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

    Neuronpedia

    INDEX
    Explanations

    the term "deep" and its variants, indicating emphasis on intensity or significance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pic
    -0.19
    lauf
    -0.16
    oline
    -0.15
    owards
    -0.15
    agn
    -0.15
    ãĥĨãĥ«
    -0.15
    761
    -0.15
    oice
    -0.15
    essor
    -0.15
    æij©
    -0.14
    POSITIVE LOGITS
     deep
    0.31
    deep
    0.29
    ening
    0.29
     deepest
    0.26
    ened
    0.26
    Deep
    0.24
     depths
    0.24
     Deep
    0.24
    æ·±
    0.24
     deeply
    0.23
    Act Density 0.028%

    No Known Activations