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

    Neuronpedia

    INDEX
    Explanations

    I've, I'll statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
                
    0.17
     当
    0.17
     বলা
    0.16
     విషయంలో
    0.16
     när
    0.15
    髀
    0.15
     Previously
    0.15
     gehabt
    0.15
     auraient
    0.15
                            
    0.15
    POSITIVE LOGITS
     provides
    0.23
     provide
    0.23
     will
    0.22
     consists
    0.21
     are
    0.21
    utiliser
    0.20
    은
    0.19
    les
    0.19
     wholeheartedly
    0.19
     hereby
    0.18
    Act Density 0.422%

    No Known Activations