© 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. 34-GEMMASCOPE-RES-131K
    7. 60926
    Prev
    Next
    INDEX
    Explanations

    error messages and missing files

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

    The neuron fires on numeric tokens (error codes, line numbers, and similar digit strings) in programming error messages or stack traces.

    oai_token-act-pair · o4-miniTriggered 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
    relenting
    -0.90
     alemão
    -0.90
    QUA
    -0.90
     تفاوت
    -0.88
    ysacchar
    -0.88
     commenti
    -0.87
     Зачем
    -0.87
     Witch
    -0.83
     luf
    -0.83
    ifc
    -0.83
    POSITIVE LOGITS
     at
    1.46
     while
    1.16
     occurred
    1.12
     remedy
    1.05
     Perhaps
    0.97
     make
    0.96
     caused
    0.95
     occurs
    0.95
    Weitere
    0.94
     during
    0.92
    Activations Density 0.125%

    No Known Activations