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

    Neuronpedia

    INDEX
    Explanations

    references to the concept of importance in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arken
    -0.16
    ucas
    -0.16
    uci
    -0.15
    iska
    -0.15
    opping
    -0.15
    ucci
    -0.15
    .DEFAULT
    -0.14
     cul
    -0.14
    add
    -0.14
    aho
    -0.14
    POSITIVE LOGITS
     importance
    0.30
    /import
    0.28
     Importance
    0.25
     Attached
    0.24
     attached
    0.22
    Attached
    0.22
    attached
    0.21
     significance
    0.20
     placed
    0.20
     role
    0.19
    Act Density 0.013%

    No Known Activations