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

    Neuronpedia

    INDEX
    Explanations

    code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    rež
    0.47
     乾燥
    0.42
    ódio
    0.41
    ombres
    0.41
    Sumber
    0.40
     schéma
    0.40
    burst
    0.40
     Peral
    0.39
     цвето
    0.39
    brecht
    0.39
    POSITIVE LOGITS
     foo
    0.61
    foo
    0.58
    Foo
    0.48
     Foo
    0.48
     confusing
    0.48
     problematic
    0.47
     ambigu
    0.45
     unrelated
    0.45
     misused
    0.44
     mistaken
    0.43
    Act Density 0.026%

    No Known Activations