© 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

    exception variable assignment

    python-code · gemini-2.5-flash-lite

    assignment operator (negative activations)

    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
     function
    -0.04
     between
    -0.03
     algorithm
    -0.03
    ffe
    -0.03
    over
    -0.03
    The
    -0.02
    imum
    -0.02
    iven
    -0.02
     the
    -0.02
    thod
    -0.02
    Positive Logits
     q
    0.04
    q
    0.03
    gs
    0.02
     r
    0.02
    .d
    0.02
    Q
    0.02
    text
    0.02
     id
    0.01
     dist
    0.01
    par
    0.01
    Activations Density 99.302%

    No Known Activations

    PrevNext
    Close
    Circuit Sparsity
    No connected neurons data available
    Explanations

    json loading or dumping

    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
    ~
    -0.03
    age
    -0.02
    ge
    -0.02
     out
    -0.02
     any
    -0.01
    ho
    -0.01
    ater
    -0.01
     file
    -0.01
     points
    -0.01
    er
    -0.01
    Positive Logits
    on
    0.10
     represe
    0.07
     representing
    0.07
     represent
    0.07
    uples
    0.04
    ble
    0.04
     Returns
    0.04
     contain
    0.04
    ink
    0.04
    one
    0.03
    Activations Density 83.088%

    No Known Activations