© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 58325
    Prev
    Next
    INDEX
    Explanations

    Joshua Rodgers Scala Merritt

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

    This neuron activates on capitalized identifiers—proper nouns or class/module names (e.g. person names, programming‐language names, class names).

    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
    持つ
    -2.61
    
    -2.56
    </h4>
    -2.55
    さまざま
    -2.48
    
    -2.45
    蠖
    -2.41
     Voraussetzung
    -2.39
    ሏ
    -2.39
     głó
    -2.39
     oft
    -2.39
    POSITIVE LOGITS
    '
    3.19
    1
    3.17
    <
    3.14
    B
    3.02
    0
    3.00
    2
    2.80
    ه
    2.75
    ции
    2.53
    Even
    2.44
    In
    2.41
    Activations Density 0.012%

    No Known Activations