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

    Neuronpedia

    INDEX
    Explanations

    past tense verbs

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "to be" in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    now
    -0.73
    _______
    -0.70
    ethy
    -0.70
    opath
    -0.66
    erno
    -0.66
    uras
    -0.65
    ——
    -0.64
     anymore
    -0.64
    Trade
    -0.64
    Dialogue
    -0.63
    POSITIVE LOGITS
     originally
    0.95
     previously
    0.92
    hes
    0.89
     pione
    0.87
     recently
    0.85
     subur
    0.84
     formerly
    0.84
    EStream
    0.81
    hers
    0.78
    Þ
    0.78
    Act Density 0.749%

    No Known Activations