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

    Neuronpedia

    INDEX
    Explanations

    location and state indicators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    煂
    0.46
    лювання
    0.46
    mitting
    0.45
    <unused2115>
    0.45
     용
    0.45
     losses
    0.44
    ٰی
    0.43
    rolyte
    0.43
     त्य
    0.42
     ballon
    0.42
    POSITIVE LOGITS
    oc
    0.55
    CK
    0.51
    ar
    0.51
    ோ
    0.51
    AR
    0.49
    AG
    0.47
    an
    0.47
    ad
    0.47
    os
    0.47
    HA
    0.47
    Act Density 0.001%

    No Known Activations