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

    Neuronpedia

    INDEX
    Explanations

    words related to things that are considered challenging or impossible

    oai_token-act-pair · gpt-3.5-turbo

    negative prefixes related to trustworthiness and reliability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anwhile
    -0.85
    姫
    -0.84
    phrine
    -0.77
     chants
    -0.76
    SHIP
    -0.76
    cium
    -0.72
    hyde
    -0.71
     Pigs
    -0.70
    ŃĶ
    -0.68
     briefs
    -0.68
    POSITIVE LOGITS
    ruly
    1.28
    itled
    1.25
    rave
    1.18
    ested
    1.14
    ainted
    1.07
    ribut
    1.07
    ired
    1.05
    apped
    1.03
    ravel
    1.03
    rained
    1.02
    Act Density 0.018%

    No Known Activations