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

    Neuronpedia

    INDEX
    Explanations

    adverbs conveying intensity

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "very" used to emphasize a quality or condition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    peak
    -0.63
    igi
    -0.61
    Afee
    -0.60
    amel
    -0.59
    isec
    -0.59
    osal
    -0.58
    CHA
    -0.56
    iture
    -0.56
    ses
    -0.55
    mare
    -0.55
    POSITIVE LOGITS
     much
    1.01
     seldom
    0.92
    much
    0.88
     rarely
    0.87
     similar
    0.83
     close
    0.83
     important
    0.82
     MUCH
    0.81
     difficult
    0.80
     unlikely
    0.79
    Act Density 0.081%

    No Known Activations