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

    Neuronpedia

    INDEX
    Explanations

    determiner phrases that indicate focus or emphasis

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

    a focus on

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ProtoMessage
    -0.41
     asamblea
    -0.40
     desmotivaciones
    -0.39
     impresiones
    -0.37
    InjectAttribute
    -0.37
    Hentet
    -0.37
     createContext
    -0.36
     ciudadana
    -0.35
     escenas
    -0.35
     vêtement
    -0.35
    POSITIVE LOGITS
     focus
    0.64
     emphasis
    0.62
     focused
    0.57
     flair
    0.56
     mission
    0.56
    focus
    0.52
     Emphasis
    0.52
     Focused
    0.51
    Emphasis
    0.51
    focused
    0.50
    Act Density 0.045%

    No Known Activations