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

    Neuronpedia

    INDEX
    Explanations

    arrangements

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     journalism
    -0.08
     Dani
    -0.07
    Income
    -0.07
    income
    -0.07
     Journalism
    -0.07
     style
    -0.07
    商
    -0.07
     dbo
    -0.07
     mentorship
    -0.07
    Sponsor
    -0.07
    POSITIVE LOGITS
     scrambling
    0.14
     permutation
    0.14
     permutations
    0.13
     scrambled
    0.13
     scramble
    0.13
    Permutation
    0.12
    Shuffle
    0.11
     состояние
    0.11
    shuffle
    0.11
     parity
    0.11
    Act Density 0.016%

    No Known Activations