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

    Neuronpedia

    INDEX
    Explanations

    prime numbers and counts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Inputs
    0.73
    Input
    0.72
    Especific
    0.67
    anego
    0.66
     fournis
    0.66
    νον
    0.66
     मुझ
    0.66
    schutz
    0.65
     exempel
    0.64
    款
    0.63
    POSITIVE LOGITS
     counts
    1.53
     sorted
    1.50
     dictionaries
    1.39
     dictionary
    1.35
     unique
    1.35
     uniques
    1.34
     keys
    1.33
    counts
    1.33
     histograms
    1.28
     duplicates
    1.27
    Act Density 1.448%

    No Known Activations