© 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. 85473
    Prev
    Next
    INDEX
    Explanations

    sub-Saharan Africa

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

    The neuron activates on geographic references containing “Saharan,” such as “Western Sahara” or “Sub-Saharan Africa.”

    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
    絎
    -3.16
    樒
    -3.16
    饪
    -3.03
    橚
    -2.92
    
    -2.91
    纐
    -2.86
    鏷
    -2.86
    AdvEx
    -2.84
    沄
    -2.83
    訁
    -2.83
    POSITIVE LOGITS
    ↵↵
    2.78
    to
    2.73
    ↵
    2.61
     Our
    2.59
    te
    2.48
     If
    2.33
    and
    2.31
     I
    2.25
     WAY
    2.23
    но
    2.17
    Activations Density 0.003%

    No Known Activations