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

    Neuronpedia

    INDEX
    Explanations

    start with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     протяжении
    0.93
     початку
    0.90
    Starter
    0.90
     podczas
    0.86
     zaradi
    0.86
    ваемые
    0.83
     restantes
    0.82
    本身
    0.82
     остались
    0.82
    рованные
    0.81
    POSITIVE LOGITS
     off
    1.29
     afresh
    1.28
     anew
    1.12
     fresh
    1.07
     humbly
    1.04
     experimenting
    1.01
     preparations
    0.95
     conversations
    0.94
     modestly
    0.94
     somewhere
    0.94
    Act Density 0.187%

    No Known Activations