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

    Neuronpedia

    INDEX
    Explanations

    brand

    np_max-act · gemini-2.0-flash

    This neuron activates specifically on German subword fragments (stems/pieces) found in compound words and morphological constructions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Sharper
    -0.08
    ضای
    -0.07
    (IR
    -0.07
     yapılan
    -0.07
    무
    -0.07
     якій
    -0.07
    Activation
    -0.06
    ノ
    -0.06
    なん
    -0.06
    Monitoring
    -0.06
    POSITIVE LOGITS
    ).↵
    0.08
    )).↵
    0.07
    ').↵
    0.07
    .↵
    0.07
    ',)↵
    0.07
     lug
    0.06
     wanted
    0.06
    の大
    0.06
     LoginForm
    0.06
    ”.↵
    0.06
    Act Density 0.260%

    No Known Activations