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

    Neuronpedia

    INDEX
    Explanations

    adjectives followed by descriptors or judgments about people or things

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating disparity or variance among groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uggest
    -0.79
     +++
    -0.68
     Hover
    -0.67
    None
    -0.65
    not
    -0.65
    maybe
    -0.64
     instead
    -0.63
     Missing
    -0.63
    çĶ°
    -0.62
     Quote
    -0.62
    POSITIVE LOGITS
     necessarily
    0.90
     agree
    0.76
    icable
    0.75
     agrees
    0.74
     stellar
    0.72
     equally
    0.71
     fortunate
    0.71
     appreciated
    0.70
    avorable
    0.69
    escent
    0.68
    Act Density 0.191%

    No Known Activations