© 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. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49808
    Prev
    Next
    INDEX
    Explanations

    code variable assignments

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is picking out identifier names (variable, attribute, and configuration keys) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    programming identifiers in code—especially Python variable and attribute names, including snake_case and dunder names, and names used in indexing or dictionary keys.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    variable assignments in Python code, particularly when a variable is being set equal to a value.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ;
    -2.30
    }$;
    -2.00
    );
    -1.83
    ();
    -1.82
     "";
    -1.77
     [];
    -1.73
    ];
    -1.62
    %;
    -1.59
    ();
    -1.57
    [];
    -1.53
    POSITIVE LOGITS
    ,
    1.50
    ()
    1.28
     {}
    1.25
     ""
    1.23
     ''
    1.20
    =[]
    1.12
    )”
    1.03
    ”
    1.03
     None
    1.02
    '
    1.02
    Activations Density 0.018%

    No Known Activations