© 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. 123467
    Prev
    Next
    INDEX
    Explanations

    role in or of

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

    The neuron fires on occurrences of “role” (especially in phrases like “possible role” or “role in ...”), marking statements about a factor’s function or significance.

    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
     restes
    -1.85
     cremallera
    -1.85
    ܊
    -1.73
    
    -1.65
     capucha
    -1.65
     MATTERS
    -1.59
    
    -1.59
     dimensiones
    -1.59
     uſe
    -1.59
    辂
    -1.59
    POSITIVE LOGITS
     in
    2.09
    5
    2.02
    8
    1.99
    6
    1.92
    7
    1.77
    4
    1.73
    ↵↵
    1.56
    0
    1.56
    <strong>
    1.46
     began
    1.41
    Activations Density 0.286%

    No Known Activations