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

    Neuronpedia

    INDEX
    Explanations

    evidence

    np_max-act · gemini-2.0-flash

    The neuron activates on terms related to legal evidence and proof (e.g. “proof,” “evidence,” “beyond reasonable doubt”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    !I
    -0.07
     @{↵
    -0.06
     могли
    -0.06
     pou
    -0.06
    139
    -0.06
    .Speed
    -0.06
    ливий
    -0.06
     //}↵
    -0.06
    >>
    -0.06
    .pg
    -0.06
    POSITIVE LOGITS
     evidence
    0.07
     perish
    0.07
     güncel
    0.07
     floral
    0.06
    ibilidad
    0.06
    aways
    0.06
     institutional
    0.06
     长
    0.06
     visceral
    0.06
    _BOOLEAN
    0.06
    Act Density 0.030%

    No Known Activations