© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44297
    Prev
    Next
    INDEX
    Explanations

    special characters and foreign words

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

    The neuron fires on tokens that close code blocks—particularly closing curly braces “}.”

    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
    .
    -3.14
    $
    -3.06
     ideen
    -2.92
    if
    -2.81
    Two
    -2.75
    卝
    -2.70
    I
    -2.67
    齪
    -2.66
    ኧ
    -2.64
    providedIn
    -2.63
    POSITIVE LOGITS
     kterou
    2.92
    髕
    2.89
    
    2.61
    ellido
    2.59
     berühmten
    2.50
    tividade
    2.44
    )$
    2.41
    rugada
    2.36
     který
    2.34
     dalších
    2.34
    Activations Density 0.012%

    No Known Activations