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

    Neuronpedia

    INDEX
    Explanations

    anything not specifically mentioned elsewhere in the text

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express inclusion or consideration of various alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Runner
    -0.71
    haw
    -0.67
    gers
    -0.66
     Roose
    -0.66
    Upload
    -0.65
     Derby
    -0.64
    hai
    -0.62
    gets
    -0.61
    oku
    -0.59
    past
    -0.59
    POSITIVE LOGITS
    worldly
    1.07
     imaginable
    0.93
     besides
    0.89
     includ
    0.77
     describ
    0.75
     mattered
    0.74
     happens
    0.69
    nces
    0.68
     happened
    0.68
     afforded
    0.67
    Act Density 0.017%

    No Known Activations