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

    Neuronpedia

    INDEX
    Explanations

    visual patterns

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     лучший
    0.61
    ళ్
    0.60
    。”
    0.58
     মন্দ
    0.57
     نه
    0.57
     preparing
    0.55
    ।”
    0.55
    忡
    0.55
     मंदिर
    0.55
     үч
    0.55
    POSITIVE LOGITS
     Fall
    0.62
     Dot
    0.61
     Nature
    0.57
     Wired
    0.57
     dotted
    0.56
     Seriously
    0.55
     wire
    0.55
     Its
    0.55
     Saf
    0.55
     Branch
    0.55
    Act Density 0.000%

    No Known Activations