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

    Neuronpedia

    INDEX
    Explanations

    concepts related to darkness and negative emotions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    estone
    -0.86
    odium
    -0.82
    kus
    -0.82
    amins
    -0.82
    ergy
    -0.77
    estones
    -0.76
    nesium
    -0.76
    yton
    -0.75
    uscript
    -0.72
    ãĥ£
    -0.71
    POSITIVE LOGITS
     arises
    1.01
     plag
    0.99
     caused
    0.94
     arising
    0.93
     arose
    0.93
     stemming
    0.90
     inflicted
    0.85
     perv
    0.85
     abound
    0.84
     plagued
    0.83
    Act Density 0.192%

    No Known Activations