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

    Neuronpedia

    INDEX
    Explanations

    RGB values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     resulting
    0.45
     achieves
    0.44
    thi
    0.42
     outfile
    0.39
    ɪ
    0.38
     activated
    0.38
     simply
    0.38
     increases
    0.38
     offset
    0.38
     achieved
    0.38
    POSITIVE LOGITS
    plum
    0.43
    樯
    0.42
     Besuch
    0.41
     Dixie
    0.40
     Bobcats
    0.39
    墉
    0.39
     Ducks
    0.38
     expropri
    0.38
    hooter
    0.38
     Sisters
    0.37
    Act Density 0.003%

    No Known Activations