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

    Neuronpedia

    INDEX
    Explanations

    concluding or contrasting statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     이건
    0.21
     jeans
    0.20
     họ
    0.20
    это
    0.20
     이거
    0.20
     swirls
    0.19
     headlights
    0.19
    Didn
    0.19
     bunnies
    0.19
     combos
    0.19
    POSITIVE LOGITS
     Therefore
    0.52
    Therefore
    0.49
     Consequently
    0.48
     Furthermore
    0.48
     Accordingly
    0.43
     Поэтому
    0.42
     therefore
    0.42
     Moreover
    0.41
    Consequently
    0.41
    因此
    0.40
    Act Density 0.086%

    No Known Activations