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

    Neuronpedia

    INDEX
    Explanations

    authors after "by"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,
    0.72
    0.69
    -
    0.55
     H
    0.52
    .
    0.50
    '
    0.50
     F
    0.49
    ...
    0.48
     in
    0.48
     Man
    0.48
    POSITIVE LOGITS
    <unused646>
    0.59
    <unused1142>
    0.58
    <unused289>
    0.57
    <unused579>
    0.56
    <unused169>
    0.55
     þat
    0.55
    <unused581>
    0.55
    <unused619>
    0.55
    <unused477>
    0.55
    <unused481>
    0.54
    Act Density 0.000%

    No Known Activations