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

    Neuronpedia

    INDEX
    Explanations

    highland topics and associations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    л
    1.52
    у
    1.44
    м
    1.38
    ли
    1.26
    д
    1.23
     as
    1.17
    ра
    1.12
    ле
    1.11
    7
    1.10
    lar
    1.09
    POSITIVE LOGITS
    та
    1.34
    P
    1.20
    ுக
    1.20
    M
    1.18
    ิ
    1.05
     invariance
    1.05
    T
    1.02
    TERN
    1.01
    ambarkan
    1.01
    tries
    1.01
    Act Density 0.005%

    No Known Activations