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

    Neuronpedia

    INDEX
    Explanations

    expressions of honesty and candidness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     kapturem
    -0.51
     miniaturka
    -0.50
     deseo
    -0.50
     recours
    -0.49
     outcomes
    -0.48
     Sünde
    -0.47
     Vergnügen
    -0.47
     dragón
    -0.47
     Gnade
    -0.46
    izante
    -0.45
    POSITIVE LOGITS
     probably
    0.89
     really
    0.78
     barely
    0.77
     honestly
    0.74
     prolly
    0.71
     probable
    0.70
    probably
    0.70
     Probably
    0.68
     Really
    0.67
    Probably
    0.66
    Act Density 0.292%

    No Known Activations