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

    Neuronpedia

    INDEX
    Explanations

    highly relevant content or sections in a document indicating key contributions and important findings

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

    foreign language words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ValueStyle
    -0.75
     defaultstate
    -0.70
    <unused14>
    -0.68
    <unused47>
    -0.68
    <unused28>
    -0.68
    <unused41>
    -0.67
    <unused74>
    -0.67
    [@BOS@]
    -0.67
    <unused79>
    -0.67
    <unused8>
    -0.67
    POSITIVE LOGITS
     acceptez
    0.44
     igång
    0.38
     malades
    0.37
     všetkých
    0.36
     limba
    0.35
     tuturor
    0.35
     godk
    0.33
     certificación
    0.32
     ļ
    0.32
     dinámico
    0.31
    Act Density 0.001%

    No Known Activations