© 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. 104296
    Prev
    Next
    INDEX
    Explanations

    legalistic judgment or observation

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

    The neuron fires on formal legal or regulatory language—words naming or describing official rules, orders, procedures, or judgments (e.g. ordinance, law, permit, custody, coercive, chaos).

    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
     for
    -1.50
     in
    -1.43
     our
    -1.41
     through
    -1.38
    ׇ
    -1.23
     thru
    -1.20
     οὐ
    -1.20
     semacam
    -1.18
     chrétienne
    -1.16
     блюдо
    -1.16
    POSITIVE LOGITS
    颜色
    1.62
    选举
    1.41
    baomidou
    1.39
    很赞
    1.39
    racene
    1.35
     choses
    1.34
     كيل
    1.32
    德國
    1.31
    ⎯
    1.30
    燉
    1.29
    Activations Density 0.018%

    No Known Activations