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

    Neuronpedia

    INDEX
    Explanations

    phrases contrasting two different ideas or sides of an argument

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     للمعارف
    -0.66
     OFDb
    -0.61
    principalColumn
    -0.59
    ésultats
    -0.57
    iffance
    -0.56
    GEBURTSDATUM
    -0.55
     snippetHide
    -0.55
     Administrativna
    -0.54
     pleaſure
    -0.53
    iempos
    -0.52
    POSITIVE LOGITS
     whereas
    0.70
    whereas
    0.69
     dagegen
    0.68
     hingegen
    0.64
     أما
    0.63
    omiast
    0.63
    أما
    0.63
    Whereas
    0.60
     Whereas
    0.59
     while
    0.55
    Act Density 0.197%

    No Known Activations