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

    Neuronpedia

    INDEX
    Explanations

    references to entities and their relationships within a discussion

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

    the + noun/entity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    owskiego
    -0.28
    륨
    -0.27
     dibu
    -0.26
     externi
    -0.25
     guapos
    -0.24
    且
    -0.24
     plantes
    -0.23
     typelib
    -0.23
     Económica
    -0.23
     actualidad
    -0.23
    POSITIVE LOGITS
    <unused14>
    0.82
    [@BOS@]
    0.82
    <unused74>
    0.82
    DoubleQuotes
    0.82
    <unused28>
    0.82
    <unused68>
    0.82
    <unused41>
    0.82
    <unused42>
    0.82
    <unused16>
    0.82
    <unused8>
    0.82
    Act Density 0.129%

    No Known Activations