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

    Neuronpedia

    INDEX
    Explanations

    old words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ле
    0.97
    ın
    0.85
    يد
    0.82
     nossos
    0.80
    on
    0.78
     humains
    0.77
     importe
    0.75
     enzym
    0.75
    u
    0.75
    ের
    0.73
    POSITIVE LOGITS
    time
    0.70
    td
    0.68
    가
    0.66
    I
    0.64
    town
    0.61
    thesis
    0.61
    tag
    0.61
    ta
    0.59
    し
    0.59
    sum
    0.59
    Act Density 0.012%

    No Known Activations