© 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. 68154
    Prev
    Next
    INDEX
    Explanations

    code syntax or programming keywords

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

    This neuron responds to deeply nested or indented lines in code, effectively detecting high indentation levels.

    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
    Материалы
    -0.91
    ܖ
    -0.90
    ولوج
    -0.88
     mikina
    -0.87
     понравилось
    -0.86
    tivas
    -0.85
    сында
    -0.85
    さんから
    -0.84
    uration
    -0.84
    ヨタ
    -0.84
    POSITIVE LOGITS
    
    0.90
     “
    0.82
     추가
    0.82
    刷新
    0.81
    stringBuilder
    0.80
     letz
    0.78
     augen
    0.75
     tantos
    0.75
    咱
    0.75
    が増
    0.74
    Activations Density 0.001%

    No Known Activations