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

    Neuronpedia

    INDEX
    Explanations

    the word "that" followed by statements or clauses

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    izont
    -0.89
     pione
    -0.82
    aukee
    -0.81
    ãĤ´ãĥ³
    -0.75
    entary
    -0.73
    ocre
    -0.73
    istor
    -0.72
    ocaly
    -0.71
     exting
    -0.71
    yna
    -0.70
    POSITIVE LOGITS
     consequently
    1.14
     they
    0.98
     therefore
    0.98
     furthermore
    0.94
     there
    0.89
     moreover
    0.86
     it
    0.85
    soever
    0.84
     includes
    0.82
    's
    0.82
    Act Density 0.114%

    No Known Activations