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

    Neuronpedia

    INDEX
    Explanations

    phrases beginning with the word "So" as a transition or continuation

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "So", indicating a conversational or explanatory tone

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    intosh
    -0.64
     Mehran
    -0.62
     neighb
    -0.58
    Winged
    -0.58
    /
    -0.57
    ASED
    -0.57
     Unleashed
    -0.55
    uc
    -0.55
     Featured
    -0.54
     ])
    -0.53
    POSITIVE LOGITS
    oner
    1.50
     far
    1.15
     yeah
    1.13
     unless
    1.01
    fter
    0.98
     if
    0.97
     basically
    0.95
     yes
    0.93
     whereas
    0.91
     why
    0.90
    Act Density 0.067%

    No Known Activations