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

    Neuronpedia

    INDEX
    Explanations

    improvement requirement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    idot
    -0.09
    xon
    -0.09
    /IP
    -0.09
    orama
    -0.09
    ìĨ¡
    -0.08
    spm
    -0.08
    linger
    -0.08
    naires
    -0.08
    cape
    -0.08
    assy
    -0.08
    POSITIVE LOGITS
    ments
    0.18
    ment
    0.17
     upon
    0.15
    /en
    0.12
    put
    0.12
    upon
    0.11
     Upon
    0.11
     Pend
    0.11
    ance
    0.11
    Upon
    0.10
    Act Density 0.032%

    No Known Activations