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

    Neuronpedia

    INDEX
    Explanations

    phrases or references related to authorship or sources in written text

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

    the word "from"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    IContainer
    -0.59
     errorCode
    -0.56
     ſche
    -0.55
     ſol
    -0.52
     pleaſure
    -0.51
     resell
    -0.50
     eventId
    -0.49
    pell
    -0.48
     taskId
    -0.48
    analy
    -0.48
    POSITIVE LOGITS
    From
    1.56
     From
    1.48
    FROM
    1.42
    from
    1.34
     FROM
    1.32
     from
    1.19
    Från
    1.00
    getFrom
    0.95
     från
    0.93
    setFrom
    0.90
    Act Density 0.043%

    No Known Activations