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

    Neuronpedia

    INDEX
    Explanations

    instances of quotation marks or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    <bos>
    -1.01
    Jeografia
    -0.98
    хьтан
    -0.95
    Portale
    -0.94
     betweenstory
    -0.91
    Datuak
    -0.90
     Catawiki
    -0.88
    enterOuterAlt
    -0.88
    ########.
    -0.87
    SharedCtor
    -0.85
    POSITIVE LOGITS
    [
    0.82
     [
    0.69
     wasn
    0.67
    <eos>
    0.63
     isn
    0.60
     don
    0.60
     aren
    0.59
    mathrm
    0.57
     didn
    0.57
     wouldn
    0.55
    Act Density 0.013%

    No Known Activations