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

    Neuronpedia

    INDEX
    Explanations

    but, exit, quit

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mengungkapkan
    0.76
     tells
    0.74
    ：（
    0.74
     abrang
    0.74
     dizer
    0.73
     menyebut
    0.71
     บอก
    0.69
    （
    0.69
     پور
    0.69
    Siempre
    0.67
    POSITIVE LOGITS
     but
    1.12
     lakini
    1.01
    exit
    1.00
     mutta
    0.94
    continue
    0.89
    quit
    0.89
     But
    0.89
     quit
    0.87
     exit
    0.86
     pero
    0.85
    Act Density 0.237%

    No Known Activations