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

    Neuronpedia

    INDEX
    Explanations

    Pronouns

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the Portuguese feminine third-person singular pronoun “ela.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     much
    -0.07
     Much
    -0.07
    POINT
    -0.07
     took
    -0.06
    “No
    -0.06
    Kit
    -0.06
     Kaepernick
    -0.06
     fifteen
    -0.06
     VAT
    -0.06
     charismatic
    -0.06
    POSITIVE LOGITS
     она
    0.10
     she
    0.09
     He
    0.09
     They
    0.09
     они
    0.09
     Он
    0.09
     оно
    0.08
     She
    0.08
     he
    0.08
     он
    0.08
    Act Density 0.040%

    No Known Activations