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

    Neuronpedia

    INDEX
    Explanations

    names of concepts and their explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    confident
    0.50
     .
    0.47
    à
    0.47
     {
    0.47
    für
    0.45
    $.
    0.45
    å
    0.44
     in
    0.44
     בין
    0.44
     (%)
    0.43
    POSITIVE LOGITS
    ла
    0.63
    erar
    0.52
     WOULD
    0.48
     વરસ
    0.47
     দারি
    0.47
     COMPANIES
    0.47
    ⊟
    0.46
    ERING
    0.45
     sabbam
    0.44
     бизне
    0.44
    Act Density 0.000%

    No Known Activations