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

    Neuronpedia

    INDEX
    Explanations

    pose a threat/risk/violation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    黴
    0.62
     ዝ
    0.58
    高峰
    0.57
     주의
    0.56
     cautious
    0.56
     ow
    0.56
    افظ
    0.56
    ah
    0.55
    ాలంటే
    0.55
     Junction
    0.54
    POSITIVE LOGITS
     teeth
    0.80
     existential
    0.69
     tooth
    0.69
     Trojan
    0.68
    ɬ
    0.67
     Transform
    0.67
     Teeth
    0.66
     attack
    0.66
     dientes
    0.66
     camada
    0.65
    Act Density 0.147%

    No Known Activations