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

    Neuronpedia

    INDEX
    Explanations

    when

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    consumption
    0.57
    n
    0.54
    re
    0.52
    al
    0.50
    por
    0.50
    probably
    0.49
    really
    0.47
    و
    0.47
    choices
    0.45
    specific
    0.45
    POSITIVE LOGITS
     lashes
    0.47
    iczna
    0.46
     mettent
    0.44
    ặn
    0.44
     modulates
    0.44
     دولت
    0.43
     দুটি
    0.43
    ıştır
    0.43
     substituting
    0.42
    чными
    0.42
    Act Density 0.001%

    No Known Activations