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

    Neuronpedia

    INDEX
    Explanations

    cassette tapes

    np_max-act · gemini-2.0-flash

    This neuron activates on floating-point number tokens (decimal numerical strings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Republicans
    -0.07
    community
    -0.06
    IFA
    -0.06
     Ali
    -0.06
    istically
    -0.06
    .process
    -0.06
     Proper
    -0.06
    actual
    -0.06
    ari
    -0.06
     attraction
    -0.06
    POSITIVE LOGITS
    -то
    0.07
    cdn
    0.07
     औ
    0.07
    まで
    0.07
    تد
    0.06
     z
    0.06
     hızla
    0.06
     σε
    0.06
     ideally
    0.06
     callbacks
    0.06
    Act Density 0.022%

    No Known Activations