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

    Neuronpedia

    INDEX
    Explanations

    how things are presented or operate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Become
    0.42
     превра
    0.40
     ഉണ്ടാ
    0.40
     menjadi
    0.39
     дворе
    0.39
     achievements
    0.38
     becomes
    0.38
    Different
    0.38
     Different
    0.37
    变为
    0.37
    POSITIVE LOGITS
     behaved
    0.57
     worded
    0.55
     configured
    0.53
    看待
    0.51
     presented
    0.50
     orientated
    0.49
     behaving
    0.49
    behaved
    0.48
     behave
    0.48
    configured
    0.48
    Act Density 0.012%

    No Known Activations