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

    Neuronpedia

    INDEX
    Explanations

    underscore blank

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     take
    0.47
     goes
    0.44
     going
    0.43
    take
    0.43
     stress
    0.43
     Going
    0.41
     Take
    0.41
     slam
    0.41
    Going
    0.37
     prendre
    0.37
    POSITIVE LOGITS
    blank
    0.46
    Blank
    0.45
     intelligents
    0.43
    blos
    0.41
     renovación
    0.40
    бели
    0.39
     sebe
    0.39
    図柄
    0.39
    空白
    0.39
     blanks
    0.38
    Act Density 0.000%

    No Known Activations