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-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    <bos>
    -2.25
     maintain
    -0.72
     continue
    -0.71
     pour
    -0.69
     stay
    -0.69
     also
    -0.68
     will
    -0.64
     be
    -0.64
     finally
    -0.64
     are
    -0.63
    POSITIVE LOGITS
     lele
    1.79
     meis
    1.76
     bandung
    1.72
     Juf
    1.72
     Minang
    1.66
     maroc
    1.64
     unlaw
    1.64
     fta
    1.64
     Græ
    1.61
     mef
    1.58
    Act Density 0.732%

    No Known Activations