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

    Neuronpedia

    INDEX
    Explanations

    constantly

    np_max-act · gemini-2.0-flash

    This neuron appears never to activate—it doesn’t respond to any tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     relocation
    -0.07
    /false
    -0.07
     θ
    -0.07
    .ArgumentParser
    -0.07
    Conversation
    -0.06
     parsed
    -0.06
    _dirs
    -0.06
     {},
    -0.06
     commons
    -0.06
     utilise
    -0.06
    POSITIVE LOGITS
    jed
    0.08
    punk
    0.07
    ้าส
    0.06
     swal
    0.06
    ed
    0.06
     совет
    0.06
    usty
    0.06
    .definition
    0.06
     Roe
    0.06
    าหาร
    0.06
    Act Density 0.024%

    No Known Activations