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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating multiple ways or reasons for something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that suggest methods or reasons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    livion
    -0.82
     Tsukuyomi
    -0.75
    ··
    -0.73
    ĸļ
    -0.70
    ãģĻ
    -0.64
    Phones
    -0.62
    Mos
    -0.60
    UME
    -0.60
    onomic
    -0.59
    iture
    -0.59
    POSITIVE LOGITS
     pecul
    0.75
     recurring
    0.73
     inspir
    0.72
     notable
    0.71
     downside
    0.68
     difference
    0.66
     Redd
    0.65
     highlights
    0.64
     noteworthy
    0.64
     redeem
    0.64
    Act Density 0.148%

    No Known Activations