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

    Neuronpedia

    INDEX
    Explanations

    phrases starting with "first"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.51
    ?
    0.39
    '।
    0.35
     wounds
    0.33
     chuyển
    0.32
    ’
    0.32
    ]
    0.31
     pedibusque
    0.29
    =
    0.29
    يد
    0.29
    POSITIVE LOGITS
    native
    0.41
    b
    0.40
    ink
    0.38
    essa
    0.37
    ick
    0.36
    ن
    0.35
    add
    0.35
    uma
    0.33
    val
    0.32
    text
    0.32
    Act Density 0.332%

    No Known Activations