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

    Neuronpedia

    INDEX
    Explanations

    Identifiers involving 2 or letters

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on the placeholder token referring to the second character (NAME_2).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pain
    -0.07
     Martins
    -0.06
     món
    -0.06
    Ant
    -0.06
     analyse
    -0.06
     Iss
    -0.06
     polarization
    -0.06
    风
    -0.06
    -awesome
    -0.06
    _Un
    -0.06
    POSITIVE LOGITS
    .V
    0.07
    월
    0.07
    532
    0.07
    work
    0.07
    ."<
    0.07
    .ov
    0.06
    ležit
    0.06
     qx
    0.06
    κρι
    0.06
     result
    0.06
    Act Density 0.044%

    No Known Activations