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

    Neuronpedia

    INDEX
    Explanations

    block compounds and specific terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    وون
    0.54
    쪽에
    0.48
    gle
    0.47
    conflict
    0.45
    cabin
    0.44
    deficit
    0.44
    ෙ
    0.43
    ካ
    0.43
    گیر
    0.42
    гле
    0.41
    POSITIVE LOGITS
     block
    1.08
     Block
    1.05
     Blocks
    0.96
     blocks
    0.95
    Block
    0.91
     blok
    0.89
     Blok
    0.88
     ব্লক
    0.88
     blocked
    0.88
     ब्लॉक
    0.87
    Act Density 0.015%

    No Known Activations