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

    Neuronpedia

    INDEX
    Explanations

    units of measurement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Parsing
    0.80
     改
    0.79
    ៥
    0.75
     で
    0.73
     forKey
    0.73
     Including
    0.72
     Seventy
    0.72
    ;
    0.72
    七
    0.72
     katika
    0.72
    POSITIVE LOGITS
    -
    0.70
     rule
    0.66
     stronghold
    0.66
     placeholder
    0.63
    짜
    0.62
     shipment
    0.60
     unit
    0.60
     loophole
    0.59
    ‑
    0.59
    -{\
    0.58
    Act Density 0.062%

    No Known Activations