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

    Neuronpedia

    INDEX
    Explanations

    years

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     l
    0.59
    ↵
    0.55
    us
    0.53
     and
    0.51
     in
    0.49
    5
    0.47
    4
    0.46
    ne
    0.44
     Whit
    0.43
    /
    0.43
    POSITIVE LOGITS
    atorias
    0.49
     безопас
    0.48
    стояние
    0.46
    дә
    0.44
     услуги
    0.44
    edics
    0.42
    ară
    0.42
    َة
    0.42
    ointing
    0.41
    arang
    0.40
    Act Density 0.012%

    No Known Activations