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

    Neuronpedia

    INDEX
    Explanations

    random strings or abstract concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     čak
    0.49
     êtes
    0.49
     除了
    0.48
     Pire
    0.48
     aportar
    0.48
     dudes
    0.47
     Bullet
    0.46
    zynarod
    0.46
     aad
    0.46
     respetar
    0.45
    POSITIVE LOGITS
    Behavior
    0.43
     dismissal
    0.39
     overthrow
    0.39
     pronoun
    0.39
    თავ
    0.39
     convexity
    0.39
    褥
    0.39
    </u>
    0.38
     nanny
    0.38
    ovich
    0.38
    Act Density 0.009%

    No Known Activations