© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 74775
    Prev
    Next
    INDEX
    Explanations

    Russian and foreign grouping terms

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

    The neuron fires on inflectional suffixes in highly inflected languages (e.g. Baltic and Slavic case or tense endings).

    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
    );
    -2.44
    """
    -2.13
     formidable
    -2.09
    )\\
    -2.08
    )$
    -1.99
    ),
    -1.88
     oscura
    -1.85
     infinita
    -1.84
     pesa
    -1.83
     besonder
    -1.82
    POSITIVE LOGITS
    逦
    2.58
    dersfield
    2.34
    窳
    2.31
     VERY
    2.27
    횃
    2.23
    —
    2.22
    垰
    2.20
    during
    2.20
    柺
    2.19
     DURING
    2.17
    Activations Density 0.188%

    No Known Activations