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

    Neuronpedia

    INDEX
    Explanations

    end of sentences

    np_max-act · gemini-2.0-flash

    The neuron detects numeric word‐count constraints (e.g. the “no more than 50 words” limit).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     anche
    -0.07
    Middle
    -0.07
    čního
    -0.07
    (proto
    -0.06
    InputModule
    -0.06
    ber
    -0.06
    adiens
    -0.06
    fusion
    -0.06
    过来
    -0.06
     Burlington
    -0.06
    POSITIVE LOGITS
     cipher
    0.06
    \DB
    0.06
     مدت
    0.06
                         ↵
    0.06
    ิภาพ
    0.06
                      ↵
    0.06
    :title
    0.06
    ังจาก
    0.06
     rộng
    0.06
    Away
    0.06
    Act Density 0.010%

    No Known Activations