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

    Neuronpedia

    INDEX
    Explanations

    M

    np_max-act · gemini-2.0-flash

    The neuron is detecting radio technical identifiers, especially “FM” labels, call‐sign suffixes (e.g. “-FM”), and related frequency tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     firearm
    -0.07
     ولد
    -0.07
     Charleston
    -0.06
    conda
    -0.06
    рати
    -0.06
    _words
    -0.06
    ΙΚΗ
    -0.06
     Soldiers
    -0.06
    aras
    -0.06
    克斯
    -0.06
    POSITIVE LOGITS
    ấp
    0.06
    Gesture
    0.06
     TERMIN
    0.06
    。
    0.06
     repl
    0.06
     explain
    0.06
     свеж
    0.06
    Dual
    0.06
     multic
    0.06
     efficiencies
    0.06
    Act Density 0.002%

    No Known Activations