© 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. 9589
    Prev
    Next
    INDEX
    Explanations

    decorative separators or lines

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

    This neuron detects box‐drawing line characters (e.g. ═, ╔, ╦, ╠, ╚) used to form ASCII/Unicode table borders.

    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
    鑁
    -2.73
     For
    -2.50
     \&
    -2.34
     Они
    -2.30
    ⏑
    -2.28
    陇
    -2.17
    ↀ
    -2.17
    mlung
    -2.16
     в
    -2.13
    ᠊
    -2.11
    POSITIVE LOGITS
    *
    2.88
    _
    2.78
    2
    2.75
    or
    2.67
    1
    2.67
    >
    2.58
    鹮
    2.56
    茑
    2.44
    錞
    2.41
     bunch
    2.30
    Activations Density 0.003%

    No Known Activations