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

    Neuronpedia

    INDEX
    Explanations

    dangerous threats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Voraussetzungen
    0.44
    Conditions
    0.43
     disulfide
    0.43
    Bracelet
    0.41
    hetics
    0.40
     ayudar
    0.40
    铯
    0.39
    参考文献
    0.39
    炼
    0.39
     profund
    0.39
    POSITIVE LOGITS
     menacing
    0.91
     threats
    0.89
     threat
    0.89
     threatening
    0.84
     menace
    0.80
     amea
    0.80
    威胁
    0.79
     attacks
    0.77
    threat
    0.77
     enemy
    0.75
    Act Density 0.441%

    No Known Activations