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

    Neuronpedia

    INDEX
    Explanations

    exclamatory expressions

    oai_token-act-pair · gpt-3.5-turbo

    exclamatory statements or expressions of enthusiasm

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     manif
    -0.84
     destro
    -0.81
     tyr
    -0.77
     glim
    -0.76
     restraints
    -0.74
     surpr
    -0.72
     metic
    -0.70
     arrang
    -0.70
     charact
    -0.69
     dams
    -0.68
    POSITIVE LOGITS
    @#&
    1.38
    #$
    1.15
    @#
    0.93
    ?!
    0.93
    important
    0.91
    ctory
    0.81
    olkien
    0.72
    ?,
    0.71
     Shine
    0.71
    ~
    0.70
    Act Density 0.069%

    No Known Activations