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

    Neuronpedia

    INDEX
    Explanations

    referring to, relationships, or analysis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ق
    0.54
    ش
    0.51
    ند
    0.50
    ны
    0.49
    ções
    0.47
    缭
    0.47
    க்க
    0.46
    خي
    0.46
    م
    0.45
    ائن
    0.44
    POSITIVE LOGITS
     was
    0.53
     ਲਈ
    0.52
     tests
    0.51
     asks
    0.51
    size
    0.48
     Wag
    0.47
    stats
    0.46
    だけど
    0.46
     size
    0.46
     जताई
    0.46
    Act Density 0.000%

    No Known Activations