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

    Neuronpedia

    INDEX
    Explanations

    numerical values and specific formatted data structures

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

    code or math expressions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    transQ
    -0.52
     betweenstory
    -0.47
     ✭✭
    -0.47
    ungszeit
    -0.47
    MLLoader
    -0.47
    fromnode
    -0.46
    [@BOS@]
    -0.46
    <unused14>
    -0.45
    kegaard
    -0.45
    <unused23>
    -0.45
    POSITIVE LOGITS
    Personensuche
    0.61
    Referanser
    0.48
    Hentet
    0.47
    GOTREF
    0.47
    farwyddwr
    0.44
    principalColumn
    0.42
    DoubleQuotes
    0.41
    Manbalar
    0.41
    tagHelperRunner
    0.41
    ьаж
    0.40
    Act Density 0.000%

    No Known Activations