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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating obsolescence or no longer being relevant

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     indescri
    -0.63
     unspeak
    -0.58
     withal
    -0.57
     Shakspeare
    -0.57
     impelled
    -0.55
     perchance
    -0.52
     Relaxed
    -0.51
    Sneaky
    -0.51
     Chronology
    -0.50
     exult
    -0.50
    POSITIVE LOGITS
     abbra
    0.86
     ideolog
    0.85
     reger
    0.84
     tramont
    0.84
     morire
    0.81
     Jerusal
    0.78
     utop
    0.78
     ormai
    0.76
     sabato
    0.76
     prega
    0.74
    Act Density 0.353%

    No Known Activations