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

    Neuronpedia

    INDEX
    Explanations

    issue

    np_max-act · gemini-2.0-flash

    The neuron selectively lights up on floating‐point number tokens (i.e. strings containing a decimal point and digits).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     nộp
    -0.07
     roast
    -0.06
     rug
    -0.06
     sev
    -0.06
    .sec
    -0.06
    _io
    -0.06
    .atom
    -0.06
    das
    -0.06
    _don
    -0.06
     Sev
    -0.06
    POSITIVE LOGITS
     coefficient
    0.07
     assists
    0.07
     *↵
    0.07
    สอบ
    0.06
    ัญญ
    0.06
     Ariel
    0.06
    _ASSUME
    0.06
    Coefficient
    0.06
     coherence
    0.06
     Gradient
    0.06
    Act Density 0.010%

    No Known Activations