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

    Neuronpedia

    INDEX
    Explanations

    measuring efficiency and positive tone

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ings
    0.53
    નાર
    0.48
    un
    0.46
    preprocessing
    0.43
    INGS
    0.42
    amatsu
    0.42
    uk
    0.41
    abbe
    0.41
    ers
    0.41
    u
    0.40
    POSITIVE LOGITS
     драй
    0.46
     حوالہ
    0.45
     تريد
    0.43
     يتح
    0.40
    莜
    0.40
    問題
    0.39
     dónde
    0.39
     المُ
    0.39
     favore
    0.38
    裡
    0.38
    Act Density 0.007%

    No Known Activations