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

    Neuronpedia

    INDEX
    Explanations

    code-related syntax patterns, particularly involving "->"

    oai_token-act-pair · gpt-3.5-turbo

    code-related symbols and structures, particularly arrows and operators commonly used in programming

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Dud
    -0.78
     Guinness
    -0.72
     Fairfax
    -0.71
     Gul
    -0.70
     swear
    -0.64
    Ĥ¬
    -0.64
     hoax
    -0.61
     Nasa
    -0.61
     Oath
    -0.60
     tru
    -0.60
    POSITIVE LOGITS
    ->
    3.97
     ->
    2.23
    âĨĴ
    1.79
     âĨĴ
    1.72
    =>
    1.46
    -->
    1.44
    ::
    1.29
    ++)
    1.24
    >[
    1.21
    >>
    1.21
    Act Density 0.021%

    No Known Activations