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

    Neuronpedia

    INDEX
    Explanations

    Problems and solutions

    np_max-act · gemini-2.0-flash

    verbs related to fixing or changing something.

    oai_token-act-pair · gpt-4oTriggered by @tcai
    New Auto-Interp
    Negative Logits
     avoids
    -0.07
     link
    -0.07
    erne
    -0.06
     ohne
    -0.06
     plagiarism
    -0.06
    ряд
    -0.06
     '\\
    -0.06
     onto
    -0.06
    دمة
    -0.06
     raise
    -0.06
    POSITIVE LOGITS
    &W
    0.07
     NodeList
    0.07
     Bik
    0.06
    _FINAL
    0.06
    IGHL
    0.06
     NAN
    0.06
     vo
    0.06
    -bl
    0.06
    +B
    0.06
     goalt
    0.06
    Act Density 0.106%

    No Known Activations