© 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. 39413
    Prev
    Next
    INDEX
    Explanations

    unarmed civilian

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

    The neuron selectively activates on the overline or macron character (¯) and other similar extended‐Unicode diacritical symbols.

    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
    Now
    -3.25
    Before
    -2.92
    𖥧
    -2.69
    What
    -2.66
    ”،
    -2.58
     }]
    -2.56
    After
    -2.53
    Then
    -2.53
    </td>
    -2.52
     οποίο
    -2.52
    POSITIVE LOGITS
    蹰
    2.69
    小伙伴
    2.20
     Feinde
    2.13
     but
    2.11
    癭
    2.08
    こども
    2.05
     
    2.03
    女孩子
    2.02
    婠
    2.00
     بالإضافة
    1.98
    Activations Density 0.003%

    No Known Activations