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

    Neuronpedia

    INDEX
    Explanations

    rules and correctness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Lak
    -0.10
    ém
    -0.09
    odia
    -0.09
    /utility
    -0.09
     plaintext
    -0.09
    Utility
    -0.09
     cuck
    -0.09
     Leonardo
    -0.09
    .nih
    -0.08
     मण
    -0.08
    POSITIVE LOGITS
     memory
    0.16
     Memory
    0.12
     security
    0.12
    åħ¼
    0.11
    memory
    0.11
     pros
    0.11
    Memory
    0.10
    å¹
    0.10
     grain
    0.10
     resource
    0.10
    Act Density 0.099%

    No Known Activations