© 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. 36143
    Prev
    Next
    INDEX
    Explanations

    [[...]] bracketed phrases

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

    The neuron is tuned to detect wiki markup tokens (e.g. `[[`, `]]`, pipes, category names, file links).

    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
    </b>
    -2.63
    ’
    -2.06
    madı
    -1.95
     \
    -1.93
    卺
    -1.92
     prioridades
    -1.89
    逡
    -1.87
    段階
    -1.80
    sukienka
    -1.77
    あります
    -1.77
    POSITIVE LOGITS
    to
    2.89
    as
    2.72
    </h3>
    2.58
    л
    2.36
    c
    2.36
    ed
    2.31
    ization
    2.31
    ïdes
    2.28
    鲼
    2.27
    et
    2.20
    Activations Density 0.029%

    No Known Activations