© 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. 59861
    Prev
    Next
    INDEX
    Explanations

    sterols, Pageable, Vegetarian Recipe

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

    The neuron strongly activates on tokens that contain non-standard ASCII characters or foreign/script-specific text (e.g. Japanese hiragana, accented letters, or uncommon proper-name fragments).

    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
    '
    -2.20
     had
    -1.97
     can
    -1.90
     if
    -1.88
     for
    -1.85
     seems
    -1.82
     then
    -1.77
     amigo
    -1.77
     would
    -1.74
     could
    -1.73
    POSITIVE LOGITS
    hence
    2.16
     hefty
    1.91
    thus
    1.88
    蹓
    1.88
     fervent
    1.87
    uot
    1.84
    those
    1.82
    狲
    1.80
    they
    1.79
    唻
    1.77
    Activations Density 0.001%

    No Known Activations