© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5777
    Prev
    Next
    INDEX
    Explanations

    flags and pride symbols

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

    The neuron detects named symbolic emblems or insignia—words referring to flags, crests, mottos, heraldic shields, logos, etc.

    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
     Metabol
    0.71
     metabol
    0.68
    醂
    0.66
     Metabolism
    0.64
     容量
    0.64
     Wellness
    0.63
     Whisper
    0.62
    蓯
    0.62
     LUM
    0.62
     metformin
    0.62
    POSITIVE LOGITS
     flag
    2.08
     flags
    2.02
    flag
    1.91
    Flag
    1.89
     logo
    1.88
     emblem
    1.82
     insignia
    1.82
     Flag
    1.81
     Flags
    1.80
    flags
    1.79
    Activations Density 0.466%

    No Known Activations