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

    Neuronpedia

    INDEX
    Explanations

    conjunctions indicating contrasting information or ideas

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "But" indicating contrasting information or points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãģ®
    -0.71
    kie
    -0.66
    heads
    -0.65
    built
    -0.65
    roy
    -0.64
    UD
    -0.62
     segment
    -0.60
    broken
    -0.59
    million
    -0.58
    adr
    -0.58
    POSITIVE LOGITS
    tons
    1.36
     alas
    1.11
     beware
    0.95
     nevertheless
    0.87
    chery
    0.85
     hey
    0.83
     luckily
    0.82
     fortunately
    0.81
    chers
    0.80
    tery
    0.79
    Act Density 0.104%

    No Known Activations