© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 84026
    Prev
    Next
    INDEX
    Explanations

    US states and geographical regions

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

    The neuron detects mentions of geographic locations or regions (especially place names including states, counties, or directional qualifiers).

    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
    鼕
    -1.91
     画像
    -1.52
     กับ
    -1.49
    这个人
    -1.48
    頃から
    -1.45
     xD
    -1.45
    ⍢
    -1.45
     identificado
    -1.44
    ーん
    -1.43
     deutscher
    -1.42
    POSITIVE LOGITS
    ."
    2.14
    『
    1.70
    на
    1.70
    <em>
    1.59
     \&
    1.55
    њи
    1.55
    、
    1.51
    ны
    1.47
    autoreleasepool
    1.45
    //{
    1.43
    Activations Density 0.042%

    No Known Activations