© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27143
    Prev
    Next
    INDEX
    Explanations

    following words, code separators

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

    The neuron strongly responds to CSS rules that define a “content:” property (notably Font Awesome pseudo‐element content declarations).

    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
     by
    -1.01
    pecia
    -0.95
    ;';
    -0.94
    ++.
    -0.93
     Terrasse
    -0.90
    ebab
    -0.89
     applications
    -0.88
    {}\
    -0.88
     første
    -0.88
    blé
    -0.88
    POSITIVE LOGITS
    続いて
    1.15
    接著
    1.05
    Similarly
    1.05
    同じく
    1.02
     similarly
    1.00
    同样的
    0.98
     ditto
    0.96
    后再
    0.96
    变了
    0.95
    同樣
    0.94
    Activations Density 0.005%

    No Known Activations