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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating the creation or proposal of new ideas or solutions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    toi
    -0.15
    ieri
    -0.15
    iger
    -0.15
    esseract
    -0.14
     Parm
    -0.14
    باØŃ
    -0.14
    ikel
    -0.14
    åģ¶
    -0.14
    _dropout
    -0.14
    ndl
    -0.14
    POSITIVE LOGITS
     ways
    0.23
     solutions
    0.19
     ideas
    0.19
     nick
    0.18
     idea
    0.18
    åĬŀæ³ķ
    0.18
     strategies
    0.17
     solution
    0.17
     Ways
    0.16
    idea
    0.15
    Act Density 0.034%

    No Known Activations