© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68275
    Prev
    Next
    INDEX
    Explanations

    class="

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

    The neuron is detecting occurrences of HTML “class” attributes (i.e. the token class= in markup).

    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
    ’
    -1.81
     usually
    -1.51
     whenever
    -1.47
    缡
    -1.47
     while
    -1.42
     noOf
    -1.41
    鞑
    -1.40
     noticeable
    -1.38
    所有的
    -1.37
     birkaç
    -1.36
    POSITIVE LOGITS
     menyen
    1.86
     egent
    1.74
     dezelve
    1.70
    ributes
    1.70
     attirer
    1.69
     samti
    1.68
    🛇
    1.67
    certainly
    1.65
     jPanel
    1.62
    obviously
    1.60
    Activations Density 0.006%

    No Known Activations