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

    Neuronpedia

    INDEX
    Explanations

    class attribute

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tacos
    0.66
    ™.
    0.64
     Aztecs
    0.63
     ojos
    0.63
     thursday
    0.62
    హ్
    0.62
     ayahuasca
    0.62
     curvy
    0.61
    𝘺
    0.60
    йте
    0.60
    POSITIVE LOGITS
    from
    0.86
    name
    0.85
    for
    0.78
    da
    0.77
    1
    0.76
    map
    0.74
    e
    0.74
    title
    0.71
    ة
    0.71
    c
    0.70
    Act Density 0.001%

    No Known Activations