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

    Neuronpedia

    INDEX
    Explanations

    phrases related to specific events, states, or locations

    oai_token-act-pair · gpt-3.5-turbo

    sentences that conclude with a period

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     elbow
    -0.68
    ling
    -0.65
     monop
    -0.65
     neglig
    -0.65
     nodd
    -0.65
    lings
    -0.64
     footing
    -0.64
    raph
    -0.63
     coy
    -0.63
     optional
    -0.63
    POSITIVE LOGITS
     Their
    1.00
     Hopefully
    0.99
     Specifically
    0.98
     Unfortunately
    0.96
     Thankfully
    0.96
     Sadly
    0.94
     Whereas
    0.92
     Along
    0.92
     Fortunately
    0.91
     Luckily
    0.90
    Act Density 0.930%

    No Known Activations