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

    Neuronpedia

    INDEX
    Explanations

    superlative forms of adjectives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ujednoznacz
    -0.59
    agie
    -0.56
    o
    -0.53
     Go
    -0.53
    /
    -0.51
     o
    -0.51
    <eos>
    -0.50
     a
    -0.50
    q
    -0.49
    toHave
    -0.49
    POSITIVE LOGITS
     healthiest
    2.05
     heaviest
    2.04
     strongest
    2.04
     lightest
    1.96
    iest
    1.96
     hardest
    1.92
     loudest
    1.92
     warmest
    1.91
     slowest
    1.89
     safest
    1.89
    Act Density 0.231%

    No Known Activations