© 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. 1078
    Prev
    Next
    INDEX
    Explanations

    exception related to `cep` or `sub` (negative activations)

    python-code · gemini-2.5-flash-lite

    main loop start

    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
    ange
    -0.27
    67
    -0.18
    _node
    -0.16
    straint
    -0.16
    _edge
    -0.15
    ([
    -0.14
    15
    -0.12
    loor
    -0.11
    E
    -0.10
    ab
    -0.10
    POSITIVE LOGITS
    od
    0.26
    ¡
    0.20
    39
    0.18
    list
    0.16
    40
    0.13
    _al
    0.13
    41
    0.10
    keys
    0.09
    row
    0.09
     al
    0.09
    Activations Density 99.112%

    No Known Activations