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

    Neuronpedia

    INDEX
    Explanations

    references to looking back or reflecting on past events or memories

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     viz
    -0.87
     Hague
    -0.78
    ises
    -0.72
    icular
    -0.71
    ijn
    -0.70
     Ukrain
    -0.68
    entric
    -0.68
     Penguins
    -0.67
     pollut
    -0.67
    BLIC
    -0.67
    POSITIVE LOGITS
     nostalg
    1.32
    wards
    1.15
    dated
    1.09
    packs
    1.03
    gam
    1.00
    stab
    0.99
    trace
    0.99
    glass
    0.92
    WARD
    0.90
     fond
    0.90
    Act Density 5.510%

    No Known Activations