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

    Neuronpedia

    INDEX
    Explanations

    describing states or qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     cuyo
    0.42
    多くの
    0.42
     größeren
    0.40
     größere
    0.40
     cuya
    0.39
     whose
    0.38
     kleinere
    0.37
    大きな
    0.37
     conceptually
    0.37
     cuja
    0.36
    POSITIVE LOGITS
     intact
    0.73
     impeccable
    0.61
     immaculate
    0.60
     minimal
    0.52
     nonexistent
    0.52
     bleak
    0.50
     reversed
    0.48
     erratic
    0.47
     strained
    0.47
     uncertain
    0.46
    Act Density 0.143%

    No Known Activations