© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 104166
    Prev
    Next
    INDEX
    Explanations

    caught exception or error

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

    references to asynchronous programming constructs like "Future", "Task", or "concurrent", as well as error messages containing "Exception" or "Uncaught".

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

    The main thing this neuron does is detect asynchronous programming constructs and keywords (e.g. Task, Future, concurrency APIs).

    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
     ähnlichen
    -2.17
     zahlreichen
    -1.83
     raras
    -1.81
     soñ
    -1.80
     antiguas
    -1.78
     HERE
    -1.77
     ursprünglichen
    -1.76
     einigen
    -1.75
     dibujado
    -1.75
    拼命
    -1.74
    POSITIVE LOGITS
    $
    1.83
     bosco
    1.52
     Océ
    1.48
     fontaine
    1.44
    的
    1.43
    。
    1.38
     mignon
    1.38
     tille
    1.38
    #
    1.34
     mixte
    1.33
    Activations Density 0.001%

    No Known Activations