© 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. 72296
    Prev
    Next
    INDEX
    Explanations

    lists, lists of words

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

    The neuron strongly responds to comma characters (i.e. “,” punctuation).

    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
     داره
    -2.66
    quelize
    -2.42
     électron
    -2.36
    夲
    -2.34
    
    -2.30
    𓆩
    -2.23
    .»
    -2.20
    नलोड
    -2.20
     merciless
    -2.14
    ξει
    -2.14
    POSITIVE LOGITS
     are
    2.80
     by
    2.39
     in
    2.31
     so
    2.06
    <em>
    2.00
     There
    1.95
     is
    1.85
     Such
    1.81
     made
    1.81
     from
    1.81
    Activations Density 0.009%

    No Known Activations