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

    Neuronpedia

    INDEX
    Explanations

    begins subtly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     length
    0.41
     bias
    0.40
     inflammatory
    0.40
     magnet
    0.40
     inhibitory
    0.40
     selection
    0.40
     nào
    0.40
     prohibitive
    0.40
    selection
    0.39
    esor
    0.38
    POSITIVE LOGITS
     agrand
    0.50
     растения
    0.47
    ค่อย
    0.47
     автомоби
    0.47
    Continuing
    0.47
    roits
    0.46
     коммуна
    0.45
     नवी
    0.44
    庞
    0.44
    🔚
    0.44
    Act Density 0.001%

    No Known Activations