© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11256
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on geographic location names (e.g., “China”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    文本
    0.70
    第
    0.63
    摹
    0.60
    sort
    0.58
    Text
    0.57
    呈
    0.57
    text
    0.55
    Eval
    0.55
    现实
    0.55
    TEXT
    0.54
    POSITIVE LOGITS
     Basque
    0.68
    
    0.68
     নেয়
    0.67
     Psychiatry
    0.65
     Kyrgyz
    0.64
     Alcohol
    0.64
     Katar
    0.64
     señor
    0.63
     Teut
    0.63
     Lizard
    0.62
    Activations Density 0.000%

    No Known Activations