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

    Neuronpedia

    INDEX
    Explanations

    I can help

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     উৎ
    0.38
     divulgação
    0.36
     wm
    0.35
    reveal
    0.35
    arns
    0.35
     alta
    0.33
     подру
    0.33
    ิติ
    0.33
     разде
    0.32
     पुल
    0.32
    POSITIVE LOGITS
     assumed
    0.55
     Assuming
    0.52
     demonstrated
    0.50
    Assuming
    0.49
    ொ
    0.48
     assuming
    0.46
     exemplified
    0.46
     gave
    0.45
     assume
    0.45
     assum
    0.45
    Act Density 0.009%

    No Known Activations