© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14590
    Prev
    Next
    INDEX
    Explanations

    introducing assumptions or conditions

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

    The neuron flags question‐style words and phrases that express uncertainty or ask “how to” do something (e.g. how, could, way, sure, know how to).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     three
    -1.98
     four
    -1.95
    ↵↵
    -1.95
     nine
    -1.88
     two
    -1.86
     six
    -1.84
     eight
    -1.82
     one
    -1.80
     five
    -1.78
     There
    -1.72
    POSITIVE LOGITS
     bijvoorbeeld
    2.17
    particularly
    2.00
    茈
    1.95
    
    1.94
     intric
    1.94
     juſ
    1.87
    specifically
    1.86
    creating
    1.85
    ︲
    1.84
    potentially
    1.80
    Activations Density 0.193%

    No Known Activations