© 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. 95715
    Prev
    Next
    INDEX
    Explanations

    HTML table markup

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

    This neuron fires on numeric literal tokens in HTML markup—digits or hex‐code values (e.g. table row IDs, bgcolor hex codes, numeric attributes).

    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
     a
    -1.04
     by
    -1.02
    marcos
    -0.94
     évé
    -0.94
     enough
    -0.92
     मामले
    -0.91
     pensée
    -0.90
    ържа
    -0.90
    Calculus
    -0.89
     宝宝
    -0.87
    POSITIVE LOGITS
     САД
    1.13
    </tr>
    1.11
    歡
    0.98
    ]=\
    0.94
     members
    0.92
    naeus
    0.90
     previous
    0.89
     weiteres
    0.87
     unanimously
    0.87
     fled
    0.87
    Activations Density 0.004%

    No Known Activations