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

    Neuronpedia

    INDEX
    Explanations

    statistical analysis

    np_max-act · gemini-2.0-flash

    The neuron is tuned to detect numeric expressions—especially floating-point numbers or decimal metrics—in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    tod
    -0.06
     nv
    -0.06
    .ModelForm
    -0.06
    .ar
    -0.06
     xs
    -0.06
    departure
    -0.06
    _ix
    -0.06
    linewidth
    -0.06
     INLINE
    -0.06
     deposits
    -0.06
    POSITIVE LOGITS
    ตอบ
    0.07
    П
    0.07
    ，由
    0.06
    ımda
    0.06
    latesAutoresizingMaskIntoConstraints
    0.06
    weise
    0.06
     GOODMAN
    0.06
     classified
    0.06
     halls
    0.06
    مم
    0.06
    Act Density 0.019%

    No Known Activations