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

    Neuronpedia

    INDEX
    Explanations

    scientific writing

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric expressions—especially floating‐point numbers and other scientific numerical values—in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     paw
    -0.07
     Mü
    -0.07
     head
    -0.06
    ैन
    -0.06
                    
    -0.06
     shoulder
    -0.06
     готов
    -0.06
     echoed
    -0.06
     crossing
    -0.06
     Natur
    -0.06
    POSITIVE LOGITS
    -sheet
    0.07
    rowth
    0.07
    isory
    0.06
     approximate
    0.06
     Dialog
    0.06
     Sik
    0.06
    swick
    0.06
     AUTHORS
    0.06
    *f
    0.06
    /forms
    0.06
    Act Density 0.055%

    No Known Activations