© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130832
    Prev
    Next
    INDEX
    Explanations

    This neuron detects very high-frequency “stopword” tokens (common articles, prepositions, pronouns, etc.) across languages.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    use for introduction or need

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    }$.
    -1.00
    $.
    -0.88
    ^*)
    -0.86
    ]-\
    -0.85
     aquele
    -0.85
    那种
    -0.83
    ...).
    -0.82
    Peut
    -0.81
     tiveram
    -0.80
     essendo
    -0.79
    POSITIVE LOGITS
     them
    1.23
     it
    1.17
     additional
    0.96
     Vaters
    0.95
     Soccerway
    0.95
    Following
    0.93
     analysis
    0.91
     Following
    0.89
    Empresa
    0.88
    Gruppe
    0.87
    Activations Density 0.094%

    No Known Activations