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

    Neuronpedia

    INDEX
    Explanations

    how actions are performed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     いっ
    0.47
     tutors
    0.47
     originals
    0.43
     algebra
    0.41
     swearing
    0.41
     archives
    0.41
     starters
    0.41
    เรียน
    0.41
     eleven
    0.41
     explosions
    0.41
    POSITIVE LOGITS
     delle
    0.49
     foglie
    0.49
     belir
    0.47
     singolo
    0.46
    ượt
    0.46
    جمالي
    0.46
     nutzt
    0.45
    sluš
    0.45
     selecionar
    0.45
    ewnątrz
    0.44
    Act Density 0.001%

    No Known Activations