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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron is activated by numeric literals—especially decimal numbers and version or floating‐point values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    county
    -0.07
     Sheep
    -0.07
    already
    -0.07
    .loop
    -0.07
    .stamp
    -0.07
    vip
    -0.07
     hemorrh
    -0.07
    selectors
    -0.06
     messagebox
    -0.06
     devoted
    -0.06
    POSITIVE LOGITS
     xác
    0.07
     zal
    0.07
     Caj
    0.06
    نع
    0.06
     ingresar
    0.06
     빨
    0.06
     Bbw
    0.06
    ráf
    0.05
     защит
    0.05
    .setAlignment
    0.05
    Act Density 0.070%

    No Known Activations