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

    Neuronpedia

    INDEX
    Explanations

    phrases related to emphasizing specific information or points

    oai_token-act-pair · gpt-3.5-turbo

    the word "that" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    apolis
    -0.73
    hips
    -0.71
    YS
    -0.71
    emis
    -0.70
    istics
    -0.66
    asures
    -0.66
    tones
    -0.63
    olitan
    -0.62
    yles
    -0.62
    IAS
    -0.60
    POSITIVE LOGITS
     translates
    0.93
     includes
    0.91
     culminated
    0.91
     pesky
    0.86
     entails
    0.82
     resulted
    0.82
     extends
    0.77
     contradicts
    0.77
     culmin
    0.77
     mattered
    0.76
    Act Density 0.106%

    No Known Activations