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

    Neuronpedia

    INDEX
    Explanations

    phrases related to step-by-step instructions or process descriptions

    oai_token-act-pair · gpt-3.5-turbo

    second-person references in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ðĿ
    -0.71
    à
    -0.68
    OX
    -0.64
    idity
    -0.63
    abases
    -0.62
     Unreal
    -0.59
    istries
    -0.58
     Valiant
    -0.58
    CHAT
    -0.58
    fff
    -0.57
    POSITIVE LOGITS
    ensis
    0.74
     progresses
    0.74
     progressed
    0.72
     relates
    0.72
     transitioned
    0.71
     matured
    0.71
     unfolds
    0.70
     regards
    0.67
     evolves
    0.64
    soDeliveryDate
    0.64
    Act Density 0.190%

    No Known Activations