© 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. 54290
    Prev
    Next
    INDEX
    Explanations

    beginnings of structured text

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

    The neuron detects terminology used in corporate forward‐looking statements and risk‐factor disclaimers (e.g., “future,” “results,” “uncertainties,” “include,” “impact,” etc.).

    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
     begrenzt
    -0.99
    ֩
    -0.98
     inkább
    -0.97
    женные
    -0.94
    meia
    -0.94
     wzglę
    -0.92
     dagegen
    -0.92
    AILS
    -0.92
    を使
    -0.91
     chré
    -0.91
    POSITIVE LOGITS
     and
    1.87
     which
    1.34
     new
    1.27
     or
    1.21
    っており
    1.19
     future
    1.18
    纔
    1.15
     whether
    1.10
    给我们
    1.09
     our
    1.09
    Activations Density 0.003%

    No Known Activations