© 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. 73653
    Prev
    Next
    INDEX
    Explanations

    Locke, Åland, Åkesson, Zhan Shi, Engels

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

    The neuron responds to proper names—i.e. capitalized personal and place names (named entities).

    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
    畱
    -2.73
    纐
    -2.72
     Gedanke
    -2.50
    вгений
    -2.47
    笵
    -2.42
     нескольких
    -2.41
    
    -2.36
     alcune
    -2.31
    好像是
    -2.30
     aiutare
    -2.28
    POSITIVE LOGITS
    '
    3.05
    $
    3.05
    \
    2.89
            
    2.30
    </em>
    2.19
     '
    2.05
     mijn
    2.03
     Perhaps
    2.02
    ziná
    2.02
     theor
    2.00
    Activations Density 0.001%

    No Known Activations