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

    Neuronpedia

    INDEX
    Explanations

    instances of returning or reconnecting themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    à¸Ĺาà¸Ļ
    -0.14
    -0.13
    HandlerContext
    -0.13
    ë»
    -0.13
    ilen
    -0.13
    sez
    -0.13
    ucker
    -0.13
    lique
    -0.13
    lotte
    -0.13
    lisi
    -0.13
    POSITIVE LOGITS
     leave
    1.25
     Leave
    1.15
     leaving
    1.11
    Leave
    1.11
     leaves
    1.09
    leave
    1.06
     left
    0.92
     Leaves
    0.92
     Leaving
    0.91
    _leave
    0.88
    Act Density 0.587%

    No Known Activations