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

    Neuronpedia

    INDEX
    Explanations

    phrases with strong or emphatic language

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to sounds or vocal expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arching
    -0.77
     Janeiro
    -0.73
     Kart
    -0.71
    elsen
    -0.69
    ournal
    -0.67
    ffic
    -0.67
    rest
    -0.66
     Submission
    -0.66
    jac
    -0.61
    eve
    -0.61
    POSITIVE LOGITS
     sounding
    1.08
     alarms
    0.94
    nces
    0.92
     louder
    0.91
     nodd
    0.91
    sounding
    0.84
     voices
    0.84
    rums
    0.81
     sounded
    0.81
     noises
    0.79
    Act Density 0.015%

    No Known Activations