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

    Neuronpedia

    INDEX
    Explanations

    instances of conditional phrases and their implications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    evi
    -0.15
    елем
    -0.14
    063
    -0.14
    assel
    -0.14
    angan
    -0.13
    ijing
    -0.13
    858
    -0.13
    ermal
    -0.13
    atik
    -0.13
    elpers
    -0.12
    POSITIVE LOGITS
     anyway
    1.09
    Anyway
    0.99
     Anyway
    0.97
     anyways
    0.93
     anyhow
    0.78
     nonetheless
    0.54
     nevertheless
    0.51
     toch
    0.47
    Nevertheless
    0.47
     Nonetheless
    0.46
    Act Density 1.155%

    No Known Activations