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

    Neuronpedia

    INDEX
    Explanations

    code files and urls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    lagt
    0.39
     quinine
    0.38
     henkil
    0.38
     možnost
    0.38
    oniazid
    0.37
     centaines
    0.37
    短い
    0.37
     samma
    0.37
     misma
    0.37
     beide
    0.37
    POSITIVE LOGITS
    *.
    0.42
    ,.
    0.37
     *(
    0.36
     test
    0.35
    _.
    0.34
     भागीदारी
    0.34
    Slice
    0.34
     participación
    0.34
    元素的
    0.33
     परीक्षण
    0.33
    Act Density 0.002%

    No Known Activations