© 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. 10-GEMMASCOPE-RES-131K
    4. 41847
    Prev
    Next
    INDEX
    Explanations

    more particularly

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

    The neuron fires on formal scope- and condition-marking words (e.g. generally, particularly, substantially, pursuant, applicable, subject, etc.) characteristic of legal or technical qualifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     couvrir
    -1.69
     хочет
    -1.65
    妧
    -1.52
    铤
    -1.52
     beleid
    -1.50
    嫑
    -1.50
    まさかの
    -1.48
     ouvido
    -1.48
     которы
    -1.46
     exclusivamente
    -1.45
    POSITIVE LOGITS
     on
    1.80
     for
    1.59
    ,"
    1.57
     the
    1.56
    (
    1.55
    就
    1.54
     uite
    1.52
    ://
    1.45
     of
    1.41
    —
    1.41
    Activations Density 0.084%

    No Known Activations