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

    Neuronpedia

    INDEX
    Explanations

    ear

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of “ear” (ear-related tokens).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	img
    -0.08
    	diff
    -0.07
    建设
    -0.07
     Boston
    -0.07
    70
    -0.07
    .Blue
    -0.07
     obligated
    -0.07
     Cox
    -0.07
    “No
    -0.07
    	cmd
    -0.07
    POSITIVE LOGITS
     ears
    0.13
     ear
    0.12
     Ear
    0.12
     earrings
    0.09
    Ear
    0.08
    ra
    0.07
    earing
    0.07
    ar
    0.07
    AR
    0.07
    rees
    0.07
    Act Density 0.007%

    No Known Activations