© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 118990
    Prev
    Next
    INDEX
    Explanations

    innovating, sliding, debilitating conditions

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

    adjectives ending in "-ated", "-able", "-ible" and related forms that describe qualities or properties.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron selectively activates on longer, multisyllabic content words—especially formal, technical, or jargon terms.

    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
    s
    -3.05
     членов
    -2.70
     we
    -2.70
    嫆
    -2.67
    u
    -2.66
    厹
    -2.66
     desenli
    -2.59
     imponer
    -2.56
     izin
    -2.55
     nuevos
    -2.48
    POSITIVE LOGITS
    0
    3.80
    '
    2.69
            
    2.59
    ～～
    2.50
    ～～～
    2.39
    各样的
    2.34
    ～～～～
    2.31
     termed
    2.25
     achieves
    2.25
    ～、
    2.23
    Activations Density 0.019%

    No Known Activations