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

    Neuronpedia

    INDEX
    Explanations

    movie reviews

    np_max-act · gemini-2.0-flash

    This neuron fires on capitalized tokens that are part of names or titles—that is, proper nouns.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Marine
    -0.06
    érc
    -0.06
    .footer
    -0.06
     mineral
    -0.06
    _inverse
    -0.06
    igg
    -0.06
    esiz
    -0.06
    91
    -0.06
     contract
    -0.06
    _best
    -0.06
    POSITIVE LOGITS
    .'/
    0.07
    าฟ
    0.07
    StartupScript
    0.06
    _COMPILER
    0.06
     useDispatch
    0.06
     발표
    0.06
     puan
    0.06
    opause
    0.06
    執
    0.06
    RELATED
    0.06
    Act Density 0.060%

    No Known Activations