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

    Neuronpedia

    INDEX
    Explanations

    very, quite, intensifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    /
    0.38
    WITH
    0.38
    with
    0.37
    Factors
    0.36
    Without
    0.34
    0
    0.33
    与
    0.33
    Failure
    0.33
     with
    0.32
    POST
    0.32
    POSITIVE LOGITS
     Очень
    0.83
     veldig
    0.80
     очень
    0.77
     довольно
    0.77
     bardzo
    0.76
     खूप
    0.75
     nagyon
    0.75
     very
    0.74
     굉장히
    0.74
     très
    0.73
    Act Density 0.555%

    No Known Activations