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

    Neuronpedia

    INDEX
    Explanations

    specific capitalized sequences of characters

    oai_token-act-pair · gpt-3.5-turbo

    common phrases or expressions in writing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nesday
    -0.87
     consecut
    -0.84
     opting
    -0.81
     favour
    -0.78
     cryptoc
    -0.74
     steering
    -0.74
     transitioning
    -0.73
     princ
    -0.73
     namely
    -0.72
     shielding
    -0.71
    POSITIVE LOGITS
    Released
    0.96
    Constructed
    0.96
    ³³³³³³³³
    0.94
    ³³³³³³³³³³³³³³³³
    0.87
    ³³³³
    0.86
    Okay
    0.82
    ccording
    0.82
    ³³³
    0.79
    Ear
    0.79
    Rail
    0.79
    Act Density 0.228%

    No Known Activations