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

    Neuronpedia

    INDEX
    Explanations

    effects are observed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    それぞれ
    0.88
     각각
    0.82
     যদি
    0.74
     ഓരോ
    0.72
    यदि
    0.71
    が発生
    0.71
     касается
    0.70
    문
    0.70
     each
    0.66
    если
    0.65
    POSITIVE LOGITS
     enjoys
    1.05
     enjoy
    1.01
     fared
    0.99
     tend
    0.92
     find
    0.90
     rejoice
    0.89
     flocked
    0.89
     thrived
    0.89
     merasa
    0.89
     receive
    0.88
    Act Density 0.043%

    No Known Activations