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

    Neuronpedia

    INDEX
    Explanations

    go first

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mostly
    0.98
     majority
    0.90
     dominate
    0.89
     maximization
    0.88
     predominance
    0.88
     mainly
    0.88
     predominant
    0.86
     dominant
    0.86
     dominating
    0.86
     dominated
    0.84
    POSITIVE LOGITS
    First
    1.19
     First
    1.12
     FIRST
    0.95
    FIRST
    0.93
    Пер
    0.84
    пер
    0.82
    호
    0.82
    Kuch
    0.82
    Editor
    0.81
    RAF
    0.80
    Act Density 0.002%

    No Known Activations