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

    Neuronpedia

    INDEX
    Explanations

    the word "try" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "try to" in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    dylib
    -0.71
    ¥µ
    -0.62
    obyl
    -0.60
    rome
    -0.58
    isible
    -0.56
     Tribune
    -0.56
    otle
    -0.56
    iasis
    -0.55
    azi
    -0.55
    $$$$
    -0.54
    POSITIVE LOGITS
     unsuccessfully
    0.98
     harder
    0.88
    anke
    0.79
     desperately
    0.73
    sts
    0.69
     again
    0.68
     anew
    0.67
     hard
    0.66
     levers
    0.66
    okes
    0.66
    Act Density 0.061%

    No Known Activations