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

    Neuronpedia

    INDEX
    Explanations

    matrix, pizza, mix

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    м
    1.92
    ु
    1.71
    ा
    1.70
     extrait
    1.65
     inim
    1.61
     háb
    1.60
    ות
    1.60
     analogous
    1.57
    히
    1.56
    عت
    1.53
    POSITIVE LOGITS
    𝐞
    1.95
    𝐢
    1.91
    am
    1.86
    𝐚
    1.78
    𝐧
    1.78
    юць
    1.77
    𝐩
    1.71
    𝐫
    1.69
    𝐱
    1.67
    𝐮
    1.66
    Act Density 0.060%

    No Known Activations