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

    Neuronpedia

    INDEX
    Explanations

    parts of code surrounded by specific characters

    oai_token-act-pair · gpt-3.5-turbo

    instances of backticks or grave accents (`` ` ``)

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    phrine
    -0.76
     mills
    -0.74
     Mamm
    -0.67
     condem
    -0.64
     delinqu
    -0.64
     Jenner
    -0.64
     Glou
    -0.63
     therap
    -0.62
     dividing
    -0.62
     elim
    -0.61
    POSITIVE LOGITS
    ansas
    0.87
    taboola
    0.86
    bah
    0.81
    seq
    0.81
    daq
    0.80
    lein
    0.78
    rosis
    0.77
    pler
    0.76
    Vi
    0.75
    ns
    0.74
    Act Density 0.017%

    No Known Activations