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

    Neuronpedia

    INDEX
    Explanations

    phrases indicative of personal relationships and domestic life

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atile
    -0.16
    prung
    -0.15
    elijk
    -0.14
    elijke
    -0.14
    ellij
    -0.14
    ucu
    -0.14
    comed
    -0.14
    arez
    -0.13
    atoon
    -0.13
    ezi
    -0.13
    POSITIVE LOGITS
     sometimes
    0.21
    sometimes
    0.21
    variably
    0.19
    often
    0.17
     vždy
    0.17
    ometimes
    0.17
     Sometimes
    0.16
    Optimizer
    0.16
     often
    0.16
     invariably
    0.16
    Act Density 0.096%

    No Known Activations