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

    Neuronpedia

    INDEX
    Explanations

    hazards, dangers, and problems

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    aust
    0.45
    elan
    0.42
     افرادی
    0.41
    util
    0.41
    大幅
    0.40
     verwenden
    0.40
    abe
    0.39
    ESS
    0.39
    ignan
    0.39
    㓩
    0.39
    POSITIVE LOGITS
     hazards
    1.29
     dangers
    1.23
     threats
    1.23
     perils
    1.17
     Threats
    1.11
     Hazards
    1.08
     problems
    1.07
     проблемы
    0.99
    problems
    0.97
     evils
    0.96
    Act Density 0.021%

    No Known Activations