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

    Neuronpedia

    INDEX
    Explanations

    meeting requirements or suitability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    0.54
     are
    0.54
    ua
    0.52
    q
    0.51
    ui
    0.44
    si
    0.44
    ita
    0.44
    ari
    0.43
     l
    0.41
    ri
    0.41
    POSITIVE LOGITS
    For
    0.45
    ↵
    0.44
    ↵↵
    0.41
    あまり
    0.41
    볍
    0.41
     desempen
    0.40
    短
    0.40
    ल्हा
    0.39
     detn
    0.38
    如此
    0.38
    Act Density 0.626%

    No Known Activations