© 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

    html string literal

    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
    ī
    -0.10
     {}↵↵
    -0.08
     {}↵
    -0.06
     []↵↵
    -0.05
     collections
    -0.05
     []↵
    -0.04
    10
    -0.04
     typing
    -0.03
    _rate
    -0.03
    r
    -0.03
    Positive Logits
    ls
    0.09
    _l
    0.08
    mplo
    0.08
    _li
    0.08
    lf
    0.08
    Ä
    0.08
    les
    0.07
     l
    0.07
    lot
    0.07
     lon
    0.07
    Activations Density 19.501%

    No Known Activations

    PrevNext
    Close
    Circuit Sparsity
    No connected neurons data available
    Explanations

    dictionary key or attribute access (negative activations)

    python-code · gemini-2.5-flash-lite

    list assignment with square brackets

    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
     ==
    -0.02
    ays
    -0.02
    .y
    -0.02
    date
    -0.01
    ue
    -0.01
    _y
    -0.01
    ],
    -0.01
    ay
    -0.01
    ],↵
    -0.01
     [[
    -0.01
    Positive Logits
    split
    0.06
    ().
    0.05
    .split
    0.04
    range
    0.04
    .find
    0.03
    .edges
    0.03
    rior
    0.02
    lose
    0.02
     range
    0.02
    uration
    0.02
    Activations Density 99.996%

    No Known Activations