© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 50425
    Prev
    Next
    INDEX
    Explanations

    legal case numbers

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

    The neuron fires strongly on long runs of non‐alphanumeric characters (especially asterisks), i.e. comment or section separator 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_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
    Интере
    -0.96
    Катего
    -0.93
    mapTo
    -0.90
    いいのか
    -0.87
    Če
    -0.85
    ³.
    -0.85
    Май
    -0.84
     Guarda
    -0.83
    ERSHIP
    -0.82
    erty
    -0.81
    POSITIVE LOGITS
     Архивировано
    0.93
    ベトナム
    0.79
    ⽉
    0.74
     Publikum
    0.73
    めちゃ
    0.69
    eynman
    0.67
    ↵
    0.67
    faux
    0.66
    lighter
    0.66
    Flint
    0.65
    Activations Density 0.002%

    No Known Activations