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

    Neuronpedia

    INDEX
    Explanations

    conjunctions connecting phrases or clauses

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     prétend
    -1.24
     squa
    -1.23
     ftu
    -1.22
     soigne
    -1.19
     dispen
    -1.19
     Augu
    -1.15
     fta
    -1.14
     inev
    -1.14
     Keny
    -1.13
     secon
    -1.13
    POSITIVE LOGITS
     whose
    1.22
    whose
    1.00
     which
    0.94
     who
    0.90
     whom
    0.86
    which
    0.79
    ซึ่ง
    0.76
    Whose
    0.72
     которые
    0.72
     where
    0.71
    Act Density 0.257%

    No Known Activations