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

    Neuronpedia

    INDEX
    Explanations

    less common or desirable

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     неболь
    0.43
    合わせ
    0.40
    াইল
    0.39
    atch
    0.37
     Benefits
    0.37
    aliana
    0.37
     Increased
    0.36
    too
    0.35
     जरा
    0.35
     lots
    0.33
    POSITIVE LOGITS
     than
    0.76
    ens
    0.69
    ening
    0.66
     ніж
    0.66
     desirable
    0.61
    ENING
    0.59
    ened
    0.58
    Than
    0.58
     niż
    0.57
     glamorous
    0.57
    Act Density 0.039%

    No Known Activations