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

    Neuronpedia

    INDEX
    Explanations

    accusations or crimes

    np_max-act · gemini-2.0-flash

    This neuron flags past-participle verb forms (especially “-ed” or irregular “-en” forms) often used in passive or legal/descriptive contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     map
    -0.07
    nání
    -0.06
    _wave
    -0.06
    _motor
    -0.06
     أخ
    -0.06
    attended
    -0.06
    ие
    -0.06
    apı
    -0.06
    LEE
    -0.06
    athlon
    -0.06
    POSITIVE LOGITS
     orth
    0.08
     dě
    0.07
    �
    0.07
    (eventName
    0.06
     seals
    0.06
    .setAuto
    0.06
     paralle
    0.06
    -coordinate
    0.06
    _UClass
    0.06
    ')
↵
↵
    0.06
    Act Density 0.040%

    No Known Activations