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

    Neuronpedia

    INDEX
    Explanations

    quality loss

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Deux
    0.48
     vrijdag
    0.46
     város
    0.46
     городе
    0.45
     Macbook
    0.44
     klaar
    0.44
     Thousands
    0.43
     République
    0.43
    年末
    0.43
     December
    0.42
    POSITIVE LOGITS
    াস্থ
    0.46
     söyled
    0.45
    adina
    0.43
     tradisional
    0.43
    荠
    0.43
     тради
    0.42
     safeguards
    0.42
    ib
    0.42
     preservatives
    0.42
     commitments
    0.41
    Act Density 0.007%

    No Known Activations