© 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. 609
    Prev
    Next
    INDEX
    Explanations

    self.attribute access (negative activations)

    python-code · gemini-2.5-flash-lite

    string capitalize method

    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
    gh
    -0.20
    mit
    -0.19
    ss
    -0.16
    art
    -0.15
    ses
    -0.13
    ound
    -0.13
    is
    -0.05
    '
    0.00
    #
    0.00
    &
    0.00
    POSITIVE LOGITS
    ½
    0.30
    ents
    0.23
    ent
    0.22
    nt
    0.18
    ency
    0.17
    lust
    0.15
    ore
    0.15
     """↵
    0.15
    fe
    0.14
    lose
    0.10
    Activations Density 98.854%

    No Known Activations