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

    Neuronpedia

    INDEX
    Explanations

    not moving or reacting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ever
    0.61
     never
    0.47
    ever
    0.45
    Ever
    0.45
     EVER
    0.44
     Ever
    0.41
     nunca
    0.41
     jamás
    0.41
     jamais
    0.40
    never
    0.39
    POSITIVE LOGITS
     komment
    0.57
     comentó
    0.51
     comentario
    0.47
     koment
    0.46
     comment
    0.46
     fidget
    0.46
     unkind
    0.46
    Comment
    0.45
     invisible
    0.45
     bergerak
    0.45
    Act Density 0.013%

    No Known Activations