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

    Neuronpedia

    INDEX
    Explanations

    doing things effectively

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Immediate
    0.48
     Immediate
    0.46
    immediate
    0.45
     immediate
    0.42
     immédi
    0.40
     immediacy
    0.40
     IMMEDIATE
    0.39
     Slight
    0.39
    вної
    0.38
     खासियत
    0.38
    POSITIVE LOGITS
     efficiently
    2.69
     эффективно
    2.22
     effectively
    2.11
     efficacement
    2.11
    efficient
    1.97
     safely
    1.97
     reliably
    1.92
     efficient
    1.91
     effektiv
    1.89
     effiz
    1.89
    Act Density 0.137%

    No Known Activations