© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26
    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    Side-by-side compare needs a wider screen.
    Open other featureClose compare
    PrevNext
    Close
    Circuit Sparsity
    No connected neurons data available
    Explanations

    replace or return keyword

    python-code · gemini-2.5-flash-lite
    Auto-Interp Tools
    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
    Negative Logits
    Į
    -0.06
    e
    -0.06
    >
    -0.06
    ats
    -0.05
    ating
    -0.05
    ated
    -0.05
    at
    -0.05
    dge
    -0.04
    ator
    -0.04
    quence
    -0.04
    Positive Logits
    ...
    0.09
     main
    0.04
     initial
    0.03
    move
    0.03
    main
    0.03
     //
    0.03
    o
    0.03
    l
    0.03
    ain
    0.03
    uild
    0.03
    Activations Density 15.003%

    No Known Activations

    PrevNext
    Close
    Circuit Sparsity
    No connected neurons data available
    Explanations

    list element after comma (negative activations)

    python-code · gemini-2.5-flash-lite

    variable attribute access

    python-code · gemini-2.5-flash-lite
    Auto-Interp Tools
    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
    Negative Logits
    py
    -0.05
     run
    -0.04
    dule
    -0.04
    uil
    -0.04
    reshold
    -0.03
     _
    -0.03
    100
    -0.03
    20
    -0.03
     iter
    -0.03
     sys
    -0.03
    Positive Logits
    adjac
    0.03
    uccess
    0.03
     conne
    0.02
    duct
    0.02
    ag
    0.02
    il
    0.02
    ong
    0.02
    _sum
    0.01
     parent
    0.01
    con
    0.01
    Activations Density 88.908%

    No Known Activations