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

    Neuronpedia

    INDEX
    Explanations

    discharge

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of someone’s military discharge (especially “honorably discharged” and similar discharge statements).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Preferences
    -0.07
    cam
    -0.07
     endanger
    -0.07
    _SEARCH
    -0.07
     beneficiation
    -0.06
     establishment
    -0.06
    ">//
    -0.06
    covery
    -0.06
    voice
    -0.06
    _Man
    -0.06
    POSITIVE LOGITS
    olds
    0.06
     ét
    0.06
     scenes
    0.06
     nivel
    0.06
     karş
    0.06
     lawful
    0.06
    го
    0.06
     ode
    0.06
    omencl
    0.06
     Seattle
    0.06
    Act Density 0.005%

    No Known Activations