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

    Neuronpedia

    INDEX
    Explanations

    problem analysis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     subtraction
    0.43
     subtract
    0.39
    teman
    0.39
     inscr
    0.38
     понять
    0.38
     lambda
    0.38
     Multiply
    0.38
     cun
    0.38
    Tol
    0.38
    Trim
    0.37
    POSITIVE LOGITS
     zarówno
    0.49
    both
    0.40
     ảo
    0.40
     cả
    0.40
    சென்னை
    0.39
     πάνω
    0.38
     Kellogg
    0.38
    ारस
    0.38
     美国
    0.37
     both
    0.36
    Act Density 0.000%

    No Known Activations