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

    Neuronpedia

    INDEX
    Explanations

    Code identifiers and configurations

    np_max-act · gemini-2.0-flash

    The neuron activates on all-caps identifiers (constants, enum names, macro names) often containing underscores.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _scripts
    -0.07
    }%
    -0.07
     бума
    -0.07
    ύ
    -0.07
    iltr
    -0.06
    messages
    -0.06
    Signed
    -0.06
    oothing
    -0.06
    _heading
    -0.06
    rieving
    -0.06
    POSITIVE LOGITS
    isson
    0.08
    ,list
    0.07
    横
    0.06
    (chalk
    0.06
     Void
    0.06
    #ifndef
    0.06
    (orig
    0.06
     gra
    0.06
     approximation
    0.06
     motel
    0.06
    Act Density 0.160%

    No Known Activations