© 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. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33204
    Prev
    Next
    INDEX
    Explanations

    code return statements

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

    The neuron fires on closing curly braces “}” in code, i.e. the ends of blocks.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    bá
    -0.88
     PROCEED
    -0.88
     PREPARATION
    -0.84
    ŵ
    -0.84
     introduit
    -0.83
     veril
    -0.82
     już
    -0.81
    はこの
    -0.81
    例えば
    -0.81
     formulaire
    -0.81
    POSITIVE LOGITS
    最後は
    1.23
    return
    1.04
    次回は
    1.03
     final
    1.01
     return
    0.96
    Spes
    0.96
    为啥
    0.95
    lif
    0.92
    Viited
    0.90
    ではまた
    0.90
    Activations Density 0.002%

    No Known Activations