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

    Neuronpedia

    INDEX
    Explanations

    punctuation followed by initiating token

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     omkring
    0.50
     ب
    0.49
     වීම
    0.47
     asli
    0.46
     banyaknya
    0.46
     musik
    0.46
     ไหน
    0.45
     banyak
    0.45
     एकदम
    0.45
     hidup
    0.45
    POSITIVE LOGITS
     dossiers
    0.53
     τῶν
    0.47
    esi
    0.44
    etty
    0.42
    comboBox
    0.41
     dossier
    0.40
    izing
    0.40
    !='')
    0.39
    strate
    0.39
    Before
    0.38
    Act Density 0.008%

    No Known Activations