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

    Neuronpedia

    INDEX
    Explanations

    hazardous situations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rel
    0.39
     ug
    0.37
     Consecutive
    0.37
    种种
    0.36
     sector
    0.35
    either
    0.34
    etrot
    0.34
     Either
    0.34
    افه
    0.34
     Affair
    0.34
    POSITIVE LOGITS
     flammable
    0.98
     sharp
    0.94
     liquids
    0.93
     hazards
    0.84
     hazardous
    0.82
     fragile
    0.79
     loose
    0.77
    sharp
    0.77
     Hazardous
    0.75
     unsecured
    0.75
    Act Density 0.030%

    No Known Activations