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

    Neuronpedia

    INDEX
    Explanations

    Fear

    np_max-act-logits · gemini-2.5-flash-lite

    mentions of fear or expressions of being afraid/anxious.

    oai_token-act-pair · gpt-5-miniTriggered by @lporourke621

    The neuron responds to occurrences of the word “fear,” effectively detecting expressions of fear.

    oai_token-act-pair · o4-miniTriggered by @lporourke621
    New Auto-Interp
    Negative Logits
     frustrations
    0.64
     impatience
    0.63
     frust
    0.59
     শোক
    0.59
     frustration
    0.58
     annoyance
    0.58
    тей
    0.57
     экране
    0.56
     ढूंढ
    0.55
     Honorary
    0.55
    POSITIVE LOGITS
    mong
    0.93
     lest
    0.84
     heights
    0.80
    repr
    0.75
     repris
    0.72
     repr
    0.72
    lessly
    0.70
    fully
    0.69
    Unknown
    0.69
     mong
    0.68
    Act Density 0.038%

    No Known Activations