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

    Neuronpedia

    INDEX
    Explanations

    harmful or triggering content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    unitas
    0.88
     asymmetries
    0.86
     heresy
    0.83
     discontinuities
    0.80
    বর্তন
    0.76
    0.75
     discontinuity
    0.74
     Killed
    0.73
     pernicious
    0.72
     deus
    0.72
    POSITIVE LOGITS
     integral
    0.80
     deemed
    0.69
     considered
    0.68
     emotionally
    0.65
     indicative
    0.65
     categorized
    0.62
     capable
    0.62
     greatly
    0.62
     both
    0.62
     apporter
    0.61
    Act Density 0.263%

    No Known Activations