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

    Neuronpedia

    INDEX
    Explanations

    Code-related questions and answers

    np_max-act · gemini-2.0-flash

    The neuron detects phrases stating that something exceeds a numeric threshold (for example “has more than 3 DateTime properties”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     hiding
    -0.07
    前に
    -0.07
    Cancelar
    -0.06
     якій
    -0.06
    нт
    -0.06
    Player
    -0.06
    )=(
    -0.06
    ><
    -0.06
    Sanders
    -0.06
    .'
    -0.06
    POSITIVE LOGITS
    704
    0.07
     suggestions
    0.06
    短
    0.06
    과정
    0.06
     ماك
    0.06
    isque
    0.06
     es
    0.06
    ラス
    0.06
    791
    0.06
    702
    0.06
    Act Density 0.054%

    No Known Activations