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

    Neuronpedia

    INDEX
    Explanations

    concepts followed by modifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     divided
    0.39
     sino
    0.35
    PT
    0.35
     Sä
    0.35
     angles
    0.35
    ÍA
    0.35
     s
    0.34
     name
    0.34
     nume
    0.34
     O
    0.33
    POSITIVE LOGITS
     appropriately
    0.56
     accordingly
    0.55
    起来
    0.54
     Accordingly
    0.49
    Accordingly
    0.48
    ify
    0.48
     себя
    0.47
    ize
    0.46
    лізу
    0.45
    itize
    0.44
    Act Density 0.047%

    No Known Activations