© 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. 571
    Prev
    Next
    INDEX
    Explanations

    legal citations and numbers

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

    The neuron detects numeric tokens (multi-digit numbers and punctuation) used as identifiers—e.g. case numbers, docket numbers, or rule/citation markers.

    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
    LANG
    -0.80
    忍
    -0.73
    MARK
    -0.72
     disclosed
    -0.72
    کند
    -0.69
    LoginComponent
    -0.69
    SECTION
    -0.69
    <h3>
    -0.68
    ('../
    -0.67
     Recruit
    -0.67
    POSITIVE LOGITS
     siècle
    0.88
    applique
    0.84
    从来
    0.84
    を変更
    0.82
    穿过
    0.81
    Espa
    0.80
     piac
    0.78
     papir
    0.78
     dîner
    0.77
    ριν
    0.76
    Activations Density 0.003%

    No Known Activations