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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating intention or plans

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing intention or plans

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acks
    -0.74
    cit
    -0.72
    aud
    -0.72
    aura
    -0.70
    bury
    -0.69
    anging
    -0.61
    yards
    -0.61
    wal
    -0.60
     Solitaire
    -0.60
    visors
    -0.59
    POSITIVE LOGITS
    ments
    0.77
    mber
    0.75
    л
    0.71
    ratulations
    0.70
    orial
    0.70
    reprene
    0.70
    chwitz
    0.69
    terday
    0.68
     intends
    0.67
    lessly
    0.67
    Act Density 0.014%

    No Known Activations