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

    Neuronpedia

    INDEX
    Explanations

    empty MAX_ACTIVATING_TOKENS

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     CQL
    1.10
     চৈ
    1.10
    லிக்க
    1.08
     futebol
    1.07
    Cadastro
    1.07
    1.05
     filosóf
    1.05
    ные
    1.04
    ோ
    1.04
    汼
    1.04
    POSITIVE LOGITS
     exaggerate
    0.99
     що
    0.95
     appr
    0.95
     例如
    0.91
     &
    0.88
     遊ん
    0.87
    와의
    0.86
    屿
    0.86
     affirme
    0.85
     aren
    0.84
    Act Density 1.270%

    No Known Activations