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

    Neuronpedia

    INDEX
    Explanations

    coding style and standards

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    um
    1.26
    ва
    1.23
    ী
    1.20
    ก
    1.13
    ாக
    1.12
    ת
    1.09
    का
    1.08
    ει
    1.08
    اد
    1.05
    ec
    1.04
    POSITIVE LOGITS
     Coding
    1.20
    Coding
    1.10
     coding
    1.09
     coder
    1.00
     в
    0.89
     ува
    0.85
     huesos
    0.84
    coding
    0.83
     сове
    0.81
     информацией
    0.81
    Act Density 0.007%

    No Known Activations