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

    Neuronpedia

    INDEX
    Explanations

    dangerous, weapons, death, or financial contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     synced
    0.39
    bmod
    0.39
    optimal
    0.39
     incompar
    0.38
    max
    0.38
     devoid
    0.38
    0.37
    新增
    0.37
     habitually
    0.37
    が存在
    0.36
    POSITIVE LOGITS
     The
    0.54
     Women
    0.49
     Medical
    0.47
     National
    0.46
     Rainbow
    0.46
     Financial
    0.45
     A
    0.44
     Justice
    0.44
     the
    0.43
     Prince
    0.43
    Act Density 0.000%

    No Known Activations