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

    Neuronpedia

    INDEX
    Explanations

    negative contractions used to express denial or rejection

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Weiss
    -0.82
    adpleegd
    -0.75
    setSource
    -0.75
     merce
    -0.75
    तः
    -0.73
     Gibbs
    -0.71
    er
    -0.69
    Weiss
    -0.68
    ेष
    -0.68
    Té
    -0.68
    POSITIVE LOGITS
     isn
    1.44
     wasn
    1.37
     Wasn
    1.34
     weren
    1.33
     aren
    1.29
     shouldn
    1.28
     Isn
    1.27
     hasn
    1.26
     Shouldn
    1.25
     didn
    1.23
    Act Density 0.078%

    No Known Activations