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

    Neuronpedia

    INDEX
    Explanations

    regular expression symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ။
    0.47
    dropDown
    0.46
     сроки
    0.44
    剌
    0.42
    0.42
     disposto
    0.42
     aiguë
    0.41
     സമീപ
    0.41
     нанесе
    0.40
     ciudadano
    0.40
    POSITIVE LOGITS
     glamorous
    0.51
     powerful
    0.50
     Paj
    0.49
    มีการ
    0.49
    та
    0.48
     proper
    0.48
     pacemaker
    0.47
     energy
    0.47
     luminous
    0.46
     poured
    0.45
    Act Density 0.001%

    No Known Activations