© 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. 55594
    Prev
    Next
    INDEX
    Explanations

    table formatting, images, and numbers

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

    The neuron is highly selective for Markdown image‐embed or figure‐reference lines (e.g. “![](…){#sp…}”)—i.e. literal `![](...)` constructs with attribute blocks.

    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
    idän
    -0.96
    jét
    -0.94
     lagt
    -0.93
     sû
    -0.89
    forderungen
    -0.87
    lenme
    -0.87
    అ
    -0.86
     situe
    -0.84
     begyn
    -0.84
    usnya
    -0.84
    POSITIVE LOGITS
     table
    1.45
    テーブル
    1.28
    Table
    1.22
     Table
    1.13
    getTable
    1.06
     tables
    1.02
     雑誌
    1.02
     에서
    1.01
    以上
    1.01
     these
    1.01
    Activations Density 0.001%

    No Known Activations