© 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
    2. OpenAI · Weight-Sparse Transformers
    3. CircuitGPT-Python
    4. Residual Stream
    5. 7-RESID
    6. 1831
    Prev
    Next
    INDEX
    Explanations

    self in method definitions (negative activations)

    python-code · gemini-2.5-flash-lite

    attribute access `.in` or method calls `forward`

    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
    apac
    -0.32
    _cost
    -0.25
     cost
    -0.23
    apacity
    -0.21
    rice
    -0.12
    ure
    -0.08
     [(
    -0.06
     weights
    -0.06
     weight
    -0.04
    '
    0.00
    POSITIVE LOGITS
     depth
    0.16
    mmand
    0.16
     sim
    0.13
    _line
    0.09
     sc
    0.09
     visited
    0.08
     on
    0.05
    '
    0.00
    #
    0.00
    &
    0.00
    Activations Density 100.000%

    No Known Activations