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

    Neuronpedia

    INDEX
    Explanations

    Likely causing this

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    asih
    0.83
    ară
    0.77
    باش
    0.76
    mış
    0.76
     सभी
    0.76
     aucune
    0.75
    やはり
    0.75
     সবগুলো
    0.74
    ногие
    0.72
     trochę
    0.71
    POSITIVE LOGITS
     happening
    1.83
     happen
    1.65
     happened
    1.56
     happens
    1.46
     acontece
    1.31
     aconteceu
    1.30
     Happens
    1.27
     done
    1.26
     aconte
    1.26
     Happened
    1.23
    Act Density 0.656%

    No Known Activations