    INDEX
    Explanations

    expressions of fear and anxiety

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    fear and being scared

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    httphttps
    -0.69
    <unused51>
    -0.60
    <pad>
    -0.60
    [@BOS@]
    -0.60
    <unused47>
    -0.60
    <unused52>
    -0.60
    <unused41>
    -0.60
    <unused17>
    -0.60
    <unused8>
    -0.60
    <unused14>
    -0.60
    POSITIVE LOGITS
     fear
    1.02
    fear
    0.94
     terrified
    0.93
     scared
    0.93
     frightened
    0.90
    scared
    0.85
    Fear
    0.85
     Fear
    0.82
     afraid
    0.81
     peur
    0.80
    Act Density 0.039%

    No Known Activations