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

    Neuronpedia

    INDEX
    Explanations

    setting random seed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ysis
    -0.79
    Czym
    -0.75
     Rohan
    -0.72
    야
    -0.72
     nec
    -0.71
     Wright
    -0.71
    cett
    -0.69
    survi
    -0.69
    разуме
    -0.69
     Berliner
    -0.69
    POSITIVE LOGITS
     seed
    3.81
    seed
    3.36
     Seed
    2.98
    Seed
    2.95
     seeds
    2.91
    SEED
    2.63
     SEED
    2.45
     seeding
    2.39
     seeded
    2.38
    seeds
    2.31
    Act Density 0.015%

    No Known Activations