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

    Neuronpedia

    INDEX
    Explanations

    AI versus human

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of human-centered, empathy‐ or interpersonal‐related skills and qualities (e.g., judgment, empathy, human, interpersonal, aspects).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     coch
    -0.08
     dagger
    -0.07
     beef
    -0.06
    Ctx
    -0.06
    speech
    -0.06
    ptom
    -0.06
    	arr
    -0.06
    ocop
    -0.06
     pá
    -0.06
    CRY
    -0.06
    POSITIVE LOGITS
    TREE
    0.07
    舰
    0.06
     DIRECTORY
    0.06
    United
    0.06
     intf
    0.06
    icago
    0.06
     component
    0.06
    =true
    0.06
    ğine
    0.06
    _genre
    0.06
    Act Density 0.025%

    No Known Activations