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

    Neuronpedia

    INDEX
    Explanations

    capitals letters 'KA'

    oai_token-act-pair · gpt-3.5-turbo

    references to military designations or areas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enegger
    -0.80
    lasses
    -0.74
    vous
    -0.68
    ayn
    -0.67
     ease
    -0.66
    yi
    -0.65
    tie
    -0.64
    offer
    -0.62
    hof
    -0.62
    cards
    -0.62
    POSITIVE LOGITS
    BILITY
    1.42
    BILITIES
    1.22
    VE
    1.11
    UTH
    1.05
    qua
    1.04
    ircraft
    1.04
    verages
    0.99
    BIL
    0.97
    perture
    0.96
    ħĭ
    0.93
    Act Density 0.046%

    No Known Activations