© 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. 11008
    Prev
    Next
    INDEX
    Explanations

    css class names

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

    This neuron is keyed to CSS class selector syntax, specifically spotting the leading “.” before class 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_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
     if
    -1.29
     or
    -1.14
     These
    -1.13
     While
    -1.10
     Although
    -1.08
     Whether
    -1.03
     However
    -1.02
     Regardless
    -0.99
     Since
    -0.99
     different
    -0.98
    POSITIVE LOGITS
     ſay
    1.08
    Corte
    1.00
     Pompeo
    0.99
    loß
    0.97
     appelez
    0.96
     prenez
    0.96
    ございません
    0.95
    注意事
    0.94
     ağı
    0.94
     Mahar
    0.93
    Activations Density 0.003%

    No Known Activations