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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating contrast between two entities

    oai_token-act-pair · gpt-3.5-turbo

    prepositions and conjunctions indicating contrast or comparison

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    alities
    -0.69
    CAST
    -0.63
    igans
    -0.62
    izons
    -0.60
    ities
    -0.59
     selves
    -0.58
    ITIES
    -0.58
    adesh
    -0.58
    PLA
    -0.57
     staples
    -0.57
    POSITIVE LOGITS
     contrast
    0.87
     particular
    0.82
     nutshell
    0.82
     meanwhile
    0.78
     whom
    0.75
    whose
    0.71
     cooperation
    0.68
    ccording
    0.67
     incidentally
    0.67
     meantime
    0.66
    Act Density 0.135%

    No Known Activations