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

    Neuronpedia

    INDEX
    Explanations

    phrases or statements that mention quotations or dialogue

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

    A single quote at the beginning of abstracts

    np_acts-logits-general · gemini-2.0-flash

    academic paper abstracts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    anan
    -0.67
     Harbor
    -0.65
    Harbor
    -0.65
     Vapor
    -0.64
    chili
    -0.63
    aura
    -0.63
    __*/
    -0.62
    IntoConstraints
    -0.62
     Hartman
    -0.62
     Donahue
    -0.61
    POSITIVE LOGITS
     Meksika
    0.55
     huelga
    0.48
     Turquía
    0.48
    emplares
    0.48
     británico
    0.47
     ''){
    0.46
     médicale
    0.46
    ómetros
    0.46
    言えば
    0.45
    sweise
    0.45
    Act Density 0.230%

    No Known Activations