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

    Neuronpedia

    INDEX
    Explanations

    actions related to communication and business engagements

    oai_token-act-pair · gpt-3.5-turbo

    punctuations that indicate breaks or lists in a narrative context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ril
    -0.75
    erity
    -0.73
    ction
    -0.71
    Ì
    -0.66
    raft
    -0.66
    emin
    -0.62
    ï¸
    -0.62
    erm
    -0.61
    tsky
    -0.61
    opal
    -0.61
    POSITIVE LOGITS
     etc
    1.55
    etc
    1.28
     respectively
    0.80
    whatever
    0.75
     Flavoring
    0.75
    FontSize
    0.73
     blah
    0.72
     ect
    0.68
    artifacts
    0.66
     vitamins
    0.64
    Act Density 0.460%

    No Known Activations