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

    Neuronpedia

    INDEX
    Explanations

    mat, matic, matplotlib, matadors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    지
    1.20
    ,
    1.13
    ون
    1.09
    essä
    1.07
    나
    1.05
    కు
    1.00
    ing
    0.96
    ais
    0.95
    呫
    0.94
    in
    0.92
    POSITIVE LOGITS
    с
    1.37
     for
    1.18
    ל
    1.06
    ス
    1.05
    ו
    1.05
     mat
    1.02
     Mat
    1.01
    s
    1.01
     MAT
    1.00
    ो
    0.99
    Act Density 0.024%

    No Known Activations