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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "to," particularly in contexts indicating a return or recommitment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ütün
    -0.16
    enties
    -0.16
    ibur
    -0.15
    loff
    -0.15
    anine
    -0.14
    zim
    -0.14
    hend
    -0.14
    outcome
    -0.13
    zw
    -0.13
     lleg
    -0.13
    POSITIVE LOGITS
     basics
    0.24
     normal
    0.23
     sender
    0.22
     fold
    0.20
     roots
    0.19
     Sender
    0.19
    roots
    0.19
    normal
    0.19
     earth
    0.19
    æŃ£å¸¸
    0.18
    Act Density 0.071%

    No Known Activations