© 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. 22-GEMMASCOPE-RES-131K
    7. 71638
    Prev
    Next
    INDEX
    Explanations

    code titles and questions

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

    This neuron fires on code comment markers and preprocessor directives (e.g. “//”, “/* … */”, and “#” lines).

    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
     où
    -1.02
     rowspan
    -0.94
    anoic
    -0.90
    <h3>
    -0.90
    laria
    -0.89
     остальных
    -0.86
    KeyPressed
    -0.86
     Generator
    -0.85
    див
    -0.85
    ibli
    -0.84
    POSITIVE LOGITS
     ذلك
    1.06
     設置
    1.02
     beslu
    0.96
     貨
    0.93
    גות
    0.93
     kunne
    0.90
    льності
    0.89
    ➖➖
    0.89
     detalj
    0.89
     قاشق
    0.88
    Activations Density 0.001%

    No Known Activations