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

    Neuronpedia

    INDEX
    Explanations

    discussions related to specific domain expertise or technical content with a focus on syntax and technical terms

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    utterstock
    -1.06
    anship
    -0.99
    ometimes
    -0.95
    ulhu
    -0.92
    illas
    -0.92
    ĸļ
    -0.91
    EStream
    -0.87
    afety
    -0.83
    wright
    -0.82
    chwitz
    -0.80
    POSITIVE LOGITS
    sounding
    0.96
     ones
    0.88
    hearted
    0.82
    blooded
    0.79
     versions
    0.78
     alternative
    0.78
    enough
    0.78
     ways
    0.77
     manner
    0.77
    ly
    0.73
    Act Density 17.764%

    No Known Activations