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

    Neuronpedia

    INDEX
    Explanations

    double and duplication

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    legged
    0.96
     mult
    0.89
    级别的
    0.87
     Mult
    0.86
    多
    0.86
     multip
    0.86
     δυ
    0.86
     multi
    0.83
    stakes
    0.83
     هڪ
    0.82
    POSITIVE LOGITS
     haszn
    0.90
    andus
    0.85
     alimentaire
    0.84
     irgendwo
    0.83
     gái
    0.83
    vodu
    0.82
     lobby
    0.82
    Ironically
    0.81
    Considering
    0.81
    ন্ধে
    0.81
    Act Density 0.047%

    No Known Activations