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

    Neuronpedia

    INDEX
    Explanations

    phrases that indicate an action or procedural instruction

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

    technical "how to" actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interactiva
    -0.41
    czego
    -0.40
     wiederholt
    -0.40
     prawda
    -0.38
     gekenn
    -0.38
     humanas
    -0.38
     Eindruck
    -0.37
     mogę
    -0.36
     humano
    -0.36
     twar
    -0.36
    POSITIVE LOGITS
    GOTREF
    0.50
     StringTokenizer
    0.50
    createClass
    0.47
    RectangleBorder
    0.46
    stok
    0.46
     تضيفلها
    0.46
     Вікіпе
    0.43
    verifyException
    0.43
    illez
    0.43
    makeConstraints
    0.43
    Act Density 0.228%

    No Known Activations