© 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. 22-GEMMASCOPE-RES-131K
    4. 13428
    Prev
    Next
    INDEX
    Explanations

    local government and places

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

    The neuron is highly activated by formal administrative or governmental terms (e.g. “proclaimed,” “Council,” “Municipality”) often occurring in the context of official dates or civic records.

    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
     first
    -1.32
     many
    -1.21
     we
    -1.20
     خوبه
    -1.18
     Políticas
    -1.16
     focusing
    -1.15
     our
    -1.12
     ملابس
    -1.10
    carlos
    -1.10
     ἡ
    -1.09
    POSITIVE LOGITS
    rifft
    1.61
     الحكومة
    1.54
    쫀
    1.47
     correos
    1.45
     السياس
    1.42
    ţe
    1.40
    动静
    1.40
    举动
    1.40
    脚步
    1.33
    权力
    1.31
    Activations Density 0.000%

    No Known Activations