© 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. 22-GEMMASCOPE-RES-131K
    7. 380
    Prev
    Next
    INDEX
    Explanations

    specific names and places

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

    This neuron fires on the dot-separated name tokens you find in author contact lines (e.g. the parts of an email username like “christian.schulz”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    ‪.
    -1.44
    粜
    -1.43
     bluza
    -1.41
     bricola
    -1.40
     花纹
    -1.40
    Ꭾ
    -1.39
    鵙
    -1.38
    釤
    -1.36
    ܆
    -1.33
     inteligen
    -1.32
    POSITIVE LOGITS
     in
    1.69
     for
    1.67
     of
    1.63
     on
    1.62
     just
    1.60
     at
    1.52
     (
    1.49
     from
    1.46
    :
    1.43
     no
    1.41
    Activations Density 0.109%

    No Known Activations