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

    Neuronpedia

    INDEX
    Explanations

    expressions of personal preferences and favorites

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

    expressing strong preference

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ſammen
    -0.80
    +#+#
    -0.79
     propOrder
    -0.77
     Numerade
    -0.74
    IntoConstraints
    -0.74
     queſta
    -0.72
     imagui
    -0.72
    <unused68>
    -0.71
    <unused14>
    -0.71
    <unused28>
    -0.71
    POSITIVE LOGITS
    mergeFrom
    0.36
     gustan
    0.34
     paixão
    0.34
     particularly
    0.34
     Especially
    0.33
     love
    0.33
     especially
    0.32
     Particularly
    0.32
    Especially
    0.30
     encantan
    0.30
    Act Density 0.021%

    No Known Activations