© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Weight-Sparse Transformers
    5. CircuitGPT-Python
    7. Residual Stream
    9. 7-RESID
    11. 341
    Prev
    Next
    INDEX
    Explanations

    assignment and comparison operators

    python-code · gemini-2.5-flash-lite

    method calls or attribute access (negative activations)

    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
     dp
    -0.43
    _comb
    -0.28
    _counts
    -0.24
     possible
    -0.21
     comb
    -0.20
    _sum
    -0.19
    inations
    -0.19
     best
    -0.16
     paths
    -0.15
    ination
    -0.15
    POSITIVE LOGITS
    _ob
    0.11
     cont
    0.10
    ource
    0.08
     rows
    0.07
     items
    0.06
    _from
    0.06
    _name
    0.04
     create
    0.04
     set
    0.02
    '
    0.00
    Activations Density 100.000%

    No Known Activations