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

    Neuronpedia

    INDEX
    Explanations

    designed for premium

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Cheap
    0.66
     Cheap
    0.64
     inexpensive
    0.59
    Simple
    0.55
     humble
    0.55
    便宜
    0.55
     cheap
    0.54
     simple
    0.50
    简单
    0.49
     beginner
    0.49
    POSITIVE LOGITS
     고급
    1.37
     premium
    1.20
    高端
    1.20
    高級
    1.17
     elite
    1.05
    高级
    1.05
     luxury
    1.02
    premium
    1.02
     sophisticated
    1.01
     Premium
    1.01
    Act Density 1.616%

    No Known Activations