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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating an investigation or deeper exploration

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "look into."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĺħ
    -0.73
    warning
    -0.70
    bender
    -0.69
    end
    -0.68
    enegger
    -0.68
    patrick
    -0.66
    sticks
    -0.65
    quez
    -0.65
    sworth
    -0.64
    stick
    -0.64
    POSITIVE LOGITS
     infinity
    0.80
     adulthood
    0.74
    tnc
    0.62
    dfx
    0.61
    Winged
    0.60
     Ved
    0.60
     eternity
    0.59
    Medic
    0.59
     feasibility
    0.58
     hus
    0.57
    Act Density 0.067%

    No Known Activations