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

    Neuronpedia

    INDEX
    Explanations

    instances of phrases involving interconnected actions

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "and" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    owered
    -0.80
    floor
    -0.75
    uction
    -0.73
    actionDate
    -0.73
    uta
    -0.71
    UCT
    -0.69
    STEM
    -0.69
    ince
    -0.69
    hoe
    -0.68
    hov
    -0.67
    POSITIVE LOGITS
     therefore
    1.04
     thus
    1.03
     deserve
    1.01
     hence
    0.94
     consequently
    0.94
     thereby
    0.93
     hopefully
    0.93
     deserves
    0.92
     furthermore
    0.85
     secondly
    0.84
    Act Density 0.288%

    No Known Activations