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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "the" and variations of it as well as possessive pronouns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     således
    -0.78
     précie
    -0.78
     keenly
    -0.74
    ="#"><
    -0.73
     increasingly
    -0.72
     soigneusement
    -0.72
     cuidadosamente
    -0.71
     ferdig
    -0.70
     nunmehr
    -0.70
    niksi
    -0.69
    POSITIVE LOGITS
     guy
    1.07
     thing
    0.85
     kids
    0.82
     guys
    0.82
     damn
    0.82
     whole
    0.80
     fucking
    0.80
     WHOLE
    0.77
     stupid
    0.77
     funniest
    0.73
    Act Density 0.501%

    No Known Activations