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

    Neuronpedia

    INDEX
    Explanations

    phrases related to contrasting concepts

    oai_token-act-pair · gpt-3.5-turbo

    contrasting ideas or dualities in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abase
    -0.78
    ="#
    -0.75
    enhagen
    -0.74
    asma
    -0.74
    ienne
    -0.74
    obook
    -0.74
    wordpress
    -0.73
    ograp
    -0.70
    enary
    -0.67
    ixir
    -0.66
    POSITIVE LOGITS
     respectively
    2.27
     alike
    1.48
    depending
    1.32
     depending
    1.09
     whichever
    0.95
     together
    0.87
     totaling
    0.85
     etc
    0.81
     moreover
    0.80
    together
    0.78
    Act Density 0.428%

    No Known Activations