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

    Neuronpedia

    INDEX
    Explanations

    instances where the word "sat" occurs

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sat"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    peat
    -0.92
    ascript
    -0.88
    itialized
    -0.80
    alog
    -0.75
    amac
    -0.73
    dylib
    -0.72
    assets
    -0.71
     "$:/
    -0.71
    acters
    -0.71
    Diamond
    -0.70
    POSITIVE LOGITS
     exting
    0.72
     proble
    0.68
    bane
    0.68
     destro
    0.66
     recre
    0.65
    hens
    0.63
    nv
    0.63
    kat
    0.60
     bru
    0.60
     Kre
    0.59
    Act Density 0.000%

    No Known Activations