© 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. 10-GEMMASCOPE-RES-131K
    7. 103085
    Prev
    Next
    INDEX
    Explanations

    debugging

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

    references to debugging in technical or programming contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The main thing this neuron does is detect occurrences of “debug” (and its inflected forms like “debugged” or “debugging”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     It
    -3.33
    "
    -1.97
    </h2>
    -1.97
     württemberg
    -1.95
    </h4>
    -1.86
     ehemaligen
    -1.81
    比べて
    -1.80
    比べ
    -1.76
     KNOW
    -1.74
    
    -1.73
    POSITIVE LOGITS
    鹛
    2.20
    涐
    2.08
     parachoque
    1.84
    ྕ
    1.83
    TextWatcher
    1.80
     Debugging
    1.80
     semacam
    1.80
     Troubleshooting
    1.78
    there
    1.73
    вшие
    1.71
    Activations Density 0.038%

    No Known Activations