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

    Neuronpedia

    INDEX
    Explanations

    information about people, such as names, actions, and personal details

    oai_token-act-pair · gpt-3.5-turbo

    third-person singular pronouns, particularly "he" and "she"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    γ
    -0.74
    rame
    -0.74
    Rush
    -0.63
     Pair
    -0.62
    earch
    -0.60
    itaire
    -0.60
    Chance
    -0.59
    rain
    -0.59
    ogle
    -0.59
     "+
    -0.59
    POSITIVE LOGITS
     encount
    0.95
    'd
    0.82
    'll
    0.81
     redes
    0.78
     reluct
    0.76
    zbollah
    0.75
     detractors
    0.72
     juven
    0.70
     experien
    0.69
     lamented
    0.68
    Act Density 0.651%

    No Known Activations