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

    Neuronpedia

    INDEX
    Explanations

    describing physical features or qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ikut
    0.40
    </u>
    0.40
    memcpy
    0.39
    om
    0.39
     очеред
    0.38
     imate
    0.38
    ходе
    0.38
    udas
    0.37
    =")
    0.37
     થી
    0.37
    POSITIVE LOGITS
    独特
    0.61
     distinctive
    0.61
     durability
    0.57
     sufficient
    0.56
     slits
    0.56
     slight
    0.55
     holes
    0.55
     grooves
    0.54
     rounded
    0.54
     excellent
    0.53
    Act Density 0.023%

    No Known Activations