© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 61402
    Prev
    Next
    INDEX
    Explanations

    legal cases and authors; states

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

    This neuron activates on title‐like or heading text—i.e. standalone names of works, legal case names, section headings, or other document titles.

    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
    閦
    -1.81
    祧
    -1.78
    廡
    -1.77
    ()");
    -1.72
     våra
    -1.72
     leſs
    -1.71
    饪
    -1.67
    幛
    -1.66
     dezelve
    -1.60
    »،
    -1.59
    POSITIVE LOGITS
    ing
    1.50
    up
    1.43
     I
    1.43
    心底
    1.42
    before
    1.39
    section
    1.38
     
    1.38
    k
    1.35
     in
    1.35
    ’
    1.33
    Activations Density 0.002%

    No Known Activations