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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing a sense of being significant or exceptional, often in comparison to others

    oai_token-act-pair · gpt-3.5-turbo

    phrases that imply a minimal or insignificant quantity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     pause
    -0.71
    aline
    -0.69
    inet
    -0.61
    oline
    -0.61
    erva
    -0.59
    alin
    -0.59
    activated
    -0.58
    ynthesis
    -0.58
    aturation
    -0.57
    ync
    -0.57
    POSITIVE LOGITS
    MpServer
    0.92
     Carbuncle
    0.71
    astically
    0.70
    èĪ
    0.68
    ubes
    0.67
    geoning
    0.65
     feat
    0.65
    âĶĢ
    0.64
    astic
    0.63
     stylish
    0.63
    Act Density 0.100%

    No Known Activations