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

    Neuronpedia

    INDEX
    Explanations

    examples of tests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cardiff
    0.39
     schwier
    0.39
    艰
    0.38
     constraining
    0.37
    東西
    0.37
     रीजन
    0.36
     Taxonomy
    0.36
     foreclose
    0.36
     Hend
    0.36
     የሚ
    0.36
    POSITIVE LOGITS
    init
    0.47
    sep
    0.46
     init
    0.45
    fit
    0.44
    arnya
    0.43
    Init
    0.42
    heated
    0.40
    ceased
    0.39
    Sep
    0.38
     sep
    0.38
    Act Density 0.002%

    No Known Activations