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

    Neuronpedia

    INDEX
    Explanations

    technical terms

    np_max-act · gemini-2.0-flash

    This neuron detects multiword technical or proper‐noun phrases (e.g. “Content distribution Network,” “Low earth Orbit,” “Space stations,” “M&A”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Of
    -0.06
     RESPONS
    -0.06
    WithData
    -0.06
     plainly
    -0.06
    .-
    -0.06
     lesbische
    -0.06
    _TH
    -0.06
    _se
    -0.06
    linear
    -0.06
    oooooooo
    -0.06
    POSITIVE LOGITS
    lobe
    0.07
    に向
    0.07
     
    0.07
    nas
    0.07
    (gc
    0.06
     سان
    0.06
    (vc
    0.06
     wear
    0.06
    pread
    0.06
     Amp
    0.06
    Act Density 0.366%

    No Known Activations