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

    Neuronpedia

    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. 1906
    Prev
    Next
    INDEX
    Explanations

    permutations or capitalize methods (negative activations)

    python-code · gemini-2.5-flash-lite

    `capitalize` or `connect`

    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
     g
    -0.12
    ar
    -0.07
    ut
    -0.07
    up
    -0.05
     of
    -0.04
    '
    0.00
    #
    0.00
    &
    0.00
    %
    0.00
    *
    0.00
    POSITIVE LOGITS
    ke
    0.24
    si
    0.17
     th
    0.16
    ch
    0.16
    ct
    0.12
    a
    0.09
    ing
    0.09
    ro
    0.08
    g
    0.05
    '
    0.00
    Activations Density 99.693%

    No Known Activations