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

    Neuronpedia

    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron activates on decimal version-number tokens (e.g. the “1.3447…” parts of “,v” revision identifiers) in source‐control metadata.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    +i
    -0.07
    iedo
    -0.06
    UpperCase
    -0.06
     evenings
    -0.06
    Implement
    -0.06
    _minor
    -0.06
    _equals
    -0.06
    (z
    -0.06
    lil
    -0.06
    �
    -0.06
    POSITIVE LOGITS
     감사
    0.07
     sue
    0.07
     الإن
    0.07
    ()}}↵
    0.06
    esseract
    0.06
    _SENSOR
    0.06
     MOTOR
    0.06
     النو
    0.06
     Marxist
    0.06
    ctrine
    0.06
    Act Density 0.002%

    No Known Activations