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

    Neuronpedia

    INDEX
    Explanations

    comparisons between different situations or entities

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases that highlight preferences or contrasts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    estern
    -0.86
    illary
    -0.75
    ero
    -0.72
    umbn
    -0.70
    roo
    -0.68
    ahime
    -0.68
    OSP
    -0.67
    ango
    -0.67
    eri
    -0.67
    uron
    -0.66
    POSITIVE LOGITS
     anything
    1.26
     any
    1.05
     ever
    0.96
     anybody
    0.94
     anyone
    0.94
     vice
    0.81
     necessarily
    0.80
     usual
    0.75
     actual
    0.75
     anywhere
    0.74
    Act Density 0.102%

    No Known Activations