© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Weight-Sparse Transformers
    5. CircuitGPT-Python
    7. Residual Stream
    9. 7-RESID
    11. 1065
    Prev
    Next
    INDEX
    Explanations

    function definition or `if` statement (negative activations)

    python-code · gemini-2.5-flash-lite

    code block closing brace

    python-code · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    16,384 prompts, 256 tokens each
    Dataset (Dashboard)
    neuronpedia/python-code-simplified
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     best
    -0.35
    .random
    -0.30
     random
    -0.30
    random
    -0.26
    ateg
    -0.20
     sele
    -0.20
    valu
    -0.19
    bab
    -0.19
    jective
    -0.18
    itness
    -0.17
    POSITIVE LOGITS
     char
    0.04
    gs
    0.04
    The
    0.04
    '
    0.00
    #
    0.00
    &
    0.00
    %
    0.00
    *
    0.00
    )
    0.00
    !
    0.00
    Activations Density 100.000%

    No Known Activations