© 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. 90213
    Prev
    Next
    INDEX
    Explanations

    fast focusing

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

    The neuron is specialized for spotting technical specification tokens—model numbers, numeric values with units (e.g. USD, mm, pixel-size), and related product-spec jargon.

    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
    Lucifer
    -0.82
     甲
    -0.75
     verbonden
    -0.70
     annak
    -0.68
     baño
    -0.67
    urée
    -0.67
    Athens
    -0.67
     Lucifer
    -0.66
    (_.
    -0.66
    οιν
    -0.66
    POSITIVE LOGITS
     focus
    3.41
     autofocus
    3.25
    focus
    3.03
    Focus
    2.83
     Focus
    2.80
     focusing
    2.52
    FOCUS
    2.50
     focuses
    2.39
     FOCUS
    2.39
     focused
    2.19
    Activations Density 0.061%

    No Known Activations