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

    Neuronpedia

    INDEX
    Explanations

    numeric values and symbols associated with statistical data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    number-number ranges or identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    VersionUID
    -0.32
     tonel
    -0.32
     tunn
    -0.31
     solas
    -0.31
     following
    -0.31
     통해
    -0.30
     hadas
    -0.30
     keres
    -0.30
     fita
    -0.30
     génér
    -0.29
    POSITIVE LOGITS
    WireFormatLite
    0.65
    Datuak
    0.64
    [@BOS@]
    0.62
    <unused3>
    0.62
    <unused74>
    0.62
    <unused79>
    0.62
    <unused51>
    0.62
    <unused41>
    0.62
    <unused14>
    0.62
    <unused28>
    0.62
    Act Density 0.248%

    No Known Activations