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

    Neuronpedia

    INDEX
    Explanations

    News articles

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens that introduce or label an individual actor or role—words like “agent,” “manager,” “player,” or the relative pronoun “who” marking a person doing something.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Imp
    -0.07
    _ext
    -0.06
     الأمر
    -0.06
     ekonomik
    -0.06
    038
    -0.06
    -Am
    -0.06
    SUP
    -0.06
    _dx
    -0.06
    Normally
    -0.06
     makes
    -0.06
    POSITIVE LOGITS
    buttonShape
    0.07
    puted
    0.06
    пе
    0.06
    (sk
    0.06
    (Screen
    0.06
    (primary
    0.06
    club
    0.06
    .field
    0.06
    ------------
    0.06
    ￥
    0.06
    Act Density 0.381%

    No Known Activations