© 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. 10-GEMMASCOPE-RES-131K
    7. 57620
    Prev
    Next
    INDEX
    Explanations

    patent and URL components

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

    The neuron fires on tokens containing embedded underscores (e.g. URL query or cookie parameter 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
     ！
    -1.55
     ';
    -1.52
     situe
    -1.52
    !";
    -1.51
     \
    -1.50
     rê
    -1.46
     cso
    -1.46
     berbagai
    -1.46
     fís
    -1.44
     decorativa
    -1.44
    POSITIVE LOGITS
    :
    1.91
    4
    1.91
    5
    1.81
     (
    1.76
    3
    1.72
     –
    1.70
     as
    1.66
    1
    1.63
    <td>
    1.63
    ?
    1.60
    Activations Density 0.023%

    No Known Activations