© 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. 13769
    Prev
    Next
    INDEX
    Explanations

    foreign or special characters

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

    The neuron primarily detects double‐quote characters (marking quoted text).

    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
    gewiesen
    -1.01
    ayangkan
    -0.99
     this
    -0.98
     meil
    -0.97
     gâteau
    -0.97
     refuser
    -0.97
    aarr
    -0.96
     syndical
    -0.96
     maillot
    -0.96
     politika
    -0.94
    POSITIVE LOGITS
    []{"
    0.97
    说实话
    0.96
     "
    0.91
    Ａ
    0.91
    ('
    0.88
     Soldaten
    0.88
    による
    0.88
    שׁ
    0.87
     字符
    0.87
    racht
    0.86
    Activations Density 0.018%

    No Known Activations