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

    Neuronpedia

    INDEX
    Explanations

    mechanics

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the substring “mechanic” (as in “quantum mechanic(s)”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    }|
    -0.07
     Indo
    -0.07
    茶
    -0.07
    _UINT
    -0.07
    	write
    -0.07
     XY
    -0.06
    XY
    -0.06
    ций
    -0.06
     alpha
    -0.06
     Marty
    -0.06
    POSITIVE LOGITS
    (screen
    0.07
    (cin
    0.07
     jogador
    0.07
     âm
    0.06
    oops
    0.06
     culp
    0.06
     Frequently
    0.06
     cours
    0.06
    lias
    0.06
    otherapy
    0.06
    Act Density 0.004%

    No Known Activations