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

    Neuronpedia

    INDEX
    Explanations

    references to rot or rotting

    oai_token-act-pair · gpt-3.5-turbo

    references to decay or deterioration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nect
    -0.77
    inez
    -0.68
    ynthesis
    -0.68
    gdala
    -0.67
    dress
    -0.66
     EntityItem
    -0.65
    racuse
    -0.63
    ledged
    -0.63
     Annotations
    -0.62
    ħĭ
    -0.62
    POSITIVE LOGITS
    ations
    1.08
    ational
    1.04
    unda
    0.91
    ional
    0.88
    Rot
    0.87
     rot
    0.84
    oscope
    0.80
    atory
    0.76
    uten
    0.76
    ular
    0.76
    Act Density 0.008%

    No Known Activations