© 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. 34-GEMMASCOPE-RES-131K
    7. 70225
    Prev
    Next
    INDEX
    Explanations

    CSS and HTML structures

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

    The neuron fires on numeric literals and unit tokens in code (e.g. numbers, “px,” “em,” vendor-prefix numbers) – i.e. it picks out numeric values in code/CSS.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     vorsichtig
    -0.82
     TikTok
    -0.80
    ampe
    -0.76
    fhir
    -0.76
     milli
    -0.76
    liff
    -0.75
    вы
    -0.75
    Curs
    -0.74
     camel
    -0.74
    hamburg
    -0.74
    POSITIVE LOGITS
     clearfix
    1.13
     actions
    0.95
    Controls
    0.89
     Controls
    0.87
    Wrapper
    0.85
    controls
    0.84
     список
    0.82
    ော
    0.81
    ACTIONS
    0.80
    row
    0.79
    Activations Density 0.059%

    No Known Activations