© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130048
    Prev
    Next
    INDEX
    Explanations

    The neuron is highly activated by words referring to armor or armored protective gear.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    armor, rumor, tumor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     troublesome
    -2.78
    他们在
    -2.34
    ൎ
    -2.33
    
    -2.33
     solidaridad
    -2.22
    什么呢
    -2.16
     cumbersome
    -2.16
     prefer
    -2.14
    喽
    -2.14
    
    -2.11
    POSITIVE LOGITS
    "}}>
    2.39
    ї
    2.34
    ſty
    2.27
    蓥
    2.14
    ICOLON
    2.13
     Selasa
    2.11
    ная
    2.11
    on
    2.08
     apapun
    2.08
    袮
    2.05
    Activations Density 0.157%

    No Known Activations